QassTiv\ 5" \ _ Book i—i ■flo\ ELECTRICAL ENGINEER fS POCKET - BOOK A Hand - "book of useful data for Electricians and Electrical Engineers Horatio A. Poster with the Collaboration of Eminent Specialists. Third Edition, Corrected, llinth Thousand. New York: D. Van No strand Company London: E. & P. N. Spon, Ltd. 1903 -'"'i t »+■■.*■.<.„ j2 "" (^PREFACE. It is with some little trepidation that this book is put before the public, in view of the frequent important, and even radical, changes that up to the present have char- acterized the development of electrical engineering. It has, however, been thought that the science has now reached a stage which renders necessary some manual that will be of assistance to the active worker in the various branches. This book is not an encyclopedia, nor is it intended for a text-book, but it is hoped that as a compendium of useful data it may assist the practicing electrician and engineer. The matter included is representative of American practice, and no effort has been made to include any other, except in special cases. No excuse is offered for the very considerable amount of matter taken from trade publications of the larger electrical manufacturers, as in this country the engineers retained by such works are specialists — often the best in their various branches ; and it is an accident of condition only that in some cases has compelled the use of more of the publi- cations of one company than of another, based upon available published material. Manufacturers have been most kind in supplying any special data and descriptions asked for ; and the author's thanks are in particular due to a large circle of asso- iii IV PREFACE. ciates for suggestions, revisions, critical proof-reading, and the various other details involved in a compilation of this kind, of whom the following deserve especial mention for valuable aid rendered : Messrs. E. E. Idell, W. D. Weaver, T. C. Martin, Prof. Samuel Sheldon, E. B. Raymond, John S. Griggs, Jr., William Wallace Christie, J. J. Crain, G-rahame H. Powell, Prof. Erancis B. Crocker, A. 1ST. Mansfield, E. M. Hewlett, C. E. Scott, H. S. Putnam, Charles Henry Davis, Townsend Wolcott, Walter S. Moody, Herbert Laws Webb, Charles Thorn, William Maver, Jr., Joseph Appleton, Prof. Alex. G-. McAdie, Thorburn Reid, Max Osterberg, Max Loewen- thal, J. G-. White & Co. The especial thanks of the author are due to the indefatigable co-operation of Mr. Charles E. Speirs, of the D. Van ISTostrand Co., who has rendered most valuable assistance in properly get- ting the matter into shape for publication. In closing, the author begs that readers will not hesi- tate to point out errors found in the text or tables, as many will doubtless crop out in the close examination by numerous readers. LIST OF CONTRIBUTORS. SECTION. Symbols, Units and Instruments. Resistance, Electrical Measurements. Cable Testing (re-Written). Conductors (Properties of). Conductors (Relation and Dimension of). Electric Lighting. Lightning Arresters. Electric Street Railways. Storage Batteries. Telephony. Magnetic Properties of Iron. Electromagnets. Determination of Wave Form. Electricity Meters. Dynamos and Motors. Dynamos and Motors Standard and Test. Static Transformers. Telegraphy. Switchboards and Switching Devices. Transmission of Power. Electricity in U. S. Navy. Certain Uses of Electricity in U. S. Army. Electro-chemistry, Electro-metallurgy. Electric Heating, Cooking, and Electric Welding. Mechanical Section. Lightning Conductors. Miscellaneous Section. Underwriter's Code. Index Electrical Section. Index Mechanical Section. REVISED BY Prof. Samuel Sheldon. Prof. Samuel Sheldon. Mr. William Maver, Jr. Prof. Samuel Sheldon. Prof. Samuel Sheldon. Mr. Townsend Wolcott. Mr. Townsend Wolcott. Mr. John S. Griggs, Jr. Mr. Townsend Wolcott. Mr. Herbert Laws Webb. Prof .Samuel Sheldon. Mr. Townsend Wolcott. Mr. Townsend Wolcott. Prof. Samuel Sheldon. Mr. E. B. Raymond. Mr. E. B. Raymond. J Mr. Walter S. Moody. I Mr. Townsend Wolcott. ( Mr. Chas. Thorn. I Mr. Herbert Laws Webb. Mr. E. M. Hewlett. Mr. T. C. Martin. Mr. J. J. Crain. Mr. Grahame H. Powell. Prof. Francis B. Crocker. | Mr. Max Osterberg. * Mr. Max Loewenthal. ( Mr. Wm. Wallace Christie. t Mr. F. E. Idel. Prof. Alex. G. McAdie. Mr. Townsend Wolcott. Mr. Max Loewenthal. Mr. Win. Wallace Christie. J CONTENTS. PAGE Symbols, Units, Instruments 1 Resistance Measurements 38 Magnetic Properties of Iron 64 Electro-magnets 81 Relation and Dimensions of Conductors 92 Properties of Conductors 140 Cable Testing 220 Dynamos and Motors 230 Dynamo and Motor Standards 293 The Static Transformer 331 Electric Lighting 38G Electric Street Railways 423 Transmission of Power 548 Storage Batteries 552 Switchboards 585 Lightning Arresters 601 Electricity Meters 615 Telegraphy 636 Telephony 645 Electro-Chemistry and Electro-Metallurgy 675 Electric Heating, Cooking, and Welding 683 Operation of Electric Mining Plants 696 Lightning Conductors 701 Determination of Wave Form , 705 Electricity in the IT. S. Army 711 Electricity in the U. S. Navy 727 Miscellaneous 757 National Code Rules and Requirements 762 Mechanical Section 791 Index 977 vii SYMBOLS, UNITS, INSTRUMENTS. CHAPTER I. ELECTRICAL ENGINEERING SYMBOLS The following list of symbols has been compiled from various sources as being those most commonly in use in the United States. Little variation will be found from similar lists already published except the elimination of some that may be considered exclusively foreign. The list has been revised by competent authorities and may be considered as representing the best usage. fundamental. I, Length, cm. = centimeter ; in., or //=inch, ft. or ' = foot. M, Mass. gr. = mass of 1 gramme ; kg. = 1 kilo- gramme. T, t, Time. s = second. Derived: geometric. 5, s, Surface. V, Volume. a, /3, Angle. Mechanical. v, Velocity. m, Momentum. -£ ft g ft ~ S< ^ ^ o © «2i§ |.Sd-s £ 60 © ^ © h -d ~'3 ©02 *;5 © § Sg 1 3 3 ©t Q S 28 § 1/2 , ^ © • © ft © © Q^ © « ■ -5 -C bfl 1 So H ^ tt0 tf O < 5 "3 " © *> =0 l-fl **&■ SS | t (> £ &i £ eft 3 ^ Kj •5 C? <1 bc~ .3.2 ^ihi»w*i&i ig^ mi ii i II 1 ii in 11 flH 05 ^ « § 3 8 (^ «h •, ^ -SJ, *. Vj ^ v 02,0 ■4^S> 02 « a 3 8 fe^te A, ^ © >> ' s • ."£ ^ a G? I u g o o '3 be © > © £ ■ i c fi fl © o o Pm s < ii iE- © 1 - < 3 > s J 5 < c; 3 © s 1 a 3 o 1 ELECTRICAL ENGINEERING UNITS. o>H P e « e ert c ^ w E*< ^ 8 r 53 s £ "S e g e e ~ S Q> | ^ o 6 o ii* -i-- i & ,Q £* £ £ * < < < ii ii ii l^l^^jSl^i^l^^hlT^^'^,: £ll II II II I II ,ii II II || || II II II II II II a, ^ •■ *■* $&^'H»®U^£ * 6 '. ■ s DQ «. K >> o a © be s 3 ^ § a S C c Moment tftnsitv (Magnetic) Force .... c Induction zing Forcet nnotive Force ice (Magnetic Resistance) . ->ilitv ......... 2 05 33 ° CSC Ck 3 O" Q Pk P3 Occ 35 S h Pm^SSSm Oh «} P4 SYMBOLS, UNITS, INSTRUMENTS. Abbrevia- tions of the Practical Units. > a ft .a iS S © 0 W c © •is * *o G ^ ? ±5 '3 t3 © ©S . e3 c3 fl © 6 £ i 08 "o a £ -* a A _© © 63 C > "J J3 3 ft ■^ =3 © a a li C O Pn c !* ,a 0 © «2 o5 h 0 0 0 0 0 0 1 t> dD © . > C5|N n ^ hC 2? a ^ « ^ ^ ^ ^ tq w>a Eh .3.2 K)j S Aj b;::: Eh 0>|Kjbq fcj^l^ ^|ft^,a^- ■*, 6qi^ Si 1 II II II 1 1! 1 II 1 II 1 1 °w ft K] *h O" O ^ ft «■ C ?- ^ N) a o5 & w 01 3^ 05 5. £>0 CO ^2 ft " ftq* ^ c? Q- c * *■ 4 N A "© c g © ^ a a j 1 0 SB 3 © a 73 a 0 a fl O t»» O 0 6C oa ©~ ^ Ph © . «a •• 3 G? "3 s a 0 o| © 3 Q O 0 © > s c » i c p. © a t i •a » co ^ a I ^ V '■5 £ a ~ 5 © c © 83 I - © a T "2 ^ ( < eS C 0 I % H M 3" c £ ps C 0 0 P^ 1 INTERNATIONAL ELECTRICAL UNITS. watt equals 107 absolute units, and 746 watts equals 1 horse-power. In elec- tric lighting and power the unit kilowatt, or 1000 watts, is considerably used to avoid the use of large numbers. Resistivity (symbol p) is the specific resistance of a substance, and is the resistance in ohms of a centimeter cube of the material to a flow of cur- rent between opposite faces. Conductance (symbol G) is that property of a metal or substance by which it conducts an electric current, and equals the reciprocal of its resistance. The unit proposed for conductance is the Mho, but it has not come into prominent use as yet. Conductivity (symbol v) is the specific conductance of a material, and is therefore the reciprocal of its resistivity. It is often expressed in compari- son with the conductivity of some standard metal such as silver or copper, and is then stated as a percentage. Inductance (symbol L), or coefficient of self-induction, of a circuit is that coefficient by which the time rate of charge of the current in the circuit must be multiplied in order to give the E.M.F. of self-induction in the circuit. The practical unit is the henry, which equals 109 absolute units, and exists in a circuit when a current varying 1 ampere per second produces a, volt of electro-motive force in that circuit. As the henry is so large as to be seldom met with in practice, 1 thousandth of it, or the milli-henry , is the unit most in use. Below will be found a few rules for reducing values stated in electrostatic units to units in the electro-magnetic system. To reduce electrostatic potential to volts, multiply by 300 ; " capacity to micro-farads, divide by 900,000 ; " quantity to coulombs, divide by 3 x 109 ; " current to amperes, divide by 3 x 109; " resistance to ohms, multiply by 9 X 1011. IHTTERHfATIOMAL ELECTRICAL "UHFITS. At the International Congress of Electricians, held at Chicago, August 21, 1893, the following resolutions met with unanimous approval, and being approved for publication by the Treasury Department of the United States Government, Dec. 27, 1893, and legalized by act of Congress and approved by the President, July 12, 1894, are now recognized as the International units of value for their respective purposes. RE SOL VED, That the several governments represented by the delegates of the International Congress of Electricians be, and they are hereby, recommended to formally adopt as legal units of electrical measure the following : — 1. As a unit of resistance, the International ohm, which is based upon the ohm equal to 10 9 units of resistance of the C.G.S. system of electro-magnetic units, and is represented by the resistance offered to an unvarying electric current by a column of mercury at a temperature of melting ice, 14.4521 grammes in mass, of a constant cross-sectional area, and of the lengtb 106.3 centimeters. 2. As a unit of current, the International ampere, which is one-tenth of the unit of current of the C.G.S. system of electro-magnetic units, and which is represented sufficiently well for practical use by the unvarying current which, when passed through a solution of nitrate of silver in water, in accordance with the accompanying specification (A) deposits silver at the rate of 0.001118 gramme per second. 3. As a unit of electro-motive force the international volt which is the E.M.F. that, steadily applied to a conductor whose resistance is one Inter- national ohm, will produce a current of one international ampere, and which is represented sufficiently well for practical use by — — of the E.M.F. between the poles or electrodes of the voltaic cell known as Clark's cell at a temperature of 15° C, and prepared in the manner described in the ac- companying specification (B). 4. "As the unit of quantity, the International coulomb, which is the quan- tity of electricity transferred by a current of one international ampere in one second. 5. As the unit of capacity the international farad, which is the capacity 10 SYMBOLS, UNITS, INSTRUMENTS. of a conductor charged to a potential of one international volt by one inter- national coulomb of electricity. 6. As the unit of work, the joule, which is 10 7 units of work in the C.G.S. system, and which is represented sufficiently well for practical use by the energy expended in one second by an international ampere in an inter- national ohm. 7. As the unit of power, the watt, which is equal to 10 7 units of power in the C.G.S. system, and which is represented sufficiently well for practical use by the work done at the rate of one joule per second. 8. As the unit of induction, the henry, which is the induction in the cir- cuit when the E.M.F. induced in this circuit is one international volt, while the inducing current varies at the rate of one international ampere per second Specification A. In employing the silver voltameter to measure currents of about one ampere, the following arrangements shall be adopted : The kathode on which the silver is to be deposited shall take the form of a platinum bowl not less than 10 cms. in diameter, and from 4 to 5 cms. in depth. The anode shall be a disk or plate of pure silver some 30 sq. cms. in area, and 2 or 3 cms. in thickness. This shall be supported horizontally in the liquid near the top of the solution by a silver rod riveted through its center. To prevent the disintegrated silver which is formed on the anode from falling upon the kathode, the anode shall be wrapped around with pure filter paper, secured at the back by suitable folding. The liquid shall consist of a neutral solution of pure silver nitrate, con- taining about 15 parts by weight of the nitrate to 85 parts of water. The resistance of the voltameter changes somewhat as the current passes. To prevent these changes having too great an effect on the current, some resistance, besides that of the voltameter, should be inserted in the circuit. The total metallic resistance of the circuit should not be less than 10 ohms. Method of making- a Measurement. — The platinum bowl is to be washed consecutively with nitric acid, distilled water, and absolute alcohol ; it is then to be dried at 160° C, and left to cool in a desiccator. When cold it is to be weighed carefully. It is to be nearly filled with the solution, and connected to the rest of the circuit by being placed on a clean copper support to which a binding-screw is attached The anode is then to be immersed in the solution so as to be well covered by it, and supported in that position ; the connections to the rest of the circuit are then to be made. Contact is to be made at the key, noting the time. The current is to be allowed to pass for not less than half an hour, and the time of breaking contact observed. * The solution is now to be removed from the bowl, and the deposit washed with distilled water, and left to soak for at least six hours. It is then to be rinsed successively with distilled water and absolute alcohol, and dried in a hot-air bath at a temperature of about 160° C. After cooling in a desiccator it is to be weighed again. The gain in mass gives the silver deposited To find the time average of the current in amperes, this mass, expressed m grammes, must be divided by the number of seconds during which the current has passed and by 0.001118. In determining the constant of an instrument bv this method the current should be kept as nearly uniform as possible, and the readings of the instru- ment observed at frequent intervals of time. These observations give a curve from which the reading corresponding to the mean current (time average of the current) can be found. ^"he cu.rrent is calculated from the voltameter results, corresponding to this reading. ^ & The current used in this experiment must be obtained from a battery and not from a dynamo, especially when the instrument to be calibrated is an electrodynamometer. Specification B. — The Volt. The cell has for its positive electrode, mercury, and for its negative elec- trode, amalgamated zinc ; the electrolyte consists of a saturated solution of SPECIFICATION B. 11 zinc sulphate and mercurous sulphate. The electromotive force is 1.434 volts at 15° C., and, between 10° C. and 25° C, by the increase of 1° C. in tempera- ture, the electromotive force decreases by .00115 of a volt. 1. .Preparation of the Mercury. — To secure purity it should be first treated with acid in the usual manner, and subsequently distilled in vacuo. •£. Preparation of the Zinc Amalgam. — The zinc designated in commerce as "commercially pure" can be used without further prepara- tion. For the preparation of the amalgam one part by weight of zinc is to be added to nine (9) parts by weight of mercury, and both are to be heated in a porcelain dish at 100° C. with moderate stirring until the zinc has been fully dissolved in the mercury. 3. Preparation of the Mercurous Sulphate. — Take mercurous sulphate, purchased as pure, mix with it a small quantity of pure mercury, and wash the whole thoroughly with cold distilled water by agitation in a bottle ; drain oil' the water and repeat the process at least twice. After the last washing, drain off as much of the water as possible. (For further de- tails of purification, see Note A.) 4t. Preparation of the Zinc Sulphate Solution. — Prepare a neutral saturated solution of pure re-crystallized zinc sulphate, free from iron, by mixing distilled water with nearly twice its weight of crystals of pure zinc sulphate and adding zinc oxide in the proportion of about 2 per cent by weight of the zinc sulphate crystals to neutralize any free acid. The crystals should be dissolved by the aid of gentle heat, but the temperature to which the solution is raised must not exceed 30° C. Mercurous sulphate, treated as described in 3, shall be added in the proportion of about 12 per cent by weight of the zinc sulphate crystals to neutralize the free zinc oxide remaining, and then the solution filtered, while still warm, into a stock bottle. Crystals should form as it cools. 3. Preparation of the Mercurous Sulphate and Zinc Sul- phate Paste. — For making the paste, two or three parts by weight of mercurous sulphate are to be added to one by weight of mercury. If the sulphate be dry, it is to be mixed with a paste consisting of zinc sulphate crystals and a concentrated zinc sulphate solution, so that the whole con- stitutes a stiff mass, which is permeated throughout by zinc sulphate crys- tals and globules of mercury. If the sulphate, however, be moist, only zinc sulphate crystals are to be added ; care must, however, be taken that these occur in excess, and are not dissolved after continued standing. The mercury must, in this case also, permeate the paste in little globules. It is advantageous to crush the zinc sulphate crystals before using, since the paste can then be better manipulated. To set un the Cell. — The containing glass vessel, represented in the accompanying figure, shall consist of two limbs closed at bottom, and joined above to a common neck fitted with a ground-glass stopper. The diameter of the limbs should be at least 2 cms. and their length at least 3 cms. The neck should be not less than 1.5 cms. in diameter. At the bottom of each limb a platinum wire of about 0.4 mm. in diameter is sealed through the glass To set up the cell, place in one limb mercury, and in the other hot liquid amalgam, containing 90 parts mercury and 10 parts zinc. The platinum wires at the bottom must be completely covered by the mercury and the amalgam respectively. On the mercury, place a layer one cm. thick of the zinc and mercurous sulphate paste described in 5. Both this paste and the zinc amalgam must then be covered with a layer of the neutral zinc sul- phate crystals one cm. thick. The whole vessel must then be filled with the saturated zinc sulphate solu- tion, and the stopper inserted so that it shall just touch it, leaving, however, a small bubble to guard against breakage when the temperature rises. Before finally inserting the glass stopper, it is to be brushed round its upper edge with a strong alcoholic solution of shellac, and pressed firmly in place. (For details of filling the cell see Note B.) 12 SYMBOLS, UNITS, INSTRUMENTS. r. rh >^ © -c fco'i* o> !•* « s 3 CO Is CO s w»§ O t-i SI £ in oi 2 o q sr q g ® H.P. sec, Eng- lish. lis CO i-l 11© 1 io id 1 s o CO © q ill. IO q ^ q , -d £ m si r~ £ CO O) « si oo rH © CO o g © -i° £a £ si" q 5 53 t2 C] m o£3 i-< beg —*|o SL 8 IO -H o 1 I ^ in c"t m Hr s|- ■<# r3 OS o o © S3 O ,H rH o © l"" o8g os c t> IO C 8 o s t~ p co © OS i 8 o d © § 'v.1 of cT CO o" CS © 8 d c 1 CO 8 1-1 rH '""' 1-1 1 L~ N -IS © CO CO q ^ 1 s d CO CO 8 *# 1 IO 1 ■* 1-1 1-1 rH ^ t- §8* 111 si- 8 OS co os 2 ^ in s o §iS © m o 1 © Oh^ CO ~ © ^S IO ,_, CO _ © ^_ m m IS l|s I1 q o oi ib in 8 o q III in 8 I CO © Kilo- gram- degree 2 3 3 ohms are now on the market, and are known as the Reichsanstalt form. They are made to carry very heavy currents. Fig. 23 shows such a resistance supplied with heavy contact terminals and a cooling coil. When this resistance is carrying a current, the drop between the two small terminals is such as would result from passing the same current through I0J*:\»i:it*. If one terminal of a source of E.M.F. be connected to a conductor, and the other terminal be connected to another con- ductor adjacent to the first but insulated from) it, it will be found that the two conductors ex- hibit a capacity for ab- sorbing a charge of elec- tricity that is somewhat; analogous to the filling ol a pipe with water before a pressure can be exerted The charge will remain ir the conductors after the, removal of the source oi ^ „, , supply. This capacity oil irj ilG. 22. Standard Resistance ('nils with "Wheat- rhe conductors to hole fet stone Bridge (Anthony Form;. under a given E.M.F. 2 |d| CONDENSERS. 29 charge of electricity is governed by the amount of surface exposed, by the nearness of the surfaces to each other, by the quality of the in- sulating material, and by the degree of insulation from each other. If the terminals of a battery be con- nected, through a battery and sensi- tive galvanometer, to a long sub- marine cable conductor and to the earth, it will be found that a very i considerable time will elapse before 1 the needle will settle down to a steady point. This shows that the cable insulation has been filled with electricity ; and it is common in so imeasuring the insulation resistance of a cable to assume a standard length iof time, generally three minutes, during which time such electrifica- tion shall take place. A condenser is an arrangement of metallic plates and insulation so made up that it will take a standard charge of electricity at a certain pressure. The energy represented by the charge seems to be stored up in the insulation between the conduct- ing plates in the form of a stress. This property of insulating materials to take on a charge of static electricity is known as inductive capacity, and a table in the section on the testing of capacity shows the specific in- ductive capacities of different substances. The unit of capacity is the international farad, which is defined as the capacity of a condenser which requires one coulomb (1 ampere for 1 second) to raise its potential from zero to one volt. Fig. 23. Figs. 24 and 25. Queen Standard Condensers. As the farad is far larger than ever is met in practice, the practical unit is taken as one-millionth farad or the micro-farad. The commercial standard most in use is the \ micro-farad, although adjustable condensers are often used, arranged so as to combine into many micro-farads or fractions of the same. Fig. 24 shows the ordinary a micro- farad condenser, and Fig. 25 one that is adjustable for different values. Diagram 26 shows an outline of the connections inside an adjustable con- denser. The ordinary commercial condenser is most usually made up of 30 SYMBOLS, UNITS, INSTRUMENTS. Fig. 27. Modified Mascart Electrometer. CONDENSERS. 31 sheets of tin foil separated from each other hy some insulator such as paraffined paper or mica. Every alternate sheet of foil is connected to a common terminal. As the capacity of a condenser depends upon the near- ness of the conductors to each other, and upon the area of the same, the insulating material is made as thin as possible, and still be safe from leakage or puncture. Many sheets of foil are joined together as described to make up the area. In adjustable condensers, the sheets are separated into bundles, and arranged so that any of them can be plugged in or out to add to or lessen the total capacity. If connected in multiple as shown, or if the positive side of one condenser be connected to the negative side of another, or a number of them are thus added together, then the condensers are said to be arranged in " cascade" or in series. This is seldom done unless it be to obtain greater variation in capacity. Electrometer. — Another instrument used somewhat in cable work, or where the measurement of electrostatic capacities or potentials is common, is the electrometer. A type of electrometer commonly used is the quadrant electrometer, for which we are indebted to Lord Kel- vin. The needle is a thin, flat piece of aluminium sus- pended in a horizontal po- sition by a thin metallic wire, in close proximity to four quadrants of thin sheet brass, that are supported on insulators without touching each other. Opposite quad- rants are connected by fine wires. A charge of elec- tricity is given the needle by connecting the suspension filament with a Leyden jar or other condenser. If the needle be charged positively it will be attracted by a negative charge and re- pelled by a positive charge. If, therefore, there be a dif- ference of potential between UCTA]¥CI3. Let Af=the mutual inductance between rJ^* two coils, Let L =z the self-inductance of one coil, Let Lt = the self-inductance of the other coil, Let Ln = the self-inductance of both coib connected in series, Let LJ/t = the self-inductance of both coil-:. connected in opposition to each other. Then, since Llt — L + Lx -f 2 M and L/// = L-\-L/ — 2M Another Method with battery is as follows : connect as in Fig. 30 where A and D are the two coils whose mutual induc- tance, Mv is required. R and Rx are two non-inductive resistances, and C is a con- MM 50 ME AS U REM EXTS . denser placed in shunt to R -\- Rv Closing and opening the key k produces deflections of the galvanometer G by the mutual induction of the coils and proportional to M — CRRV Varying Ogives different deflections in which, a being the first deflection and d[ a second deflection, M— CRRl _M— C\RRt d ~ dx d being the second value of the capacity of the condenser. Then M = CRRX when d is reduced to zero. measuring} the iufdhctawce ©e aerial mutes. In the following figure a line is shown Avith a load of lamps or other trans- lating devices, although for the purpose of getting the line inductance alone, it Avould most likely be i | | closed on itself. Connect up for a Wheat- stone's bridge method as shown in the cut ; close the key, and manipulate the sliiler p until a balance is obtained ; then vary the capacity of the condenser C until there is no movement of the needle when the battery circuit is broken with the key. Then, disregarding line ca- pacity, the inductance is L = cr2, and, if C =r capacity of the line, and R be the resistance of the same, then L = cr2 + § CR2. MEASUREMENT OF HCTCAI ITOUCTABfCE OE AERIAL LOES. To measure the mutual inductance of a pair of parallel lines, connect up as in the cut below. Earth both ends of each line separately, and, to avoid trouble from earth currents, put a small battery in secondary line with ad- justable shunt as shown. Adjust R and C until there is no movement of BATTERY the galvanometer needle, when the circuit of the battery is opened with the key ; then, if R = the resistance of the rheostat R as finally arranged, Rt = the resistance of secondary line, C — the capacity of the condenser as finally arranged, and M= mutual inductance, M = CRRX. MEASUREMENT OF CURRENT CIRCUITS. 51 CURREiTfT CIBCUITS. In circuits carrying alternating currents, and having an inductance in some part of their length, either in the shape of motors or other inductive load, as unloaded transformers, and the self-induction of the wires them- selves, the ordinary methods of measurement of the power or watts con- veyed are not available, as the current is seldom exactly in phase with the E.M.F., and therefore the value of the current multiplied by the E.M.F. will not be the true watts of the circuit. In all alternating circuits the power, at any instant of time, is equal to the product of the instantaneous values of the current and voltage at that time. If the current be in phase with the voltage, each will have zero values at the same instant of time, and will have maximum positive and maximum negative values simultaneously. Inasmuch as the product of two negative quantities is a positive quantity, the power of the circuit, Avith no phase dif- ference, is made up of positive pulsations varying in magnitude from 0 to a maximum. The latter is equal to the product of the maximum values of the current and E.M.F. If, however, the current differ by 90° in phase from Si* u." \ A » Uj =8 £ ~= ($ 1 M ly /f s^ V &/ k ♦; ■/i/ "« f /{ 9'L 5* \% Y 1 K\ K? ■* ^ ■©. \ ^ /\ 6 \ ^ the voltage, i.e., each having 0 value when the other has a maximum value, the power will consist of a series of pulsations, first positive and then nega- tive, and the algebraic value of the work done, i.e., power times its dura- tion, would be equal to zero. The result is that no permanent work is done, and the circuit is said to have a " Power Factor " of 0. The current which flows is called a wattless current. If the phase difference be less than 90° and more than 0°, at some instants of time the product of the volts and am- peres will be negative, but oftener will be positive. The fractional part of the whole which is positive is called the power factor. It can be shown that the power factor is equal to the cosine of the angle of phase difference. Inasmuch as an ampere of alternating current is one whose maximum value is 1.41 amperes (V^), and a volt of alternating current is one whose maximum value is 1.41 volts, the following relations hold true : — If True Watts =' /= maximum value of E.M.F., d =: maximum value of current, 6 = angle of lag of current behind the E.M.F., 2 - x Cos e. 52 M E A S V R E M E X T S . : E.M.F. by voltmeter : Vinean2, : current by ammeter : Vmean2, : angle of lag, z watts measured by watt meter,, then ; Cos 6 = Power factor, W - w Exl~ or tbe power factor is tbe value by which the observed volt-amperes must be multiplied to give the true watts. If a wattmeter be without self-induction in its fine wire coils, and tbe supporting part be not subject to eddy currents, then it may be used for measuring the value of power in A. C." circuits ; in, fact, in all full tests of alternating-current work it is necessary to have wattmeter, ammeter, and voltmeter readings. Three Voltmeter Method. Ayrton & Sumpner. This method is good where the voltage can be regulated to suit the load. In the above figure let the non-induc- tive resistance li be placed in series with the load a b ; take the voltage V across the terminals of li ; Vx across the load a b, and F2 across both, or from a to c. Then the J7 2 _ yi _ T'2 True watts — — Fig. 34. W- V 21i The best conditions are when V = > and, if li — \ ohm, Three Ampere Meter Method {not recommended). This method, due to Fleming, can be used when it is not convenient to regulate the potential of load a b. In Fig. 35 R is a non-inductive resistance _ x connected in shunt to the inductive load a b, with the three ammeters connected as shown, Then True watts = f (J22 — A2 — At2). Comhined Voltmeter and ter Method. Fig. 35. This method, devised also by Fleming, is quite accurate, and enables the accuracy of instruments in use to be a x checked. In Fig. 36 R is a non-inductive resistance connected in shunt to the induc- tive load a b, and the voltmeter V measures the p. d. across x y. A and Ax are ammeters connected as shown ; then True watts = f (a? — An- — (^f) ■ If the voltmeter V takes an appreciable amount of current, it may be tested as fol- lows : disconnect E and Fat y, and see that A and .-/, are alike ; then con- nect /,' and V at y again, and "disconnect the load a b. Then At = current taken by li and V in multiple. As regards all the above mentioned tests with 3 voltmeters, ammeters, etoi, it may be said that they were developed at a time when no good alternating current instruments were available. Since then a number of good A. C. voltmeters have been developed, and more recently the inclined coil instru- ments of the General Electric Co., and Sehallenberger instruments of the Westinghauss Co., have placed instruments in our hands that make alternat- ing-current testing nearly as easy as d. c. testing. TESTS WITH VOLTMETER. 53 TESTS WITH VOLTMETER. The following are a few of the more important tests for which a voltmeter is especially adapted, and have mostly been condensed from a very fine article by H. Maschke, Ph.D. published in the Electrical World in April, 1892. The scales of the better known portable instruments of to-day read in gen- eral from 0 to 150 volts, or from 0 to 750 volts, and in special instruments the two scales are combined, so that by connecting one wire to one or the other of two binding posts either scale is available. Instruments for battery use read from 0 to 15 volts with a second scale reading as low as J-0, or 1.5 volts. Millivoltmeters reading from 0 to J^, or 0 to TJn, etc., with divisions capable of being read as low as TooVoo volt, are also obtainable. None of the refined laboratory methods will be given here, as the reader is referred to the text-books for such tests. ELECTROMOTIVE FORCE OE BATTERIES. The positive post of voltmeters is usually at the right, and marked -4- In a battery the zinc is commonly neg- ative, and should therefore be" con- nected to the left or negative binding post. For single cells or a small number, a low-reading voltmeter, say one read- ing to 15 volts, will be used, the con- nections being as per diagrams. FJi|i|i|i|i|i|( + Fig. 37. Fig. 38. ELECTROMOTIVE EORCE OE D1SAMOS. For voltage within range of the instrument available for the purpose, it is only necessary to connect one terminal of the voltmeter to a brush of one polarity, and the other terminal to a brush of the opposite polarity, and read direct from the scale of the instrument. As continuous current volt- meters usually deflect forward or back according to which pole is connected, it is necessary sometimes to reverse the lead wires, in which case the polar- ity of the dynamo is also determined. Of course the voltage across any cir- cuit may be taken in the same way, or the dynamo voltage may be taken at the switchboard, in which case the drop in the leads sometimes enters into the calculations. Following are diagrams of the connections to bipolar and multipolar dynamos : — /\4- m Fig. 39. Fig. 40. In the case of arc dynamos or other machines giving high voltage, it is necessary to provide a multiplier in order to make use'of the ordinary in- strument; and the following is the rule for determining the resistance which, when placed in series with the voltmeter, will provide the necessary multiplying power. u MEAS HKEMENTS. e = upper limit of instrument scale, for example 150 volts, E = upper limit of scale required, for example 750 volts, R =z resistance of the voltmeter, for example 18,000 ohms, r = additional resistance required, in ohms. r = R ^ll or r — 18,000 75°~„15° = 72,000 ohms. 150 The multiplying power : E 750 B Should the exact resistance not be available, then with any available resistance i\ the regular scale readings must be multiplied by ( -^ + 1 ) • IMPOBTOCE ©_F HriCJH BHIISTAH'CE I OH VOXTMETEKSi. It is highly important, as reducing the error in measurement, that the in- ternal resistance of a voltmeter be as high as practicable, as is shown in the following example : — Let E in the figure be a dynamo, battery, or other source of electric energy, sending current through the resistance r ; and vm. be a voltmeter indicating the pressure in volts between the terminals A and B. Be- fore the vm. is connected to the terminals A and B there will be a certain difference of potential, which will be less Avhen the voltmeter is connected, owing to the les- sening of the total resistance between the two points ; if the resistance of the vm. be high, this difference will be very small, and the higher it is the less the error. Following are the formulas and computations for de- termining the error. In the above figure let E be the E.M.F. of the dynamo, r the resistance of the circrut as shown between A and B, and r-. be the resistance of the leads A and B plus that of the dynamo, and let R be the resistance of the voltmeter ; then before the vm. is connected the difference between A and B will be — 'Wvwvw — , Fig. 41. r -f-rj X E, and after connecting the voltmeter it will be R X r 1 R X r + r X rx + rx X R The difference between the two results e and ex is then X - " X rx - X elt and this difference will be smaller the greater the resistance R of the vn Example : — Let E =10 volts r = 10 ohms rx = 2 ohms R = 500 ohms then and 500 X 10 X 500 X 10 = .0333. 1 — 500 X 10 + 10 X 2 + _ _ 2_ 10 X 2 6 ~ Cl - 500 X 10 + 2 If R be made 1000 ohms, then 1000 x 10 €l ~" 1000 X 10 + 10 X 2 -)- 2 X 1000 X 10 = 8.3056, X 10 = 8.32, S ~~ el — -mnn X : ,10X2 X 10 = .0166, POMPARISOX OF E.M.F. OF BATTERIES. or just one half of fche error ; it may be said that the error is therefore in inverse proportion to the resistance of thevm. If the error of measurement is not to exceed a stated per cent p, then r and r, must be such that — , — 1 is smaller than „.,„'' ohms. 1 r + rt 100 It the circuit is not closed by a resistance i between A and B then with vm. connected and the error between the true value and that shoAvn on the vm. is and this error decreases in inverse proportion to the increase of the ratio between It and the internal resistance of the current generator rx. If the error is not to exceed p per cent, then the internal resistance rx must be less than ohms. TheE.M.F. of high-resistance cells cannot be correctly measured by the above method, even with voltmeters of relatively high resistance, but it is better done by one of the methods mentioned below. COMPARISON OIT E.M.I\ &W BATTERIES. Wlieatstf one's JfletStod. — To compare E.M.F. of two batteries A and A", with low-reading voltmeters, let E be the E.M.F. of A ; and E, the E.M.F. of X. — n/VVVVVVVVV Fig. 42. First connect battery A in series with the voltmeter and a resistance r, switch B being closed, and note the deflection V; then open the switch B, and throw in the resistance rx, and note the deflection Vx. Now connect bat- tery X in place of A, and close the switch B, and vary the resistance r until the same deflection F"of voltmeter is obtained and call the new resistance r2 ; next open the switch B, or otherwise add to the resistance r2 until the deflec- tion Vx of the voltmeter is produced ; call this added, resistance r3, then E:E1::r1 : ra. If E be smaller than Et, the voltmeter resistance R may be taken as r, and it is better to have rx about twice as large as the combined resistance of r and the resistance of A. It is not necessary that the internal resistance of the cells be small as compared with B. Poggcmlorff's Ifletliod Modified Iby Clark. To Compare the E.M.F. of a battery cell or element with a standard cell. Let S be a standard cell, Tbe a cell for comparison with the standard, B~be a, battery of higher E.M.F. than either of the above elements. A resistance r is joined in series with the battery B and a slid© wire A D. A millivoltmeter is connected as shown, both its terminals being connected to the like poles of the battery B and the Standard *9. ia 56 MEASUREMENTS. FlG. 43. Move the contact C along the wire until the pointer of the instrument stands at zero, and let r1 be the resistance of A C. Throw the switch b so as to cut out the standard S, and cut in the cell T ; now slide the contact 6\ along the wire until the pointer again stands at zero, and call the resistance of A C\ r2, Then the E.M.Fs. of the two cells T: S ::r2 : rv If a meter bridge or other scaled wire be used in place of A D, the results may be read directly in volts by arranging the resistance r so that with the pointer at zero the contact C is at the point 144 on the wire scale, or at 100 times the E.M.F. of the standard S, which may be supposed to be a Clark cell. All other readings will in this case be in hundredths of volts ; and should the location of Cx be at 175 on the scale when the pointer is at zero on the voltmeter, then the E.M.F. of the cell, being compared, will be 1.75 volts. MEASIIRIIC} CUMIIE]¥T STRENGTH WITH A. TOLTMETER. If the resistance of a part of an electric circuit be known, taking the drop in potential around such resistance will determine the current flowing by ohms law viz., I— — . In the figure let r be a known resistance be- tween the points A and B of the circuit, and / the strength of current to be determined ; then if the voltmeter, connected as shown, gives a deflection of V volts, the current flowing in r V will be 1= — . For the corrections to be applied in certain cases, see the section on Importance of High Resistance for Voltmeters. Always see that the resistance r has enough carrying capacity to avoid a rise of temperature which would change its resistance. If the reading is exact to — volt the meas- p j Fig. 44. urement of current will be exact to am- p X r peres. If r = .5 ohm, and the readings are taken on a low-reading volt- meter, say ranging from 0 to 5 volts, and that can be read to ^ volt, then the possible error will be 300 x .5 — 150 MEASURING RESISTANCE WITH A VOLTMETER. 57 If r be made equal to 1 ohm, then the volts read also mean amperes. Measurement of "Very Heavy Currents with a. Milli- voltmeter. For this purpose the method outlined above is most generally used with the substitution of a millivoltmeter for the voltmeter. Where portable instruments are used, there must be a calibrated shunt fur the millivoltmeter, the shunt being made up of a metal that does not vary in resistance with change of temperature, and which is placed in series in the circuit, the millivoltmeter simply giving the drop around this shunt, its scale being graduated in amperes. For switchboard instruments the method is the same, being varied some- times by using as a shunt a measured part of a conductor or bus bar in place of a special resistance. MEASURING RESISTAIIfCE WITH A VOLTMETEB. General Methods. — In the figure, let Ir: the unknown resistance that is to be measured, r = a known resistance, E, the dynamo or other steady source of E.M.F. When connected as shown in the figure, let the voltmeter reading be V; then connect the voltmeter terminals to r in the same manner and let the reading be Vx ; then X:r\: V : Vx and x=r^-^. If, for instance, r = 2 ohms and V = 3 volts and V1 = 4 volts then v 2x3 .' . X =. — - — = 1.5 ohms. If readings can be made to rJ-n volt, the error of resistance measurement will then be 100 XTb (-y + jr) Per cent. and for the above example would be 1 (J + A) = 0.58%. Should there be a considerable difference between the magnitudes of the two resistances X and r, it might be better to read the drop across one of them from one scale, and to read the drop across the other on a lower scale. Resistance Measurement with Voltmeter and Ammeter. The most common modification of the above method is to insert an am- meter in place of the resistance r in the last figure, in which case X=-j. where /is the current flowing in amperes as read from the ammeter. If the readings of the voltmeter be correct to T-J-ff and the ammeter read- ings be correct to the same degree, the possible error becomes : 100 x (iniU+ tsrt) = Per cent- measurement of Very Small Resistances with a HEillivolt- meter and Ammeter. By using a millivoltmeter in connection with an ammeter, very small re- sistances, such as that of bars of copper, armature resistance, etc., can be accurately measured. 58 M E A S U R K1\I E N TS . Fig. 46. In order to have a reasonable degree of accuracy in measuring resistance by the "drop" method, as this is called, it is necessary that as heavy cur- rents as may be available be used. Then, if E be the dynamo or other source of steady E.M.F., X be the required resistance of a portion of the bar, /' be the drop in potential between the points a and b, and 1 be the current flow- ing in the circuit as indicated by the ammeter, then ' x=f The applications of this method are endless, and but a few, to which it is especially adapted, need be mentioned here. They are the resistance of . armatures, the drop being taken from opposite commutator bars and not from the brush-holders, as then the brush-contact resistance is taken in ; the resistance of station instruments and all switchboard appliances, such as the resistance of switch contacts ; the resistance of bonded joints on electric railway work, as described in the chapter on railway testing. .vS«'i«.*(as-«»iii€*nt of Mig-le Resistances. With the ordinary voltmeter of high internal resistance, let R be the re- sistance of the voltmeter, X be the resistance to be measured. Connect them up in series with some source of electro- motive force as in the following figure. Close the switch b, and read the voltage V with the resistance of the voltmeter alone in circuit ; then open the switch, thus cutting in the resistance A", and take another reading of the voltmeter, Vr Then X — r[Z.— \\. If the readings of the voltmeter be cor- rect to ^ of a volt the error of the above 10 / V + V, result will be g(£^)> CTEASirnilfG ^ME JATSA2,AT*«;V OF HftEffT»'» A I¥I» POWER CIBCIITS WITH A iOITMETEH. For rough measurements, where the exact insulation resistance is not re- quired, but it is wished to determine if such resistance exceeds some stated figure or rate, then the method above given will do, when applied as fol- lows : — Let X = insulation resistance to ground as in figure, X, = insulation resistance to ground of opposite lead, /(• = resistance of voltmeter, V— potential of dynamo E, V, = reading of voltmeter, as connected in figure, Vtl =r reading of voltmeter, when connected to opposite lead. MEASURING THE INSULATION. 59 "^r Ground FIG. 48. and X, = R\ V„ The above formula can be modified to give results more nearly correct by taking into account the fact that the path through the resistance R of the voltmeter is in parallel with the leak to ground on the side to which it is connected as shown in the following figure : — In this case the voltage V of the circuit will not only send current through the lamps, but through the leaks e f to ground, and through the ground to d and c, thence through d to b, and'c to a, these two last paths being in par- allel, therefore having less resistance than if one alone was used ; thus if r be the resistance of the ground leak b d, and rx be the resistance of the leak l' ,/", and R be the resistance of the voltmeter, then the total resistance by way of the ground, between the conductors, would be R X r R + r^r» and if V= voltage of the circuit, v = reading of voltmeter from a to c, v, = reading of voltmeter from q to c. Then r = R(V-

tJ»:« Similar Device where Both Terminal!* are on the Same Base. Let ff be fuses in place on a base, V — potential of circuit, R = resistance of voltmeter, v = reading of voltmeter, required the resistance r across the base a a, to b &.. V— v Then r — R . MEASUH1XG THE IWS1T- Fig. 51. LATIOli OF BYlfAMOS. The same formula as that used for measuring high resistances (see Fig. 47) applies equally well to determining the insulation of dynamo conductors from the iron body of the machine MEASURING THE INSULATION RESISTANCE. 61 Connect, as in Fig. No. 52, all symbols having the same meaning as before. Let r = insulation resistance of dynamo, then r^-(f-l). MEA§rKIIIfG THE IlillATIOlf RESISTANCE ©e MOTORS. Where motors are connected to isolated plant circuits with known high insulation, tbe formula vised for insulation of dynamos applies ; but where tbe motors are connected to public circuits of questionable insulation it is necessary to first determine the circuit insulation, which can be done by using the connections shown in Fig. 48. Fig. 53 shows tbe connections to motor for determining its insulation by current from an operating circuit. Here, as before, the insulation total connected devices If r = total resistance of circuit and motor in multiple to ground, and r, is the insulation of the circuit from ground, then X, the insulation of the motor will be X= — . MEASUREMENT OF THE RESISTANCE OF THE HITMAN BODY. The jars jj of the following figure (No. 51) are filled with a weak solution of caustic potash ; the person whose resistance is to be measured places his "lands in the jars, if the measurement is to be made from hand to hand, or 62 MEASUREMENTS. makes an equally good connection with, any other desirable portion of the body. First take a reading of the voltmeter with the switch K closed ; then voltmet. subject ■■■it plunge his hands jars, open the switch A, -' another reading of the r. The resistance r of the ill be ■ft-* in which It is the resistance of voltmeter, Fis the reading of volt- meter alone, V, is the reading of volt- meter with switch K open and the subject in series with volt- meter. 1HEASVRE1HEIIT OF THE OTEBIA1 RE§I§TAHfCE OF A BATTFRT. 3 following figure (No. 55), let E be the cell or battery whose resistance is to be measured, A' be a switch, and r a suitable resistance. Liet V = the reading of voltmeter with the kev, A', open (this is the E.M.F. of the battery), and V, — the reading of voltmeter with key, A', closed (this is the drop across the re- sistance r), Then the battery resistance — yp — - ohms. Fig. 55. -r x - nieyiexn-froelicii keihod. In the following figure (No. 56), let E be the cell or battery to be measured, K a switch for closing resistance r to B or c ; r, rx and r2 be suitable resis- *■. tances connected as' shown. The volt- meter should of course be a low-reading one. Close by the key A", A and c, and read the voltmeter ; next close by the key A', A and B, and rend the volt- meter; then adjust r2 until the volt- meter reading is the same for either position of the key K, and r., is then equal to the resistance of the battery E. In most cases it is best to connect some known resistance in series with the cell, so that the current may not be excessive and harm the celV ; if this be done, of course it is necessary to deduct this known resistance from the final reading r2. Fig. 56. COSDECTIVITY WITH A MILIIVOITIHETEB. This is a quick and convenient method of roughly comparing the conduc- tivity of a sample of metal with that of a standard niece. In" Fig. 57, R is a standard bar of copper of 100% conductivity at 70° F. ; this bar may be of convenient length for use in the clamps, but of known CONDUCTIVITY WITH A MILLIVOLTMETER. 63 cross section. X is the piece of metal of unknown conductivity, but of the same cross section as the standard. E is a source of steady current, and if a storage battery is available it is much the better for the purpose. M is a millivoltmeter with the contact device d. The distance apart of the two points may be anything, so long as it remains unaltered and will go between the clamps on eitner 01 the bars. Now with the current tioAving through the two bars in series the fall of potential between two points the same distance apart and on the same flow- FlG, 57. line will, on either bar, be in proportion to the resistance, or in inverse pro portion to the conductivity ; therefore by placing the points of d on the bars in succession, the readings of the millivoltmeter will give the ratio of the conductivities of the two pieces. For example : — if the reading from B = 200 millivolts, and tbe reading from X = 205 millivolts, then the percentage conductivity of X as compared with R is 205 : 200 : : 100 : conductivity of X, 200 X 100 Q„ 205 =97-5%° MAGNETIC PROPERTIES OP IRON. With a given excitation the flux or flux-density (B of an electromagnet will depend upon the quality of the iron or steel of the core, and is usually rated as compared with air. If a solenoid of wire be traversed with a current, a certain number of magnetic lines of force, 3£,will be developed per square centimetre of the core of air. Now, if a core of iron be thrust into the coil, taking the place of the air, many more lines of force will flow ; and at the centre of the solenoid these will be equal to (^ lines per square centimetre. As iron or steel varies considerably as to the number of lines per square centimetre (ft which it will allow to traverse its body with a given excitation, its conductivity towards lines of force, which is called its permeability, is numerically represented by the ratio of the flux-density when the core is present, to the flux-density when air alone is present. This permeability is represented by /u.. The permeability /x of soft wrought iron is greater than that of cast iron ; and that for mild or open-hearth annealed steel castings as now made for dynamos and motors is nearly, and in some cases quite, equal to the best soft wrought iron. The number of magnetic lines that can be forced through a given cross- section of iron depends, not only on its permeability, but upon its satura- tion. For instance, if but a small number of lines are flowing through the iron at a certain excitation, doubling the excitation will practically double the lines of force ; when the lines reach a certain number, increasing the excitation does not proportionally increase the lines of force, and an excita- tion may be reached after which there will be little if any increase of lines of force, no matter what may be the increase of excitation. Iron or steel for use in magnetic circuits must be tested by sample before any accurate calculations can be made. Data for (B-3C Curves. Average First Quality American Metal. (Sheldon.) d A Cast Iron. Cast Steel. Wrought Iron Sheet Metal. -j H z!J 2£s CO M 53 . w X CD . j oi X a> ^ &a'1 2 \\\ i g > \\ ? _J D ^rr 1- ?\ tt *\ u C5 < i\ I DC UJ <■ i\ \\ \ \\ \i ^ \ \ \\ V ssS . T-r ^1 honi auvn&s aad s-naMxvwoira Fig. 1. Magnetic Properties of Iron. 6Q MAGNETIC PROPERTIES OF IRON. In large generators, having toothed armatures and large flux densities in the air-gap, the flux is carried chiefly by the teeth. This results in a very high tooth flux density, and a corresponding reduced permeability. The related values of (g,3Cand m. are given in the following table. These values are for average American sheet metal. Permeability at Hig-h Flux Densities. Ampere Ampere (B Kilomax- 3C Turns per Turns per Kilo- wells per V- cm. Length. Inch Length. grammes. Square in. 200 159 404 19.8 127 99.0 400 318 808- 21.0 135 52.5 600 477 1212 21.5 138 35.8 800 637 1616 21.8 140 27.3 1000 795 2020 22.0 142 22.0 1200 954 2424 22.3 144 1.8 1400 1113 2828 22.5 145 1.6 METHODS OF HG1EK.W B XI .X« THE HACKETIC 4tr.4LITIKM OF IRON AID §TF£I. The methods of determining the magnetic value of iron or steel for elec- tro-magnetic purposes are divided by Prof. S. P. Thompson into the follow- ing classes : Magnetometric, Balance, Ballistic, and Traction. The first of these methods, now no longer used to any extent, consists in calculating the magnetization of a core from the deflection of a magneto- meter needle placed at a fixed distance. In the Balance class, the deflection of the magnetometer needle is bal- anced by known forces, or the deflection due to the difference in magnetiza- tion of a known bar and of a test bar is taken. The Ballistic method is most frequently used for laboratory tests, and for such cases as require considerable accuracy in the results. There are really two ballistic methods, the Ring method and the Dirided-bar method. In either of these methods the ballistic galvanometer is used for measur- ing the currents induced in a test coil, by reversing the exciting current, or cutting the lines of force. Ring- Method. — The following cut shows the arrangement of instru- ments for this test, as used by Prof. Rowland. The ring is made of the sample of iron which is to undergo test, and is uniformly wound with the CELLS ^=- l_ < Fig. 2. Connections for the Ring Method. exciting coil or circuit, and a small exploring coil is wound over the excit- ing coil at one point, as shown. The terminals of the latter are connected to the ballistic galvanometer. MAGNETIC TEST METHODS. 67 The method of making a test is as follows : — The resistance, R, is adjusted to give the highest amount of exciting cur- rent. The reversing switch is then commutated several times with the gal- vanometer disconnected. After connecting the galvanometer the switch is suddenly reversed, and the throw of the galvanometer, due to the reversal of the direction of magnetic lines, is recorded. The resistance, R, is then adjusted for a somewhat smaller current, which is again reversed, and the galvanometer throw again recorded. The test is carried on with various exciting currents of any desired magnitude. In every case the exciting cur- rent and the corresponding throw of the galvanometer are noted and recorded. If i — amperes flowing in the exciting coil, nx = number of turns of wire in exciting coil, I = length in centimetres of the mean circumference of the ring, then the magnetizing force ae= § xf or 1.257 x«. If I" = length of the ring in inches, then 0C"=.495X^. If 0 = the throw of the galvanometer, K= constant of the galvanometer, R = resistance of the test coil and circuit, n2 = number of turns in the test coil, a = area of cross-section of the ring in centimetres, then ^p 10s RK9 To determine K, the constant of the galvanometer, discharge a condenser of known capacity, which has been charged to a known voltage, through it, and take the reading 01, then If c — capacity of the condenser in microfarads, e = volts pressure to which the condenser is charged, then the quantity passing through the galvanometer upon discharge in coulombs is Q^i^oo- and the galvanometer constant ~ 1,000,000 01' Dividecl-lSar Uletliod. — As it is often inconvenient or impossible to obtain samples in the form of division in» -ammeter a ring, and still more incon- venient to wind the coils on it, Hopkinson devised the di- vided-bar method, in which the sample is a long rod \" diameter, inserted in closely fitting holes in a heavy wrought iron yoke, as shown ■ C0ILS in the following cut. _ /~^\ v to v--meah.cength In the cut the exciting coils \£_J l of test piece! are in two parts, and receive ballistic current from the battery and galvanometer . , ,. through the ammeter, resist- FlG. 3. Arrangement for Hopkinson s ai- ance, and reversing switch, vided-bar method of measuring permea- as shown. bility. _ The test bar is divided near the centre at the point indicated m tlie cut, and a small light test coil is placed over it, and so arranged with springs as 68 MAGNETIC PROPERTIES OF IRON. to be thrown clear out of the yoke when released by pulling out the loose end of the test bar by the handle shown. In operation, the exciting current is adjusted by the resistance J?, the test bar suddenly pulled out by the handle, thus releasing the test coil and pro- ducing a throw of the galvanometer. As the current is not reversed, the induced pressure is due to jVonly, and the equation for (^ is 10» R K t and X— TTv X "Where L = the mean length of the test rod as shown in the cut. In using the divided-bar method, a correction must be made, for the rea- son that the test coil is much larger than the test rod, and a number of lines of force pass through the coil that do not through the rod. This cor- rection can easily be determined by taking a reading with a wooden test rod in place of the metal one. An examination of the cut will show that the bar and yoke can also be used for the method of reversals. The fourth or Traction class is exceedingly simple, and was devised by Prof. Silvanus P. Thompson. The following cut shows the method with sufficient clearness. A heavy yoke of wrought iron has a small hole in one end through which the test rod is pushed, through the exciting coil shown, and against the bottom of the yoke, which is surfaced true and smooth, as is the end of the test rod. In operation, the exciting current is ad- justed by the resistance 11, and the spring balance is then pulled until the sample or test rod separates from the yoke, at which time the pull in pounds necessary to pull them apart is read. Then (B = 1,317 X sll + JC- Fig. Where P = pull in pounds as shown on the balance, A = area of contact of the rod and yoke in square inches. J(Ms found as in the Hopkinson method preceding this. Following is a description of a practical adaptation of the permeameter to • shop-work as used in the factory of the Westinghouse Electric and Manu- facturing Co. at Pittsburgh, Pa. S. P. Thompson's per- meameter. The Permeameter, as used Ity the IVestiiigrhouse Electric and Mfg-. Co, Design and Description prepared by Mr. C. E. Skinner. A method of measuring the permeability of iron and steel known as the " Permeameter Method " was devised by Prof. Silvanus P. Thompson, and is based on the law of traction as enunciated by Clerk Maxwell. According to this law the pull required to break any number of lines of force varies as the square of the number of lilies broken.' (A complete discussion of the theory of the permeameter, with the derivation of the proper formula for calculating the results from the measurements will be found in the " Electro Magnet," by Prof. S. P. Thompson., A permeameter which has been in use for several years in the laboratory of the Westinyhouse Electric and Manufacturing Company, and which has given excellent satisfaction, is shown in the accompanying drawings. The THE PERMEAMETER, 69 yoke, A, consists of a piece of soft iron 1" x %\" x 2J", with a rectangular opening in the centre 2J" x 4". The sample, X, to he tested is %" in diam- eter and 1\" long, and is introduced into the opening through a %" hole in the yoke, as shown in the drawing. The test sample is finished very accurately to \" in diameter, so that it makes a very close tit in the hole in the yoke. The lower end of the opening in the yoke and the lower end of the sample are accurately faced so as to make a perfect joint. The upper end of the sam- ple is tapped to receive a \" screw %" long, twenty threads per inch, by means of which a spring balance is attached to it. The magnetizing coil, C, is wound on a brass spool, S, 4" long, with the end flanges turned up so that it may be fastened to the yoke by means of the screws. The axis of the coil coincides with the axis of the yoke and opening. The coil has flexible leads, which allow it to be easily removed trom the opening for the inspection of the surface where contact is made between the yoke and the test sample. The spring balance, F, is suspended from an angle iron fastened to the up- right rack, 7, which engages with the pinion. J. The balance is suspended exactly over the centre of the yoke through which the sample passes, to avoid any side pull. A spring buffer, K. is provided, which allows perfectly free movement of the link holding the sample for a distance of about \'f, and then takes up the jar consequent upon the sudden release of the sample. The frame, B, which supports the pulling mechanism, is made of brass, and has feet cast at the bottom, by means of which the complete apparatus is fastened to the table. Two spring balances are provided, one reading to 30 lbs. and the other to 100 lbs. These spring balances are of special construc- tion, having comparatively long scales. (They were originally made self- registering ; but this was found unnecessary, as a reading could be taken with greater rapidity and with sufficient accuracy without the self-register- ing mechanism.) Any good spring balance may be used. The spring should be carefully calibrated from time to time over its Avhole range ; and if there is a correction it will be found convenient to use a calibration curve in cor- recting the readings. With a sample \" in diameter, or § of a square inch area cross-section, the maximum pull required for cast iron is about 25 lbs., and for mild cast steel about 70 lbs. With the number of turns on the coil given above, the current required for obtaining a magnetizing force of JC^ 300, is about 12.5 amperes. This is as high a value as is ever necessary in ordinary work. For furnishing the current a storage battery is ordinarily used, and the variations made by means of a lamp board which has in addition a sliding resistance, so that variations of about .01 ampere may be obtained over the full range of cur- rent from 0.1 ampere to 12.5 amperes. The operation of the permeameter is as follows : — The sample to be tested is first demagnetized by introducing it into the field of an electro-magnet with a wire core, through which an alternating current is passing, and gradually removing it from the field of this electro- magnet. The sample is then introduced into the opening in the yoke, care being taken to see that it can move without friction. Measurements are taken Avith the smallest current to be used first, gradually increasing to the highest value desired. In no case should a reading be taken with a current of less value than has been reached with the sample in position, unless the sample is thoroughly demagnetized agaiti before reading is taken. It is usually most convenient to make each successive adjustment of cur- rent with the sample out of position, then introduce the sample and give it a half turn, to insure perfect contact between the sample and tiie yoke. The lower end of the sample and the surface on which it rests should be care- fully inspected to see that no foreign matter of any kind is present which might introduce serious errors in the measurements. The pull is made by turning the pinion slowly by means of a handle, E, carefully noting each position of the index of the spring balance as it advances over the scale, and noting the point of release. The mean of three or four readings is usually taken as the corrected value for pull, the current in the coil remain- ing constant. With practice the spring balance can be read to within less than 1% ; and as the square root of the pull is taken, the final error becomes quite small, especially with high readings. The evaluation of the results for the above permeameter are obtained by the use of the following formula : — .70 MAGNETIC PROPERTIES OF IRON. Where % = number of turns in the magnetizing coil = 223, i = current in amperes, I = length of magnetic circuit in centimetres, estimated in this case as 11.74. Substituting the known values in the above formula we have 3C = 23.8 i. The number of lines of force per square centimetre, (B = 1,317 y/^ + OC- Where P = pull in lbs. A = area of the sample in square inches = 0.3068. ^fC^ value of the magnetizing force for the given pull. THE PEEMEAMETER. fl Substituting the value of A in the above formula we have (£=:2,3S0Vp + 3e. There are several sources of error in measurements made by the permea- meter which should be carefully considered, and eliminated as far as possible. a. The unavoidable air gap between the sample and the yoke where it passes through the hole in the upper part of the yoke, together with the more or less imperfect contact at the lower end of the sample, increases the magnetic reluctance and introduces errors for which it is impossible to make due allowance. By careful manipulation, however, these can be reduced to a minimum, and be made practically constant. b. As the magnetization becomes greater the leakage at the lower end of the sample increases more rapidly ; and there is considerable error at very high values from this source, as the leakage lines are not broken with the rest. c. Errors in the calibration and reading of the spring balance. None but the best quality of spring balance should be used, and the average of several readings taken with the current remaining perfectly constant for each point on the (&-3C curve. As the square root of the pull is taken, the errors due to reading the spring balance make a larger and larger percent- age error in (R as P approaches zero, thus preventing accurate determina- tions being made at the beginning of the curve. 72 MAGNETIC PROPERTIES OF IRON. From the above it will be seen that the permeameter is not well adapted for giving the absolute values of the quality of iron and steel, but is especially suitable for comparative values, such as are noted in ordinary work, where a large number of samples are to be quickly measured. A complete curve can be taken and plotted in ten minutes. By suitable comparison of known samples measured by more accurate methods, the permeameter readings may be evaluated to a sufficient degree for use in the calculations of dynamo electric machinery. CORE BOSSES. These result from Hysteresis and Eddy currents. Professor Ewing has given the name Hysteresis to that quality in iron which causes the lagging of the induction behind the magnetic force. It causes a loss when the direction of the induction is reversed, and results in a heating of the iron. It increases in direct proportion to the number of reversals, and as the 1.6th power of the maximum value of the induction in the iron core. The heat produced has to be dissipated either by radiation or conduction, or by both. Steinmetz gives the following formula for hys- teresis loss in ergs per cubic centimeter, of iron per cycle ; h = -q (ft1-6, where 17 = a constant depending upon the kind of iron. M j-*teretif Constants tor Different Mat a* rial*. Material. Very soft iron wire . . Very thin soft sheet iron Thin good sheet iron . . Thick sheet iron . . . Most ordinary sheet iron Transformer cores . . Soft annealed cast steel Soft machine steel . . Cast steel Cast iron Hardened cast steel . . Hysteretic Constant. .002 .0015 .003 .0033 .004 .003 .008 .0094 .012 .016 .025 Eddy Currents are the local currents in the iron core caused by the E.M.F's generated by moving the cores in the field, and increase as the "square of the number of revolutions per second. The cure is to divide or laminate the core so that currents cannot flow. These currents cause heating, and unless the core be laminated to a great degree, are apt to heat the armature core so much as to char the insulation of its windings. Wiener gives tables showing the losses by Hysteresis and Eddy currents at one cycle per second, under different conditions. These are changed into any number of cycles by direct proportion. Following are the tables : — CORE LOSS. 73 Hysteresis factors for ^Different Core Densities. (Wiener.) Watts dissipated at A r Watts dissipated at A FREQUENCY OF ONE A a frequency of one COMPLETE MAGNETIC b1-! yA 2 ~ S e> ~ o W H « fc m ° COMPLETE MAGNETIC an K m o u e 0«H CYCLE PER SECOND. CYCLE PER SECOND. Sheet iron. Iron wire. Sheet iron. Iron wire. W change of (B, ($>. Values of Hare obtained from the formula, 47r n % JC = ' , when I = average circumference of the test ring. Change of (Bis obtained by the formula, 10s R K 9 a n2 ' where all letters have the same significance as in the formula on page 67. Remember that we started in our test with a maximum unknown value of (B, and that we gradually decreased this by steps measurable by the throw of the galvanometer, and that we afterwards raised the (Bin an opposite direc- tion to the same maximum unknown value, and still further reduced this to zero, and after commutation produced the original maximum value. Ac- cording to this, if due consideration be paid to the sign of the (B which is determined by the direction of the galvanometer throw, the algebraic sum of the changes in (B should be equal to zero ; the algebraic sum of the first or second half of the changes in (B should be equal to twice the value of the original maximum, (B- Taking this maximum value as the first under the column of the table headed (B, and applying algebraically to this the changes in (B for successive values, we obtain the completed table. Plot a curve of 3Cam"l(B* Tne area enclosed represents the energy lost in carry- ing the sample through one cycle of magnetization between* the maximum limits -(-(Band — (B- Measure this area, and express it in the same units as is employed for the co-ordinate axes of the curve. This area divided by 4ir CORE LOSS. 75 gives the number of ergs of work performed per cycle upon one cubic centi- meter of the iron, the induction being carried to the limits -f- (Band — (B- The "Wattmeter Method of Hysteresis Tests. Inasmuch as the iron, a sample of which is submitted for test, is generally to be employed in the manufacture of alternating-current apparatus, it is desirable to make tbe test as nearly as possible under working conditions. If the samples be disks, as in the previous method, and these be shellacked on both sides before being united into the composite test-ring in order to avoid as much as possible foucault current losses, the test can be quickly made according to the method outlined in the following diagram : — Fig. 8. Wattmeter Test for Hysteretic Constant. Alternating current of / alternations per second is sent through the test- ring. Its voltage, E, and current strength, i, are measured by the alternating- current voltmeter, V, and ammeter, A. If r be the resistance of the test- ring coil of 7i1 turns, then the watts lost in hysteresis W, is equal to the wattmeter reading W — i2r. If the volume of the iron be V cubic centi- meters, and the cross section of the iron ring be a square centimeters, then Steinmetz's hysteretic constant V = - 107 W Vf 'V2tt nxfa\ EW ) Foucault current losses are neglected in this formula, and the assumption is made that the current is sinusoidal. Kwiisg-',-* Hysteresis Tester. — In this in- strument, Fig. 9, the test sample is made up of about seven pieces of sheet iron §" wide and 3" long. These are rotated between the poles of a permanent magnet mounted on knife edges. The magnet carries a pointer which moves over a scale. Two standards of known hyster- esis properties are used for reference. The de- flections corresponding to these samples are plotted as a function of their hysteresis losses, and a line joining the two points thus found is referred to in subsequent tests, this line show- ing the relation existing between deflection and hysteresis loss. The deflections are practically tbe same, with a great variation in the thick- ness of the pile of test-pieces, so that no cor- rection has to be made for such variation. This instrument has the advantage of using easily prepared test samples. Fig. 9. Hysteresis Meter, Used by General Electric Co. Designed and Described by Frank Holden. During the last few weeks of the year 1892 there was built at the works of the General Electric Company, in Lynn, Mass., under the writer's direction, an instrument, shown in Fig. 10, by which the losses in sheet iron were determined by measuring the torque produced on the iron, which was punched in rings, when placed between the poles of a rotating electro-mag- net. The rings were held by a fibre frame so as to be concentric with a 7(5 MAGNETIC PROPERTIES OF IRON. vertical shaft which worked freely on a pivot bearing at its lower end They had a width of 1 centimeter, an outside diameter of 8.9 centimeters, and enough were used to make a cylinder about 1.8 centimeters high. The top part of this in- strument, which rested on a thin brass cylin- der surrounding the rings, was movable. On the upper surface was marked a degree scale, over which passed a pointer, with which the upper end of a helical spring rotated. It was so constructed that when the vertical shaft with the rings and the upper part of the instru- ment with the spring was put in place, the lower end of the spring engaged with the shaft, and consequently rotated with the rings. A pointer moving with the lower end of the spring reached to the zero of the degree scale when the apparatus was ready for use. By this ar- rangement it was found what distortion it was necessary to give the spring in order to bal- ance the effect of the rotating magnet, and the spring having been calibrated, the ergs spent Fig. 10. Hysteresis Meter, on the rings per cycle were determined by mul- tiplying the degrees distortion by a constant. A coil, so arranged that it surrounded but did not touch the rings, made contact at its ends with two fixed brushes that rested in diametrically oppo- site positions on a two-part commutator, which revolved with a magnet. The segments were connected each to a collector ring against which rubbed a brush, the latter two brushes being joined through a sensitive Weston voltmeter. If this were so arranged that the coil was at right angles to the 1000 .\-(,o 3000 4!A\! 5000 t ' X) 7000 . CYCL SUO0 - c " ex ) - r E :WJ 7000 0000 / 2000 - 3000 1000 Fig. 11. induction, when the brushes changed contact from one segment to the other, it is evident, the self-induction of the circuit being negligible, that the mean value of the current in the circuit was proportional to the total flux through the coil. Knowing the constant of the voltmeter, the deflection Avas easily calculated from the speed of the magnet, the number of turns in the coil, cross-section of the rings, and the resistance of the circuit. From an induction of 2,000 gausses to at least 10,000 gausses, the leakage across the interior space of the rings was negligible. Carried on the shaft below the magnet was a pulley around wThich passed a flat belt driven with a pulley of the same size on an electric motor, so that the speed of the magnet could be found by observing that of the motor. In operating, the deflections to be produced on the voltmeter at a certain speed, with the desired induction in the rings, were first calculated. Five hundred HYSTERESIS METER. revolutions per minute was generally adopted as the speed in this case. The motor being run at the desired speed, the magnetizing current was ad- justed until the calculated deflection was produced on the voltmeter. Keep- ing the magnetizing current constant, the speed was changed successively in value to certain values, and the corresponding distortions of the spring necessary to balance the effect of the magnet noted. When this process was carried out at different induction values, and the ergs expended per cycle on the rings plotted as a function of the speed, a series of lines was produced, as shown in Figs. 11 and 12. It was found that the slope of the lines decreased very rapidly with the decrease in thickness of the iron sheet used so as to indicate that had it been thin enough the slope would have been zero between 100 and 800 revolutions per minute, which was about the highest speed permissible. From this it would seem that, in these tests, the total loss per cycle had two components ; one remaining constant, due to hysteresis, and the other varying as the speed of the magnets, due to cur- rents induced in the iron. Fig. 15 gives observations of eddy current loss and thickness of iron sheet on this assumption. The line drawn is a parabola, so that it would appear that with the range of observations made the loss varied about as the square of the thickness of the sheets. 1000 2000 3000' 4000 5000 6000 7000 7000 = 6000- 6000- 4000; 3000^ r -; i 1 : - C 3 , / 400 500 GOO 700 REVOLUTIONS PER MINUTE Fm. 12. Fig. 11 gives lines from iron .04 centimeters thick. Speed readings were not taken lower than 250 revolutions per minute, as it had been found that the lines were always straight, and speeds below this value could not be read with the tachometer available for this particular test. Plotting the hysteresis as a function of the induction, in this case the points are all quite close to a curve whose equation is, Ergs = A constant x (Density per square centimeter)1-47, three points in the latter calculated curve being shown by the crosses. The iron, a test on which is shown in Fig. 12, was .1 centimeter thick, and shows a greater eddy current loss. The equation for the hystere- sis curve for this sample is, Ergs = A constant x (Density per square centi- meters)1-4, some points in the latter curve being shown by crosses, as before. The eddy current losses for these two samples are plotted as functions of the induction in Fig. 14. The curves drawn are parabolas; showing that in these cases the eddy current loss varied approximately as the square of the induction, although there were often greater variations from that law than these two samples show. The average exponent for the hysteresis curves was a little over 1.5, although it varied from 1.4 to 1.7. Rings tested in this manner were wound and tested with a ballistic galvanometer, using the step-by-step method. There were discrepancies of as much as 4 per cent be- tween the two results, but an average of ten tests showed the ballistic gal- vanometer method gave results 2.5 per cent lower than the other. This difference is easily attributable to experimental errors. It being noticed that for a given induction in the rings, the magnetizing currents for different samples did not vary much, it was planned shortly 78 MAGNETIC PROPERTIES OF IRON. after completing the above apparatus to construct a modified instrument which would use electro-magnets of such high reluctance that the variations of the rings would be negligible, and induction be dependent only on the current. By making the electro-magnets of suitable iron and of about one-third the cross-section of the rings used, the iron may he so highly saturated that the induction will remain quite constant Fig. 13. Modified Hyster- under considerable variation in the magnet- esis Meter. izing current, thus rendering unnecessary any accurate comparisons of magnetizing currents, and the rings can be at about their maximum permeability when thus magnetized. Such an instrument is shown in Fig. 13 in its original ex- perimental form, with the rings in position ready for test. The rings are here allowed to rotate in opposition to the action of a spring and carry a pointer over a scale, so that is is quite direct reading. Twenty-five compar- 90C0 -r-r „„„ 7000 -gAus ? " ^ *Z S *'_ * ' S 30C0-^? / 0 21 io 4 .9 - H r,Kr> F -,fi /■ /■ - * / / / / •■2 give any att were attem The instr pletion at tl ) 100 200 300 400 000 COO 700 800 900 100011001200130014 ergs per cm3 per cycle Fig. 15. ention to the matter, no further investigat pted. iment first described has been in use contii ae works of the General Electric Company, D01S00J onsin mousl inScl HI) tl ys iei EDDY CURRENT FACTORS. 79 E»I>Y CURRENT FACTORS FOR DIFFERED! CORE DENSITIES AUD EOR VARIOUS 1AHIj¥ATI01S. (Wiener.) tH O W Watts dissipated Watts dissipated PER CUBIC FOOT OF PER CUBIC FOOT OF g©H IRON AT A FRE- H ° H IRON AT A FRE- 1 § H ° QUENCY OF 1 CYCLE H ? S S O QUENCY OF 1 CYCLE PEE SECOND. PER SECOND. 0gfe -, w fe w 3 H & Thickness of lamination,8 Thickness of lamination, 8 '%%%£ .010" .020" .040" .080" < «M 02 .010" .020" .040" .080" 10,000 .0007 .003 .012 .046 66,000 .0315 .126 .503 2.013 15,000 .0016 .007 .026 .104 67,000 .0325 .130 .519 2.075 20,000 .0029 .012 .046 .185 68,000 .0335 .134 .534 2.137 25,000 .0045 .018 .072 .288 69,000 .0345 .138 .550 2.200 30,000 .0065 .026 .104 .416 70,000 .0355 .142 .566 2.265 31,000 .0070 .028 .111 .444 71,000 .0365 .146 .582 2.330 32,000 .0074 .030 .118 .472 72,000 .0375 .150 .599 2.396 33,000 .0079 .032 .126 .503 73,000 .9385 .154 .616 2.463 34,000 .0084 .034 .134 .534 74,000 .0396 .158 .633 2.530 35,000 .0089 .036 .142 .567 75,000 .0407 .163 .650 2.600 36,000 .0094 .038 .150 .600 76,000 .0418 .167 .668 2.670 37,000 .0099 .040 .158 .633 77,000 .0429 .171 .685 2.740 38,000 .0104 .042 .167 .667 78,000 .0440 .176 .703 2.810 39,000 .0110 .044 .176 .703 79,000 .0451 .180 .721 2.883 40,000 .0116 .046 .185 .740 80,000 .0462 .185 .740 2.958 41,000 .0122 .049 .194 .777 81,000 .0474 .190 .758 3.033 42,000 .0128 .051 .204 .815 82,000 .0486 .194 .777 3.108 43,000 .0134 .954 .214 .855 83,000 .0498 .199 .796 3.184 44,000 .0140 .056 .224 .896 84,000 .0510 .204 .815 3.260 45,000 .0146 .059 .234 .937 85,000 .0523 .209 .835 3.340 46,000 .0153 .061 .245 .979 86,000 .0535 .214 .855 3.420 47,000 .0160 .064 .256 1.022 87,000 .0548 .219 .875 3.500 48,000 .0167 .067 .267 1.066 88,000 .0560 .224 .895 3.580 49,000 .0174 .070 .278 1.110 89,000 .0573 .229 .916 3.662 50,000 .0181 .072 .289 1.055 90,000 .0586 .234 .937 3.745 51,000 .0188 .075 .300 1.200 91,000 .0599 .240 .958 3.830 52,000 .0195 .078 .312 1.248 92,000 .0612 .245 .979 3.915 53,000 .0202 .081 .324 1.297 93,000 .0625 .250 1.000 4.000 54,000 .0210 .084 .337 1.346 94,000 .0638 .255 1.021 4.085 55,000 .0218 .087 .349 1.397 95,000 .0651 .261 1.043 4.170 56,000 .0226 .091 .362 1.448 96,000 .0665 .266 1.064 4.257 57,000 .0234 .094 .375 1.500 97,000 .0679 272 1.086 4.345 58,000 .0242 .097 .389 1.555 98,000 .0693 .277 1.109 4.436 59,000 .0251 .101 .403 1.610 99,000 .0707 .283 1.132 4.528 60,000 .0260 .104 .416 1.665 100,000 .0722 .289 1.156 4.622 61,000 .0269 .108 .430 1.720 105,000 .0797 .319 1.274 5.095 62,000 .0278 .111 .444 1.776 110,000 .0875 .350 1.398 5.593 63,000 .0287 .115 .458 1.833 115,000 .0955 .382 1.528 6.113 64,000 .0296 .118 .473 1.891 120,000 .1040 .416 1.664 6.655 65,000 .0305 .122 .486 1.951 125,000 .1128 .451 1.806 7.222 80 MAGNETIC PROPERTIES OF IRON. SPECIFIC EXEHGY DliilPAXIOS IHT AMUIATUIIE CORE. (Wein er.) Hysteresis LOSS FOR Eddy-current loss for SHEET IRON AT PRE- 030" (.075 CM.) LAMINATION, Magnetic QUENCY OF ONE MAG- AT ONE CYCLE FER SECOND density. NETIC CYCLE FER FROFORTIONAL TO FRF- SECOND (IN WATTS). QUENCY (IN WATTS). Lilies Gaus- of force Per Per Per Per Per Per Per Per ses. per sq. in. cm.3 c. ft. kg. lb. cm.3 c. ft. kg- lb. 2,000 12,900 .00007 1.98 .0091 .0041 .0000004 .011 .000051 .000023 3,000 19,350 .00013 3.68 .0140 .0077 .0000009 .026 .000119 .000054 4,000 25,800 .00020 5.75 .0265 .0120 .0000016 .046 .000212 .000096 5,000 32,250 .00029 8.20 .0378 .0171 .0000025 .071 .000327 .000148 6,000 3S,700 .00039 11.03 .0508 .0230 .0000036 .102 .000471 .000213 7,000 45,150 .00050 14.15 .0652 .0295 .0000049 .139 .000640 .000290 8,000 51,600 .00062 17.5 .0806 .0365 .0000064 .181 .000833 .000377 9,000 58,050 .00074 20.9 .0963 .0436 .0000081 .229 .001054 .000478 10,000 64,500 .00087 24.6 .1133 .0513 .0000100 .283 .001303 .000590 11,000 70,950 .00102 28.3 .1303 .0590 .0000121 .343 .001580 .000715 12,000 77,400 .00118 33.1 .1524 .0690 .0000144 .408 .001878 .000850 13.000 83.850 .00134 37.9 .1745 .0790 .0000169 .479 .002204 .000998 14,000 90,300 .00150 42.7 .1966 .0890 .0000196 .555 .002553 .001157 15,000 96,750 .00168 47.5 .2193 .0990 .0000225 .637 .002923 .001328 16,000 103,200 .00187 52.9 .2440 .1103 .0000256 .725 .no;:;; Mi .001512 17,000 109,650 .00206 58.3 .2680 .1212 .0000289 .818 .003770 .001708 18,000 116,100 .00225 63.7 .2932 .1328 .0000324 .917 .004220 .001911 19,000 122,550 .00246 69.6 .3200 .1450 .0000361 1.022 .004710 .002130 20,000 129,000 .00267 75.6 .3480 .1575 .0000400 1.133 .005225 .002362 ELECTRO-MAGNETS, PROP£RTIE§ OF. Residual Magnetism is the magnetization remaining in a piece of magnetic material after the magnetizing force is discontinued. Retentiveness is the measure of the magnitude of residual magnetism. Coercive Force is the force which holds the residual magnetism, and is measured by the strength of the reverse field required to remove all mag- netism. Permanent magnetism is residual magnetism of great coercive force, as in hard steel, which has little retentiveness ; Avhile soft iron has great reten- tiveness but little coercive force. The following paragraphs are condensed from S. P. Thompson's " The Electromagnet." Magneto-Motive Force. — The magneto-motive force, or magnetiz- ing power of an electro-magnet is proportional to the number of turns of wire and the amperes of current flowing through them ; that is, one ampere flowing through ten coils or turns will produce the same magneto-motive force as ten amperes flowing through one coil or turn. If n = number of turns in the coil, I=z amperes of current flowing, 1.257 = -^ (to reduce to C. G. S. units). Magneto-motive force = 1.257 x nl=z $ . Intensity of Magnetic Force. — Intensity of magnetic force in an electro-magnet varies in different parts of the magnet, being strongest in the middle of the coil, and weaker toward the ends. In a long electro-mag- net, say a length 100 times the diameter, the intensity of magnetic force will be found nearly uniform along the axis, falling off rapidly close to the ends. In a long magnet, such as described above, and in an annular ring wound evenly over its full length, the value of the magnetic force, J£, is deter- mined by the following expression : — 3C — 1.257 —=- , in which 1 = centimeters. If the length is given in inches, then 3C= -495-^— , in which lu= inches. If intensity of the magnetic force is to be expressed in lines per sq. inch, 3C//= 3-193 X^. Value of £fC at the centre of a Single-turn of Conductor. — In a single ring or turn of wire of radius r, carrying / amperes of current 3C= *,: X |= -6284 X '- ■ Force on Conductor (carrying* current) in a Magnetic Field. — A conductor carrying current in a magnetic field is repelled from the field Dy a certain mechanical force acting at right angles both to the conductor itself and to the lines of force in the field ; see cut. The magnitude of this repelling force is deter- mined as follows, assuming the held to be uniform. 3C = magneto-motive force, or intensity of the held. I = length of conductor across the field in cm. l/y = ditto in inches. / = amperes of current flowing in the conductor. F = repelling force. no I j ■ vyi'J-L .Fin grains F in dynes = ~f^- 10 F in dynes - 25.4 FlG. 1. Action of Mag- netic Field, on Con- ductor carrying cur- rent. 161 ' 81 32 ELECTRO-MAGNETS. Work done by Conductor (carrying- Current) in moving: across a Magnetic Field. If the conductor described in the preceding paragraph he moved across the field of force, the work done will be determined as follows : in addition to the symbols there used, let b = breadth of field in and acrosB which the conductor is moved ; w = work done in ergs. bl = area of field, N=bl x = number of lines of force cut, ni Rotation ©f Conductor (carrying- current) around a Magnet role. If a conductor (carrying current) be so arranged that it can rotate about the pole of a magnet, the force producing the rotation, called torque, will be determined as follows : The whole number of lines of force radiating from the pole will be 4tt times the pole strength m. 4n- ml „ „„_ T 10 = — -yr— = 1.257 ml. Dividing by the angle 2 tt, the torque, T, is Every magnetic circuit tends to place itself so as to embrace the maximum flux. Tioo electric conductors carrying currents tend to place themselves in position such that their mutual flux may be maximum ; otherwise stated : if two cur- rents run parallel and in the same direction, each produces a field of its own, and each conductor tends to move across the other's field. In two coils or conductors lying parallel to each other, as in a tangent gal- vanometer, the mutual force varies directly in proportion to the product of their respective ni, and inversely as the axial distance they are apart. Principle of tlie Mag-netic Circuit. — The resistance that a mag- netic circuit offers to the passage or flow of magnetic lines of force or flux, has been given the name of reluctance, symbol (ft., and is analogous to resist- ance, to the flow of electric current in a conductor. The magnetic flux or lines of force are treated as current flowing in the magnetic circuit, and denoted by the symbol 0. The above two factors, together with the magneto-motive force described in the early part of this chapter, bear much the same relation to each other as do resistance, current, and E.M.F. of electric circuits, and are expressed as follows : — .. „ -Magneto-motive force Magnetic flux = 5 — _ — , reluctance 9 (ft 10 Av.' 1.257 ni _^ An ■■*£. 1.257 EXCITING POWER AND TRACTION. 83 If dimensions are in inches, and A is in square inches, then nl=-^- X 3132. and = ($/' A". The law of Traction. — The formula for the pull or lifting-power of an electro-magnet is as follows : — Pull (in dynes) = ^- . Pull (in grammes) = Pull (in pounds) = In inch measure, Pull (in pounds) = &2A 11,183,000 " 72,134,000 ' Magnetization and Traction of Electro Mag-net*. (B (B" Dynes Grammes Kilogs Pounds Lines per Lines per per per per per sq. cm. sq- inch. sq. cm. sq. cm. sq. cm. sq. inch. 1,000 6,450 39,790 40.56 .0456 .577 2,000 12,900 159,200 162.3 .1623 2.308 3,000 19,350 358,100 365.1 .3651 5.190 4,000 25,800 636,600 648.9 .6489 9.228 5,000 32,250 994,700 1,014 1.014 14.39 6,000 38,700 1,432,000 1,460 1.460 20.75 7,000 45,150 1,950,000 1,987 1.987 28.26 8,000 51,600 2,547,000 2,596 2.596 36.95 9,000 58,050 3,223,000 3,286 3.286 46.72 10,000 64,500 3,979,000 4,056 4.056 57.68 11,000 70,950 4,815,000 4.907 4.907 69.77 12,000 77,400 5,730,000 5,841 5.841 83.07 13,000 83,850 6,725,000 6,855 6.855 97.47 14.000 90,300 7,800,000 7,550 7.550 113.1 15,000 96,750 8,953,000 9,124 9.124 129.7 16,000 103,200 10,170,000 10,390 10.390 147.7 17,000 109,650 11,500,000 11,720 11.720 166.6 18,000 116,100 12,890,000 13,140 13.140 186.8 19,000 122,550 14,360,000 14,630 14.630 208.1 20,000 129,000 16,920,000 16,230 16.230 230.8 Exciting- Power and Traction. — If we can assume that there is no magnetic leakage, the exciting power may he calculated from the follow- ing expression ; all dimensions being in inches, and the pull in pounds. w/=:2661 X — X If dimensions are in metric measure, w/=3951 - V_Pullinkilos , \/ Pull in lbs. > 1 : : Area in sq. ms. Area in sq. cms. £ = 4965 "\yPull in kilos Area sq. cm. 84 ELECTRO-MAGNETS. Winding- of Mag-in't Coils. The following nomenclature is employed : — D = diameter of insulated wire in mils. d = diameter of bare wire in mils. t = thickness of insulation on wire in inches ( i.e., - — - — j • L ■=. total length of wire in coil in feet, a, b, h, and Z = coil dimensions in inches. K= ratio of diameter of insulated wire to bare wire. V= volume of winding space in cubic inches. N= total number of convolutions on spool. Tzr number of layers of wire on spool. n = number of convolutions per linear inch. p = resistance in international ohms of mil-foot of pure copper wire. (10.35 ohms at 20° C.) li — total resistance of coil in ohms. r = resistance per foot of wire in ohms. f =.-■=. feet in one ohm. r lm = mean length of convolution in inches. The winding will vary between two extremes, one the " square" winding in which it is assumed that the convolutions lie to- gether as if the wire was of square cross-section, and the other the " conical " winding in which it is as- sumed that the wires lie together as if the wire was of hexagonal cross-section. On the assumption that the same volume is occu- pied by insulating mate- rial about 15 percent more copper volume is obtained Tby the " conical " method of winding. The squar - 1- * f t b 1 a I 1 t h 1 assumed in the following, unless otherwise specified. The diameter of wire necessary to fill a given coil space with ber of convolutions is /lOOOUOO / li /500UU0 / (a — h) D=V-n*— = v n ' winding is i given num- y* 50U(n i, , / (a— b) K*N K*N The total length of wire of given diameter which can be wound in a given coil space is L 65450 I (a- — 1 2) From the above formula the dimensions of a spool to hold a specified length of wire of given diameter may be determined. If a and b are known l- If b and I are known If a and / are known 0545 Mrt3 — b-) — JD'iL + G545° lhZ a— V 65450 I _ k /65450 lb* — DH b— V 05450/ WINDING OF MAGNET COILS. 85 The l-esistance of a coil expressed as a function of the volume is p _ 862500 V Ix - DM* ' If the volume of wire is increased ten per cent to allow for the layers fit- ting into one another, _ W8700 V K- Did* ' Hence the diameter of i resistance is : necessary to fill a given volume with a given 948700 V The last three formulae are general, whatever the shape of the spool, i.e., whether the core is of circular, square, rectangular, elliptical, etc., cross- section. The next smaller gauge number than the diameter corresponding to the formula should be used in order to allow for irregularities in winding and for insulation between the layers. If R is taken at other than 68° F (20° C), a new value of R, i.e., R', must be taken, where R' = R (1 -f 0.0022 0/), where 6/ is the rise in temperature above 68° F. A formula known as Brough's formula is often applicable to the calcula- tions of the diameter of wire necessary to give a stater resistance. For circular cores, d = F y/677400(^- -fr, / + fiJ _tm Hi For square cores, Fig. 3, d _ |~ ^862500 (a2 - ft2) / + ^1* _ For rectangular cores, Fig. 4, d = |~ /431250 (A - a) (A + B + cTTt>) + ^"1* For core made up of square and two semi-circles, Fig. 5, radius of core-circle, b. radius of outer-circle, b. 86 ELECTRO -MAGNETS. d _ I 4/862500 (B — b) [n (B -f 6) + 2 a] + t* Thickness of Wire B nsulat ion. — The thickness of insulation upon wire varies with the manufacturer, and no fixed value can be given to cover all cases. The following table represents the practice of several large man- ufacturers. To determine the diameter of insulated wire, add to the dia- meter of the bare wire. FOR COTTON. FOE SILK. B & S Gauge Single Double Single Double OtolO 10 to 18 18 up 7 mils 5 mils 4 mils 14 mils 10 mils 8 mils 2 mils 4 mils The above values correspond to It in the formulas. Relation of Ampere-turns to Dimensions of Coil. For a coil of static dimensions it can be shown that NI— 1AG lm (1. + 0.0022 0/) ' where E = difference of potential across terminals of coil. The ampere-turns are independent of the length of the coil, of the thick- ness of insulation, and of the method of winding, depending upon the diameter of the wire, the mean length of a turn, and the temperature of the coil. To keep the number of ampere-turns constant in a coil of given volume, d* of the wire must vary inversely as E. Relations Holding- between Constants of Coils. In the following it is assumed that the thickness of insulation is propor- tional to the diameter of wire, and that all coils are uniformly wound. The results obtained under this consideration are practically but not strictly correct. The weight of copper required to fill a given coil volume is constant, whatever the size of the wire used. The resistance in a given volume varies inversely as the fourth power of the diameter of the wire used. The resistance in a given volume varies inversely as the square of the cross-sectional area of the wire used. The number of convolutions in a fixed volume varies inversely as the square of the diameter, or inversely as the cross-sectional area of the wire used. The resistance of a coil of given volume varies directly as the square of the number of turns. The magnetic effect produced by an electro-magnet of given shape, size, and construction is proportional 'to the product of the current into the square root of the resistance of the coil. If two coils of same dimensions are wound with different size wire, the current must vary with the cross-sectional area of the wire, in order tc obtain the same heating effect, or same temperature rise. For same energy loss E2 must vary inversely as (area)2 of Avire, or foi same heating effect the voltage across terminals of coil must vary inversely as the cross-sectional area of the wire used. . AMPERAGE AND DEPTH OF WINDING FOR MAGNETS. 87 AETER]¥ATJL]¥G-C1JRREj¥T ELECTHO-MAGSETi. The cores of electro-magnets to be used with alternating currents must be laminated, and the laminations must run at right angles to the direction in which eddy currents would be set up. Eddy currents tend to circulate par- allel to the coils of the wire, and the laminations must therefore be longitu- dinal to or parallel with the axis of the cores. The coils of an alternating-current electro-magnet offer more resistance to the passage of the alternating current than the mere resistance of the con- ductor in ohms. This extra resistance is called inductance, and this com- bined with the resistance of the conductor in ohms produces the quality called impedance. (See Index for Impedance, etc.) If L = coefficient of self-induction, Ar= periods per second, E = resistance, and, Impedance = V^+^Wi Maximum E.M.F. Maximum current = Mean current = Impedance Mean E.M.E. Impedance. If the current lags behind the E.M.F. by the angle , then _ . Mean E.M.F. Mean current r= -=- — —, X cos = ohms per mil-foot ; M= circular mils ; T= turns per linear.inch ; n = number of layers in depth. The emissivity is taken at .4 watt per sq. in. for stationary magnets for a rise of temperature of 35° C. (63° F.). For armatures, according to Esson's experiments, it is approximately correct to say that .9 watt per sq. in. will be dissipated for a rise of 35° 0. The insulation allowed is .007 inch on No. 0 to No. 11 B. and S. ; .005 inch on No. 12 to 24 ; and .0045 inch on No. 25 to No. 31 single ; twice these values for insulation of double-covered wires. Fifteen per cent is allowed for imbedding of the wires. The standard of resistance employed is 9.612 ohms per mil-foot at 0°. The running temperature of tables is taken at 25° + 35° = 60° C. The column giving the depth for one layer is the diameter over insulation. ELFX'TRO-MAGNETS. I OOOONCOOOMINffliiONt-iOOJM'^OlNCOnHrf* "t[^d8(I OC^r^<»t>iqCCCOC5Crjt-;<»-<*COC<]r- t- I- cd 10 ■* CO •duiy ? t- CS GO CO 00 OS - d d d d ■* •* •q^daa ocdcn ^ ooascoc co cd 06 cs'ci i-H ic c: co ob" 00 «5 ^J OS CO b; CO t-; CO iH iq C« CO •«# «# OS « t; OS t> i-J t> di>dddd'<* I go >-i o c- o c r- :■■> ~ » - ■ -■ -t< o ci co ■"*< go co lo go co q^dad I ocoooioio^(NNqo>ooi-;tB(OTi|i|nNDjiHqo S3 C3 (M IM (N n cj o4 c-j c 'loit-co-torii-ci-j q ■>* cq cq oi oj nqqa •duiy ^i>-iqqcqqocqoqiqt-;cqq5jq^ StOcdcONHOOOOlOlOMrirtHOO C CI CI d 01 CI — HrtHHHHHrl TL'-HOOJ'^COtMOOt-CSwi •q^daQ qcqaqi>.qqiq'#cqcqo}CN;'-Hqqasqoo •duty CNoqqT-fOST^qoqt^cqcqqqiqcAc: Tj< as as d ■* co i-<" d t^ t-' io <*" CO CN t-h c ,_ CCClClClClClClClrt HHHrtHHr I mcir-HCHcc ci HOooMHino *md9(X iq-*-* cqtNCNrHi-^qq qqoocq t-c- w^rn«oqqiNq^q«D(Nq'*05!NqNiooTi*H CO CO CO CO C ' I fiUin^-n-li-l^^rHOSCOt^t^ •Tlldaa I q LO CO CO OS CD c MrttOlOrHOl qqqoqoooqt^t^qqqqiqiqiq-* TfTf -i •duiy •fNccrtcirtqqqioqrt'fNoqo GOOJi-Ht^inCOoiaJ^-^r-irHCOt^lCT) Tf TJH -^ CO CO CO Ol Qj f)CK^(NHHHr •q^dea | ©cocot>yociO'*cocN'-Hd •mdaa ., ■dmy toooo-f woati SSSSSSS* q t> iq i>; q q •<# ho* io(et>HOai>NaiOHri (M«(N«(S«N^ co co eo ■* ■* ■* t)< ■*' iri loujujocc ^ool--* ci : -- 2 co co ci q -h co oi »L-t>c6odosos© aad -sq^i s § '0 S99.I§8p 09 ye c^ooj aad siuqo nwanoiN OOcNCOOSOSCSCMincOCOOSiH oooooooc ooooooo caoHcor) iiiiii 'SITM -nqno.no 5 co CD 01 ! 388335 ©■S83S •seqoui 'ajBq -wbict »£co3 *co~ S^c*? to CO t~O0 OS O »-i CflCDTtT)- -* CN C: i ^i* co^ cn cn cn cn i-5 i-i r-5 r-5 °9 ^ ^ ^ ^ **. °" •q^a CNOSOSC^Tf-HasoOCOlC CO J] C) M W OJ H H H T- « H Tj( W |> O Ci L*. ' +COHOOO ■^ n* co co co cn (m' th* i-h* i-h i-i* °? °9 v. •"■ 3*8888 -q^dsa O00CN*HlOCS-t<©COC0©t~''#TtiCN©0! - o co o io ^ ^ ■# o: co co ci w ci c^ h r •dray OOTfNOOOOHCOCDrlicOLOC •q;d9a ciHioooj^oocicci cq co jc iq -* n; *# co co cn cn o a ■* CO CM iH © •dray -i as © co to i-i t> •q;daci _*d id n" co" co'_co_cn cn r-5 i-5 i-5 i-5 NCOHCOt>CO o"t~ — |i"c O IO CO -h CO ii 71 -l-X-*rl»t>-*rH ■ '.; - CO CO 1_Q* Tt* jj CO* CO CN O* i-H* H r-H* -i ^ g CO CO .O Tf< T^ CO CO •q^dso; OCOCICNCNCNOCOO* CS OS 00 r- •dray oocoociHosaHHot-ii -'-hnSS CO ■# CO -* CO <* !>; CI t- CN OS CO CO r-H r-t © 00 t> CC i l>'l>'cOlOlQTi"co"co'c-*C>*i-'i-'T-"i-*iMi-? •q^daa •ipdaa •dray •q^d9Q[ •dray .tod sxi.inx aad -sq-i P-icooc-cnloo OH'jHoicoaij(iaNO)nt'«nMNooioc"oi> J5 c9 £? Sg - .' - -foit-i>ccoco*if coiqoicqiccocoioooc-ic-coGicoi-ioiaii-Hcj: d o co t>" t- co" io rf co" CO wqqN^Hqoqq^qoqqcjqoqqoT- t-H i-l CN -# -* CO l>" OS CN *#" t-* 05 CO CO i>" OS " " OS CO CO CO CO S3 5J8§§1§§§8 §§§1111 35 g a a £r N O lfi 00 -tl Cl 00 Cl >-< -f O M Q M lA o • • - IO00t0§*§CCCTNg*0O ■* •# CO CN CN i— IHH lOCNO og£gg£S^coJco£c*:c^c3^K£3S3S •sSive*) ^ CO TMOCOt>OOfflOnC 3 90 ELECTRO-MAGNETS. .lad sujiij, CI L~ l-;'l|Clr>KK>f'*HO X L- L.0 O CO 01 in" •#-#-#-* •*'fococofococ^o5c5c d to 1Q MHOHiot-«««H»ionoHOM!o tc-t^oinqqa t~ t- lo ic -r -f co ci h i-h CI CI CN CI CI CI r-i r-5 rH r-i H r-i r-i r-i r-i r-i r-i r-i •f?HOMOgO^OHTtHN(MiOOiNt- " :i- - i r. -i- X l-XNCCOH ci -t O t~ M C-j « H q o o c; X X X L - t~ t - '-O CO CO lO i^-Ht^qqqqi^t^cot^THoqr-iiqcqcoo eocor>corHO?o'oeic4c5CJ«dcO"*eoC'i'-H 00COr- cocococo^'^''*'jt> o ci § cs ^ o »o ci cs t~ co lo r- i-j © © © © _©io-t# \a co i> oo © o i-i IMJO^IO 1 2 3 4 4.5 4.87 22.1 1.84 .0004576 .24 7. 4 .75 5 5.09 5.82 29.6 2.46 .0007738 .24 9. 5 .74 6 5.66 6.41 36.3 3.02 .0011963 .24 11.5 6 .74 7 6.2 7.3 45.3 3.77 .001780 .24 14. 7 .73 8 7.05 8. 56.5 4.7 .0029654 .24 17.5 8 .73 9 7.66 8.42 64.5 5.37 .0042574 .24 22. 9 .73 10 8.54 9.6 82. 6.83 .00683 .238 27. 10 .72 11 * 9.7 11. 116.7 9.72 .012254 .236 34. 11 .72 12 11.2 12.8 143.4 11.95 .0150654 .233 42. 12 .71 13 * 12. 14. 168. 14. .03627 .23 55. 13 .71 14 13. 15.4 200. 16.66 .0431627 .227 68. 14 .70 15 15.37 17.9 275.5 22.96 .071520 .224 87. 15 .68 16 16.74 19.4 324.7 27.06 .108757 .22 110. 16 .64 17 17.74 21.33 378.4 31.53 .15980 .217 140. 17 .62 18 * 19.5 23. 448.5 37.38 .2389 .19 175. 18 .61 19 22.77 24.9 567. 47.25 .39165 .185 220. 19 .60 20 25.7 29.7 763.3 63.60 .6464 .184 280. 20 .58 21 28.3 32.5 920. 76.6 .98163 .182 360. 21 .57 22 31. 36. 1116. 93. 1.502 .18 450. 22 .55 23 34.4 40.36 1390.3 115.86 2.36 .178 560. 23 .52 24 36.9 44.6 1649. 137.4 3.53 .168 715. 24 .45 25 38. 47. 1790. 149.2 4.734 .145 910. 25 .43 26 * 42. 50.5 2100. 170. 7. .14 1165. 26 .41 27 * 48. 55.5 2600. 210. 10.5 .135 1445. 27 .40 28 53.28 61.1 3256. 271.3 17.63 .13 1810. 28 .39 29 * 59. 68. 4000. 335. 27. .125 2280. 29 30 63.26 76.8 4860. 405. 41.84 .121 2805. 30 .38 31 32 33 34 35 36 RELATION AND DIMENSIONS OP CON- DUCTORS FOR DISTRIBUTION. REJLATIOIY OF E.]?I.F. ; CFBREVT; DISTANCE, CROS§-§ECTIOX, AX1> WEICHT OP coaroucxoiis. a. Current or E.M.F. varies directly with the amount of energy trans- mitted. b. Given the work done, loss on the line, and the E.M.F. at the motor terminals and point of distribution ; then the cross-section of conductor varies directly with the distance and weight as the square of the distance. c. With the same conditions as above, the weight of conductor will vary inversely as the square of the E.M.F. at the motor terminals. d. With a given cross-section of conductor, the distance over which a given amount of power can be transmitted will vary as the square of the E.M.F. e. Given, the weight of conductor, the amount of power transmitted, and the loss in distribution ; then the distance over which the power can be transmitted will vary directly as the E.M.F. PRECI§IO]¥ OF CALCEIATIOiri OF WISTRIB1T IWCi: SYiTEMS. While it is possible and in every way the best to make complete compu- tations for the conductors for isolated plants and for plants of a permanent nature, it is practically impossible to make anything like precise computa- tions for large public systems of distributions, such as a large Edison system. In the early days of the Edison stations, exact sizes of conductors were computed for' entire systems ; but when the network system vras introduced, and it became possible to keep the E.M.F. constant all over a system by varying the number of feeders, all such exact computations were dropped ; and to-day such systems are equipped with a few standard sizes of conduc- tors, feeders being of one or two sizes only, and mains being of but two or three sizes, judgment of the management being used as to which size will best fit given conditions. ECOHOmiCAL COHTDITIO^i. In the laying out of a system of electrical distribution, there are eight points to bear in mind in order to obtain the best economy ; and they have been so well stated by Abbott, that I quote from his book the following : — " 1. The conductors must be so proportioned that the energy transmitted through them will not cause an undue rise of temperature. 2. The conductors must have such mechanical properties as to enable them to be successfully erected, and so durable as to require a minimum of annual maintenance. 3. The conductors may be so designed as to entail a minimum first cost in line construction. ECONOMICAL CONDITIONS. 93 4. The conductors may be designed to attain a minimum first cost for station construction. 5. The conductors may be so designed as to reduce first cost of plant, and cost of operation and maintenance to a minimum. 6. The conductors may be designed to secure minimum total first cost of installation. 7. The conductors may be so designed as to secure maximum conditions of good service. 8. The conductors may be so designed as to attain a maximum of income with a minimum of station first cost." 1. If cost of production of electric energy is low, and cost of conductors high, make conductors small in cross-section, and of such size that the in- terest on its cost plus the expense of maintaining it will be a minimum, and balance the cost of energy lost in heating. In no case, however, should the conductor be made of a size so small as to heat dangerously, for which see tables in " ^National Code." When the cost of electric energy is high, and that of the conductors low, then the cross-section of conductor must be larger, in order that the cost of energy lost may not be too high ; but the balance, with that of interest and maintenance, should still be maintained. 2. In all cases, conductors of sufficient size to have mechanical strength to suit the particular position they are to occupy, should be used. Due attention should be given to liability of siioav and sleet, breaking of poles, etc., if conductors are overhead. 3. When a plant is installed for a temporary purposerand the line sal- vage will be small, while no harm will be done to the generating plant, the cost of the line should be a minimum, and the conductors may well be of a size just sufficient to carry the current with safety, both as regards heating and mechanical strength. 4. The minimum first cost of station can be obtained, as far as influenced by the distribution system, by reducing the losses in the conductors to a minimum, thus calling for the smallest amount of current to do the work. 5. As a decrease in the expenditure for line and construction demands an increase in the cost of central station, and apparatus for producing the extra energy lost in the line, and increases the operating expense of the station likewise, it is evident there must be a point where the total of the interest and depreciation on the line can be made practically equal to the cost of the energy lost in the line ; and at this point the expenses will be the least. Care must be used in applying this law, which was first stated by Lord Kelvin in 1881, as follows : " The most economical area of conductor will be that for which the annual interest on capital outlay equals the annual eost of energy wasted." One side of this equation would be the interest, depreciation, repairs, and maintenance of the conductor, the other would be the cost of producing the energy at the generator terminals, in- cluding interest, depreciation, and operating expense. Kapp says that the above law only applies where the capital outlay is proportional to the weight of metal contained in the conductor, a condition seldom obtaining in practice, and states the correct rule as folloAvs : — " The most economical area of conductor is that for which the annual cost of energy wasted is equal to the annual interest on that portion of the cap- ital outlay which can be considered to be proportional to the weight of metal used." Prof. George Forbes, in his Cantor lectures in 1S85, called that portion of the cost of the distributing system which is proportional to the weight of metal used, " the cost of laying one additional ton of copper ; " and he shows that, for a given rate of interest charge (inclusive of depreciation), and a given cost of copper, " the most economical section of the conductor is indepen- dent of the E.M.F., and of the distance, and is proportional to the current." Professor Forbes at the same time published some tables to facilitate the calculations ; and Prof. H. S. Carhart has enlarged them, and reduced the values to United States money. 94 CONDUCTORS. 000S$ © © SI S CN © CO 8 CO © © © © I g § g 0081$ 8 CO cm 8 CO CO © CO © 13 CO 3 © g OS 0091 $ § OS CM 8 00 CM i I rf 13 CO 13 § 1 00tl$ | eo OS CM CM 3 1 CO CO CO 9 © 13 8 13 o 00SI$ o 3 | OS CO cm o ?5 CO 3S CO "# O 1 OOITt o 1 5 CO 00 05 o CM © CO eo OS CO O 8 0001 $ I o o o o o o 1 1 o s CO 8 5: 006$ o s 1 n a n h f B3S8£ 3 lH «D Tt< rt l> « r -t- r> c -x • 5 cr i-i oi ci t< !N O 00 CD •* gSoSggg - ",= i- :: ~) — ~ X — >~ -tiin — -x r~ i- -H t- cs u~ -t< -+i c: rt ?i - « c c r. i> t- - c; -t -i< n r: ci - - ' - v — -. i " -r -+ i- c i- k c. c i- *i k -t ooooooi-(T-iT-trHi-n-iT-ii-ii-n-iT-ic)ci-:i*i-:i i ,: - c-:ir.-rir: -i- x ~ c — m r-t -* in « - - - x — . — :n c: -t< i- c i- x r. c ti ^ -t ;-. •. 9AT'6SS ¥ffc'9IS *8Z'SSI «s -~u x c : 3 r: 2' 2 fe" ^SS CO -fi L". L- CC ~ ^i £1 ; o»atnooD« H (N -N w o ftrto ^° lO 9 5 Ift u 6 Sis o so Pressure required at Generator terminals. •rH qj cvg l§ a. 6. c. d. e. /• 9- h. m. X. lOO 125 150 175 .07460 .05968 .04973 .04262 574.58 459.68 383.06 328.28 $57,458 45.968 38.306 32.828 $2.8729 2.2984 1.9153 1.6414 .5745 .3831 .2739 .01626 .02032 .02439 .02845 $.1626 .2032 .2439 .2845 101.626 102.032 102.439 102.845 1.626 2.032 2.439 2.845 200 225 250 275 .03730 .03316 .02984 .02713 287.28 255.40 208.94 28.728 25.540 22.984 20.894 1.4364 1.2770 1.1492 1.0447 .2050 .1594 .1278 .1045 .03252 .03658 .04065 .04471 .3252 .3658 .4065 .4471 103.252 103.658 104.065 104.471 3.252 3.658 4.065 4.471 300 325 350 375 .02486 .02295 .02131 .01989 191.52 176.79 164.14 153.22 19.152 17.679 16.414 15.322 .9576 .8839 .8207 .7661 .0871 .0737 .0632 .0546 .04878 .05284 .05691 .06097 .4878 .5284 .5691 .6097 104.878 105.284 105.691 106.097 4.878 5.284 5.691 6.097 400 425 450 475 .01865 .01755 .01658 .01570 143.64 135.19 127.70 120.97 14.364 13.519 12.770 12.097 .6182 .6759 .6385 .6048 .0479 .0423 .0374 .0337 .06504 .06910 .07317 .07723 .6504 .6910 .7317 .7723 106.504 106.910 107.317 107.723 6.504 6.910 7.317 7.723 500 525 550 575 .01492 .01420 .01356 .01297 114.92 109.44 104.47 99.93 11.492 10.944 10.447 9.993 .5746 .5472 .5223 .4996 .0302 .0274 .0249 .0227 .08130 .08536 .08942 .09348 .8130 .8536 .8942 .9348 108.130 108.536 108.942 109.348 8.130 8.536 8.942 9.348 600 625 650 675 .01244 .01193 .01147 .01105 95.76 91.93 88.39 85.12 9.576 9.193 8.839 8.512 .4788 .4596 .4419 .4256 .0208 .0192 .0177 .0163 .09756 .10162 .10568 .10974 . 1.9756 | 1.0162 3 1.0568 I 1.0974 109.756 110.162 110.568 110.974 9.756 10.162 10.568 10.974 900 725 750 775 .01066 .01029 .00995 .00962 82.08 79.25 76.61 74.14 8.208 7.925 7.661 7.414 .4104 .3962 .3830 .3707 .0152 .0142 .0132 .0123 .11382 .11788 .12194 .12600 g 1.1382 53 1.1788 !H 1.2194 € 1.2600 111.382 111.788 112.194 112.600 11.382 11.788 12.194 12.600 soo 825 850 875 .00933 .00905 .00878 .00854 71.82 69.64 67.59 65.66 7.182 6.964 6.750 6.566 .3591 .3482 .3379 .3283 .0116 .0109 .0103 .0096 .13008 .13414 .13820 .14226 g 1.3008 •£ 1.3414 g 1.3820 O 1.4226 113.008 113.414 113.820 114.226 13.008 13.414 13.820 14.226 ooo 925 950 975 .00829 .00807 .00785 .00766 63.84 62.12 60.48 58.93 6.384 6.212 6.048 5.893 .3192 .3106 .3024 .2946 .0091 .0086 .0082 .0078 .14634 .15040 .15446 .15852 1.4634 1.5040 1.5446 1.5822 114.634 115.040 115.446 115.852 14.634 15.040 15.446 15.852 lOOO .00746 57.46 5.746 .2873 .0073 .16258 1.6258 116.258 16.258 Res. of 100 ft., 1 sq. in. at 80° C. = .010.0678. 98 CONDUCTORS. 6. When a plant is installed for more or less temporary work, it is, of course, policy to make the first cost a minimum ; and again, in many places, and perhaps in most places, it is impossible to predetermine the cost of power per unit, or number of hours it will be necessary to run, or the num- ber of hours of heavy and of light load, and many other items necessary to be known in order to determine and calculate the most economical form of plant to install. In such cases it is often necessary to feel one's way by installing a plant of low cost until the market is developed or its direction determined, after which it is much easier to lay out a plant that will produce the most econom- ical results. Sprague says that the least cost of plant is determined when the variation in the cost of the generator is equal to that in the cost of the line ; which is practically true, provided the cost of motors and generators per horse-power or unit capacity is the same. Sprague then develops the following law : — " With fixed conditions of cost and of efficiency of apparatus, the number of volts fall to get the minimum cost of plant, is a function of distance alone, and is independent of the E.M.F. used at the motor." " With any fixed couple and commercial efficiency, the cost of the wire bears a definite and fixed ratio to the cost of the generating plant." " The cost of the wire varies directly with the cost of the generating plant." " If we do not limit ourselves in the E.M.F. used, the cost per horse-power delivered exclusive of line erection is, for least cost and for a given commer- cial efficiency, absolutely independent of the distance." Without going into the detail, if we work out problems based on the above laws, the result shows that the law first stated by Professor Forbes, i.e., that " the most economical section of conductor is independent of the distance or E.M.F., and is proportional to the current," is correct. Badt develops the following law : — " For minimum initial cost of plant, and assuming certain prices per horse-power of motors and generators and power plant (all erected and ready for operation), and assuming a certain price per pound for copper (de- livered at the poles), the total cost of the plant, excluding line construction, is a constant for a certain efficiency of the electric system, no matter what the E.M.F. of the motor and the distance may be." " At a given efficiency of the electric system, the E.M.F. of the motor and distance will increase and decrease in the same ratio." 7. In designing for the accomplishment of the best service, series circuits can be economically laid out under some of the previous rules ; but in de- signing circuits for parallel distribution, they must be arranged for furnish- ing a constant and unvarying pressure at the lamps or motors of the customer, regardless of the cost of conductors ; and therefore service require- ments and not minimum first cost govern, as no service will be a paying investment that has not a uniform pressure and is not continuous in its character. Parallel distribution is fully treated in another chapter. 8. It is the attempt of all engineers to attain a maximum income from a minimum first cost of plant. If power is cheap and transportation costly, it is better to construct plant under Section 3. In some cases, though, so much of the station capacity might be wasted in the conductors as to leave little from which an income could be received ; but increasing the carrying capacity of the conductors somewhat, provided it did not cost too much to accommodate the extra machinery, would enable a paying income to be made. In order to determine the proper relation of line to station and plant, it is necessary to study the prospective loads. If street-lighting by series arcs is to be one of the sources of income, then a study of the hours of lighting must be made, and all the data as to number of hours burning, etc., will be found in the chapter on lighting schedules. For parallel and other methods of distribution, it will be necessary for some one acquainted with the system to make the necessary examination of the territory, and determine from its nature the probable load-curves. CALCULATION OF SIZE OF CONDUCTORS. 99 Efficiency in Electric Pow*ir transmission. From Badt's " Electric Transmission Hand-Book." 1. 2. 3. 4. 5. 6. £o 2s| II ^.SS Hz Iff w-Sl ? ° Mech. H be deliv at gene: pulle; N. I. 1.00 1.1111 0.0 1.1111 1.2346 81.00 1.00 1.1111 1.0 1.1223 1.2470 80.19 1.00 1.1111 2.0 1.1337 1.2597 79.38 1.00 1.1111 3.0 1.1454 1.2727 78.57 1.00 1.1111 4.0 1.1574 1.2860 77.76 1.00 1.1111 5.0 1.1696 1.2995 76.95 1.00 1.1111 6.0 1.1721 1.3134 76.14 1.00 1.1111 7.0 1.1947 1.3275 75.33 1.00 1.1111 8.0 1.2077 1.3419 74.52 1.00 1.1111 9.0 1.2210 1.3567 73.71 1.00 1.1111 10.0 1.2345 1.3717 72.90 1.00 1.1111 12.5 1.2698 1.4109 70.88 1.00 1.1111 15.0 1.3072 1.4524 68.85 1.00 1.1111 17.5 1.3468 1.4964 66.83 1.00 1.1111 20.0 1.3888 1.5447 64.80 1.00 1.1111 32.5 1.4336 1.5929 62.78 1.00 1.1111 25.0 1.4815 1.6461 60.75 1.00 1.1111 27.5 1.5325 1.7028 58.73 1.00 1.1111 30.0 1.5873 1.7636 56.70 1.00 1.1111 32.5 1.6464 1.8293 54.68 1.00 1.1111 35.0 1.7094 1.8993 52.65 1.00 1.1111 37.5 1.7778 1.9753 50.63 1.00 1.1111 38.3 1.8000 2.0000 50.00 1.00 1.1111 40.0 1.8518 2.0576 48.60 1.00 1.1111 42.5 1.9323 2.1470 46.58 1.00 1.1111 45.0 2.0210 2.2446 44.55 1.00 1.1111 47.5 2.1164 2.3515 42.53 1.00 1.1111 50.0 2.2222 2.4622 40.50 CAICUIATIOI OF THE SIZE OE COaTDUCTOKS FOR COITIHUOIJ8 CUit5*EI¥TS. Parallel distribution : — Resistance of one mil-foot pure copper at"0° C = 9.59 ohms ; Temp, coefficient for 70° F. = 1.084 Resistance of 1 mil-foot of pure copper at 70° F.= 10.395 ohms ; Resistance of 1 mil-foot of 96% conductivity copper wire at 70° F. = 10.81 ohms ; dia 2 — Length in feet x 10-81 RAsiat:5irn->A ^ ' 100 CONDUCTORS. Resistance of a copper wire conductor is then equal to Length in feet X 10.81 „ , diaT = ^.ohms. and the cross-section in circular mils or lgth in Resistance For lamps : — Let w — watts per candle-power ; then candle-power x w = watts per lamp, = W; and if E=i voltage, or P.D. of circuit ; W then — = 1= current in amperes per lamp. A voltage at which lamps are to he run is usually assumed, and a drop or loss of pressure of a certain percentage of this, determined on, and all wiring is calculated with those points as data. For instance, the most common voltage is 110 or thereabouts, and 5% drop, or 5.5 volts, is commonly assumed as the loss in pressure ; then the size of wire to produce this drop, with a given number of lamps, A7, taking, say, I amperes will be 10.81 x 2 distance X I ,. „ . n ... -. n~^5 ?-? = dia.2, or circular mils of copper. (3) volts drop 5.5 v ' For example : 120 lamps taking .5 amp. each are to be wired at a distance of 60 feet from the dynamo to the centre of distribution, at a drop of 3 volts. _ 10.81 x 2 x fiCK X 60 amps. „... . ., x. „ ^ ,0 Then, — — — 25944 cir. mils, or No. 6 B. and S. 3 volts. If the hot resistance of one lamp be given, and the number of lamps and distance, with the percentage of loss, then .. 10.81 x 2 distance x no. of lamps 100 — % loss cir. mils = =r — r-r tt^ ^ — X — tt^t1 (4) Resistance of one lamp % loss. Example : — Take the same case as above: 120 lamps ; distance 60 feet; drop in circuit, 3 % ; hot resistance of lamp, 200 ohms. 10.81 X 2 X 60' X 120 100 — 3 „„.. . .. Then, ^ X — „ — = 25944 cir. mils. For motors : — 1 Electric horse-power = 746 watts. Therefore, horse-power x 746 = watts. And watts -~ volts = amperes. Let E= volts at terminals of motor, v = volts lost in conductor. I4-» = E.M.F. at generator terminals. I=i current required at motor to deliver A7"mechanical h.p. at shaft of motor. D= single distance between motor and generator. AT:= number of mechanical h.p. delivered at motor shaft. A =area of cross-section of conductor in cir. mils. R — conductor resistance both ways. wt =i weight in pounds of conductor copper. m % = commercial efficiency of motor. g % = commercial efficiency of generator. I % — commercial efficiency of whole system. c % = per cent of energy lost in conductor. all % expressed as a decimal, as, 90 % = .90, A7" Then, —^ = electrical horse-power delivered at motor terminals : m% 746 & /» and 7=-^^ = amperes. (5) By formula No. 1, R = -, — '- — = resistance of conductor both ways. SIZES OF CONDUCTORS. 101 The drop or loss in the line v=zl R, or V = IXDX 21.02. (6) (7) Substituting the value for I, . 746 xi^X Dx 21.62 we have, A = ^ — „ ; (8) E X m% X v w , , . . 16128.5 XJ^XD and reducing we have, — = = . & ' EXm%xv Example : — Motor 20 h.p. m% =90%. Yolts at terminals = 500. Distance = 200 ft. Loss in conductors = 5 % . Then, E.M.F. of generator = ^ = 526.3 volts, and drop in line, v = 526.3 — 500 = 26.3 ; •o ^ t, ^ t /« t 746 V 746 X 20 But by formula (5), /= v,^m0/ » or I = = 33 amperes ; ' Exm% 500 X .90 and the National code only allows 8 amperes for No. 16, and 33 amperes would need at least No. 10 wire. The volts drop and per cent loss in No. 10 B. and S. wire, required to carry the 33 amperes as above shown, will be found as follows : — R of No. 10 B. and S. = .0009972 per foot ; R of 400 ft. = .39888 ohms ; Volts drop = IR — 33 x 39888 = 13.16 volts ; Volts at generator = 500 + 13. = 513. Per cent drop = — - = 2.5 %. 513 SIZJES OF COXDIICTORi FOR ¥1¥CA]¥I»ESCI3]¥T CIRCOTi. (By W. D. Weaver.) The most accurate method of determining the proper sizes of incandescent lamp conductors is to refer all measurements back to the dynamo, converter, or street tap. To illustrate, suppose we have an installa- tion of 150 lights, consisting of a feeder or dynamo main 20 feet long (to distributing point), and several mains, A, B, and C, their lamps and lamp centres being respectively 60, 50, and 40 in number, and 38, 60, and 90 feet from the end of the feeder. Let us calculate the sizes of the feeder and one main, and of one branch having 12 lamps, with centre 20 feet from the main, the branch starting 18 feet from the distribut- » ing point. (See cut.) ~Fig. 1. To find the size of the branch wire, refer to the appropriate table with 20 + 18 + 20 feet, or 58 feet for 12 lamps. To find the size of the main, imagine the branches on one side to be revolved (or lay them out thus on a diagram), so that all are on the same side sl B - , 1 % c c ,1 | ' . CENTRE, 90 F.EET U: sl; 10*2 CONDUCTORS. of the main ; then estimate or calculate the lamp centre of the resultant group, which in this case we will suppose to be 23 feet from the main, and 38 feet from the distributing- point measured along the main, and refer to the table with 2i> + .-;s + 2:> feet for 12 + 30 + is lamps, or 81 feet for GO lamps. To find the size of the feeder, suppose the mains to be revolved about the distributing point so that they all overlap, and with all the branches on one .side of the^overlapping mains ; then estimate or calculate the lamp centre of the resultant group (comprising all the lamps), which in this case we will suppose to be 'JO feet from the overlapping mains measured at right angles, and 48 feet from the distributing point measured along the main, and refer to the table with '20 + 48 -f- 20 feet, or 88 feet for 150 lights, or for the largest number of lights that will ever be used at one time. In simple 'cases the quantities maybe estimated either directly (especially for branches) or from rough diagrams ; and for more complex cases, or where a perfectly accurate result is desired, the following rules are given : — For 1$ randies, follow the method given above. For jfiiiiia*. multiply the number of lamps on each branch of a main by the distance of their lamp centre from the distributing point, always meas- ured along the lead of the main and branch ; add the products thus obtained for all the branches'on the main, and divide by the whole number of lamps on the branches. Add the length of feeder, and refer to the table with the resultant distance and lamps. Example : — (See cut, main A.) (18 + 20) X 12= 456 (33+30) X 30—1890 (60 + 15) X 18 = 1350 456 + 1890 + 1350 , OA Q1 , , „ „. io_l 'M _l i' +20 = 81 feet for 60 lamps. For .'Fenders, add the sum of the products obtained as above for all the mains, divide by the entire number of lamps on the feeder, add the length of the feeder, and refer to the table with this distance and all the lamps on the feeder, or the largest number that will ever be used at one time. Example : — (See cut.) Main A. 456 + 1890 + 1350= 3696 Main B. 60 X 50 =3000 Main C. 90 X 40 =3600 3696 + 3000 + 3609 OQ ! ! U20 = SX feet for 150 lamps. 150 Care must be taken not to confound a lamp centre (so-called) Avith a geo- metrical centre. For example, suppose a series of branches of equal length radiating from the end of a main like the spokes of a wheel, and having lamps at equal intervals. Here the geometrical centre is the radiating point, while the lamp centre is on a circle passing through the centres of the various groups, or the length of the radius from the radiating point. In the case of the main A given above, the geometrical centre is 15 feet from the main, while the true lamp centre is 23 feet. It is to preclude the error of geometrical centres that the branches and mains are laid down, or ima- gined, revolved. $ul»-l>raiiches and 'Taps may in general be considered as groups of lamps directly on the branch itself, and thus included in the calculation for the branch. The above method is applicable to all systems of wiring, and is particularly valuable and economical in securing proper distribution of light on low volt- age circuits having a small percentage of loss. By stringing the branches first, when possible, this method may he easily followed without the aid of a diagram, even in complex cases. With the "closet" system of wiring, diagrams and calculations as a rule will not be required. The " tree" system of wiring is to be avoided where possible, on account of the unequal distribution of light it entails. In many cases, secondary centres of distribution may be substituted ; and if carefully calculated, the weight of wire in the latter case need not exceed that in the former. The voltmeter should always be connected with the centre of distribution, and not with the feeder near the dynamo, unless it is desirable to have a steady light in a particular locality, when it should be connected with the line there. t y 1 CALCULATION OF SIZE. 108 Owing to the exceedingly small current passing through a voltmeter, the resistance oil even a very small wire in ordinary cases will not practically affect its readings. Where the line is very long, a No. 12 or 14 insulated iron wire may be used, and the voltmeter at the dynamo set once for all by comparison with a standard voltmeter temporarily attached at the point which is to be maintained at a constant potential. CAlCULATIOlf OF THE SIZE OF CO\Dt CTOHM FOR AETEJR]¥ATI]1T« CHRBEIT CIRCUITi. When alternating currents first came into use, it was customary to calcu- late the sizes of conductors by the ordinary rules used in connection with direct currents. This did very well as long as small currents were in use, and distances were comparatively short ; but before long new effects began on the lines that were unaccountable to any one not familiar with the action of such currents in a conductor, and this led to a more thorough study of the problems. Briefly stated there are, besides the ohmic resistance of the copper, the following effects, due to tli3 use of alternating currents : — Skin effect, a retardation of the current due to the property of alternating currents of apparently flowing along the outer surface or shell of the con- ductor, thus not making use of the full area. Inductive effects, a, self-induction of the current due to its alternations, in- ducing a counter E.M.F. in the conductor ; and b, mutual inductance, or the effect of other alternating current circuits. Capacity Effects, due to the fact that all lines of conductors act as electri- cal condensers, which are alternately charged and discharged with the fluctuations of the E.M.F. Skin Effect. The increase in resistance due to skin effect can be found by the use of the following table : — Skin Effect Factors, for Conductors carrying* Alternating- Currents. Note. — For true resistance, multiply ohmic resistance by factor from this table. Diam. and Frequencies. B.&S. gauge. 15 20 25 33 40 50 60 80 100 130 2" 1.111 1.160 1.265 1.405 1.531 1.682 1.826 2.074 2.290 2.560 If 1.072 1.114 1.170 1.270 1.366 1.495 1.622 1.841 2.030 2 272 13 1.042 1.064 1.098 1.161 1.223 1.321 1.420 1.610 1.765 1.983 i\ 1.019 1.030 1.053 1.084 1.118 1.176 1.239 1.374 1.506 1.694 {* 1.010 1.019 1.035 1.059 1.080 1.124 1.168 1.270 1.382 1.545 1.005 1.010 1.020 1.038 1.052 1.080 1.111 1.181 1.263 1.397 I" 1.002 1.002 1.007 1.014 1.016 1.028 1.040 1.066 1.100 1.156 \" 1.001 1.001 1.002 1.005 1.006 1.007 1.008 1.011 1.022 1.039 0000 1.001 1.003 1.005 1.005 1.006 1.010 1.015 1.027 000 1.001 1.002 1.002 1.005 1.007 1.010 1.017 00 1. 01 1.001 1.002 1.004 1.006 1.010 0 1.001 1.002 1.005 1.008 1 1.001 1.002 1.005 2 1.001 1.002 3 1.001 4 1.000 104 CONDUCTORS. For other frequencies, Emmet gives the following tahle : — Product of Cir. Mils by Cycles per sec. 10,000,000 1.00 20,000,000 1.01 30,000,000 1.03 40,000,000 1.05 50,000,000 1.08 60,000,000 1.10 70,000,000 1.13 80,000,000 1.17 90,000,000 1.20 100,000,000 1.25 125,000,000 1.34 150,000,000 1.43 Factors in the above table multiplied by the resistance in ohms will give the resistance of circular copper conductors to alternating currents. Effects of Self-induction. — Owing to the periodic variations of current in alternating-current circuits, a counter E.M.F. is set up, which does not coincide with the current, and which is not continuous, but periodic ; and, owing to the fact that such E.M.F. is the strongest when the current is increasing or decreasing most rapidly, the counter E.M.F. differs in phase with the current by 90°. If there be no inductive effect in a circuit (without considering anything else at present), the current produced by an impressed E.M.F. would be in phase, and the watts would be, as in direct currents, the product of the E.M.F. and current. Taking into account the inductive effect, the current is never in phase with the impressed E.M.F., and the watts are therefore never equal to the product of the two, but are less, according to the angle of phase difference ; and if they could be in quadrature, the product would be zero. The E.M.F. impressed on the circuit may be said to be made up of two components, one in phase with the current, as in direct currents, and the other in quadrature with it, as shown below in a right-angle triangle. ef.eect.ive or energy e.m.f. Counter or inductive E.M.F. varies with the Fig. 2. frequency of alternations ; but if the out-going and returning wires are close together, there is little induction ; if wound in a coil, the self-induction is much increased, and if an iron core be introduced into the coil, the flux is very much in- creased, and therefore the self-induction. Impedance. — In a plain, alternating-current circuit without iron, the current due to a given E.M.F. will depend upon a resistance which is the resultant of two components : its resistance as in direct currents, and its inductive resistance, or the current divided into the inductive E.M.F. These two components are compounded at right angles, and the resultant is called impedance, and can be represented by the same triangle as was used to illustrate the two E.M.F.'s and their resultant. Impedance also varies with the rate of alternations the same as does the counter or inductive E.M.F. If we have a circuit including a number of parts ____— --^'lu. *n series, each having a different angle of lag, and ir~ ^^ \. g represented as below by different triangles joined /I *■&%*>*' !■* together, it will be seen that the sum of all the /-" X^^ " E.M.F.'s impressed upon the parts or impedances jy^f ^N | is greater than the E.M.F. impressed upon the ^_ — 1 \° whole circuit ; and in order to arrive at the latter total ENtBGY e.m f. value, it is necessary to lay out each case sepa- Fig. 4. rately, all the horizontal lines representing energy ENERGY resistance Fig. 3. CALCULATION OF SIZE. 105 E.M.F.'s (or resistances), and all the vertical lines representing inductive E.M.F.'s (or resistances, now called reactances). To find the impedance equal to two impedances in parallel, construct a parallelogram, the adjacent sides of which will he the reciprocals of their values ; the diagonal of this parallelogram will be the reciprocal of the value of the resulting impedance ; and, as the lines representing the given impedances are joined at the proper phase angle with each other, the direction of the diagonal will represent the resulting phase. In the above figure -^ = §. Fig. 5. -r— = AD = 1.3 ohms. Ax If two impedances, connected in parallel, have such values as to give a phase difference of 90°, i.e., are at right angles with each other, their result- ant value can be found by constructing a right- angle triangle, whose adjacent sides represent in direction and length the values of the two impedances in parallel. Join the two ends, and a line drawn from this hypothenuse at right angles and meeting the others at their junction, will be equal to and in tbe direction of the resultant value. Yig. 6. -^ ac and ce are tw0 impedances in parallel, with a difference in phase of 90°, then cd equals in direction and in length the resultant of the two. Capacity Effects. — A condenser connected in multiple across the leads of an a. c. circuit is charged as the E.M.F. rises, and discharged as the E.M.F. falls, thus returning E.M.F. to the line just at the time that the inductive E.M.F. is opposing the line E.M.F., and both can be so arranged as to neu- tralize each other, or enough capacity can be introduced to cause a negative lag-angle, as shown in the following figure. When a condenser or a line having capacity is subjected to an alternating E.M.F., current will flow in to fill the capacity equal to E X CX w, where E is the E.M.F., C, tbe capacity in farads, and io = 2Tr N. Thus, if a line has a capacity of 3 micro-farads, 2? = 2000 volts, and .ZV=30, then — Fig. 7. Amperes 7= 1,000,000 X 2000X30X6.28 = .7536. Ceo And a condenser may be said to have a reactance of This reactance is also in quadrature with the energy E.M.F., as is the in- ductive reactance, but acting in the opposite direction to that of the induc- tance ; and may therefore be so arranged as to neutralize it. Line capacity acts like a condenser placed in multiple at the middle point of the length of the line. Lag angles and power factors of alternating-current motors of the induc- tion type vary with the load they carry and with the design and size, some of large size having power factors as high as 97% at full load, while poorly designed motors may have but 75% or less. Synchronous motors run with a separately excited field, which may be so varied as to produce a leading or lagging current, or be made to take from or return energy to the line. When running with but little load, with field cur- rent high, energy will be absorbed from the line as the impressed E.M.F. 106 CONDUCTORS. rises, and returned to the line as it falls, thus acting like a condenser, and tending to steady the E.M.F. of the circuit, which maybe disturbed and lowered by the inductance of induction motors. Closed circuit transformers with secondary open have a power factor of about 70%, and when loaded with non-inductive load, large sizes have a power factor of over'.YJ1;;,, with an induction component of say 6%, even at halt-load the power factor is over 99%. In the ordinary alternating-current lighting circuits, the elements are, the lamps, the secondary circuits, the transformers, the primary mains, and feeders. li distances are considerable and the wires large, there will be some in- duction due to the primary and secondary mains ; but most of the effect will come from the transformer, provided, of course, that nothing but incandes- cent lamps are used as load on the seci mdary. With good-sized transformers, the total power factor will be above 99%. In the following table will be found the angles of lag, together with the power-factors ami factors of induction due to each, from which may be com- puted the effects on lines of different inductances. Power IT actor* and Induction Factors for Different Angles of Lag*. Ch3 «3 £ i.? ft°£ o . £ o -£1 £ Ji.% ft -d £ bJO u - be o o ^ft U 1 S c £ " ft°^ ,2 to if. % a hi > s 4 o so ft°^ bo O O 'xfi o O o O o O .3 m 1 .9998 .0174 24 .9135 .4067 46 .6946 .7193 69 .3584 .9336 2 .9994 .0349 25 .9063 .4226 47 .6820 .7313 70 .3420 .9397 3 .9986 .0523 26 .8988 .4384 48 .6691 .7431 71 .3256 .9455 4 .9976 .0698 27 .8910 .4540 49 .6561 .7547 72 .3090 .9511 5 .9%2 .0872 28 .8829 .4695 50 .6428 .7660 73 .2924 .9563 6 .9945 .1045 29 .8746 .4848 51 .6293 .7771 74 .2756 .9613 7 .9925 .1219 30 .8660 .5000 52 .6156 .7880 75 .2588 .9659 8 .9903 .1392 31 .8572 .5150 53 .6018 .7986 76 .2419 .9703 9 .9877 .1564 32 .8480 .5299 54 .5878 .8090 77 .2249 .9744 10 .9848 .1736 33 .8387 .5446 55 .5736 .8191 78 .2079 .9781 11 .9816 .1908 34 .8290 .5592 56 .5592 .8290 79 .1908 .9816 12 .9781 .2079 35 .8191 .5736 57 .5446 .8387 80 .1736 .9848 13 .9744 .2249 36 .8090 .5878 58 .5299 .8480 81 .1564 .9877 14 .9703 .2419 37 .7986 .6018 59 .5150 .8572 82 .1392 .9903 15 .9659 .2588 38 .7880 .6156 60 .5000 .8660 83 .1219 .9925 10 .9613 .2756 39 .7771 .6293 61 .4848 .8746 84 .1045 .9945 17 .9563 .2924 40 .7660 .6428 62 .4695 .8829 85 .0872 .9962 18 .9511 .3090 41 .7547 .6561 63 .4540 .8910 86 .0698 .9976 19 .9455 .3256 42 .7431 .6691 64 .4384 .8988 87 .0523 .9986 20 .9397 .3420 43 .7313 .6820 65 .4226 .9063 88 .0349 .9994 21 .9336 .3584 44 .7193 .6946 6d .4067 .9135 89 .0174 .9998 22 .9272 .3746 45 .7071 .7071 67 .3907 .9205 23 .9205 .3907 68 .3746 .9272 Inductive Resistance of 'ILines. — As previously stated, two par- allel wires carrying alternating currents induce in each other counter or in- ductive E.M.F.'s that tend to retard the flow of current. The closer together these wires are, the less is this effect, and the more nearly the current waves are to the simple harmonic curve, the less is the retardation. The counter E.M.F. is somewhat larger for small wires than for large, ^^■^■■^■^^■^ INDUCTANCE FACTORS. 107 provided the current and distance between centres be the same, and the effect is about 150 times greater in iron wire circuits than with copper, as will be seen by reference to the following formulae, by which both are cal- culated. ODrClAIGE FACTORS. In Tables I. and II. below are given the formula? for inductance of two parallel wires of copper and of iron ; and in Table III. the inductance per mile for two copper wires has been computed for different inter-axial dis- tances. Table I. — Inductance for Parallel Copper Wires, Insulated. Formula, d — distance apart, centre to centre, of wires r = radius of wires. L = inductance of each wire in millihenrys. - .5 + ( 2 log e - j 10 b > per centimeter. L per centimeter — .000,000,5 + .000,004,6 log- L per inch = .000,001.27 + .000,011,68 log-. L per foot = .000,015,24 + .000,14 i d L per 1,000 feet = .01524 +.14 i d log — . L per mile = .08.5 + .741 i d Talile II. — Inductance for Parallel Iron Wires, Insulated. d = distance apart, centre to centre, of wires. r = radius of wires. Z = inductance of each wire in millihenrys. L =z 75. +( 2 log e - j 10-6, per centimeter. L per centimeter — : .000,075 + .000,004,6 log -. d L per inch L per foot L per 1,000 feet = .2286 + .14 log L per mile = .12,075 + .741 log-. : .000,191 + .000,011,68 log - . :. 002,286 +.000,14 log-. d 108 CONDUCTORS. each of Two Copper Wires Parallel to each other. Interaxial Distance in Inches. B. and S. gauge. 3. 6. 12. 24. 36. 48. 0000 0.907 1.130 1.353 1.576 1.707 1.799 000 0.944 1.168 1.391 1.614 1.745 1.836 00 0.982 1.205 1.425 1.651 1.784 1.874 0 1.019 1.242 1.465 1.688 1.818 1.911 1 1.056 1.280 1.502 1.725 1.856 1.949 2 1.094 1.317 1.540 1.764 1.893 1.986 3 1.131 1.354 1.577 1.800 1.931 2.023 4 1.168 1.392 1.614 1.838 1.968 2.061 5 1.206 1.429 1.652 1.875 2.005 2.099 6 1.243 1.466 1.689 1.912 2.043 2.135 7 1.280 1.503 1.727 1.949 2.079 2.172 8 1.317 1.540 1.764 1.986 2.117 2.209 9 1.355 1.578 1.801 2.025 2.155 2.248 10 1.392 1.615 1.838 2.061 2.192 2.285 11 1.429 1.652 1.875 2.099 2.229 2.322 12 1.467 1.690 1.913 2.135 2.266 2.359 Inductance in Millihenrys per lOOO feet of Copper Circuit. 2 AERIAL WIRES. Interaxial Distance. \" dia. £" dia. \" dia. B. and S. 0000 000 00 0 3" .248 .283 .333 .344 .358 .373 .386 6 .333 .369 .417 .428 .442 .456 .471 12 AM .451 .500 .513 .527 .540 .555 24 .500 .538 .587 .597 .611 .625 .640 48 .587 .621 .671 .681 .695 .710 .724 Interaxial Distance. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 3" .400 .415 .429 .442 .457 .472 .484 .499 .513 .527 6 .485 .498 .513 .527 .541 .555 .570 .583 .597 .612 12 .570 .583 .597 .612 .626 .640 .654 .668 .683 .696 24 .654 .668 .682 .696 .711 .724 .738 .753 .767 .781 48 .738 .752 .767 .781 .795 .808 .823 .837 .851 .865 Rift = millihenrys per centimeter ; = millihenrys per 1000 ft. of copper wire. Inductive resistance = 2 77 n x millihenrys from above table _ henrys per 1000 feet of circuit. Inductive drop = current X inductive resistance. INDUCTANCE OF THREE-PHASE SYSTEM. 109 IHTRTJCTAUCE PER MTEE OF CIRCUIT THREE- PHASE SYSTEM, ©O p. p. s. (Dr. P. A. C . Perrine in Trans. A .I.E. E.) 02 REACI.4KCE OF AITEHSAT- THHG CUBREiYT CIRCUIT'S. By Steinmetz. Let Ez: resistance in ohms. L = impedance. E = power E.M.F. err impressed E.M.F. (a = 2w7l. L =r coefficient of self-induction. I = current. c = capacity. Then : In circuits containing Resistance* and Inductance, Impedance, Z, — VR2 -f L2 a>2, and e = VE2 + 12 L2 w2 ; or diagrammatically, Fig. 8. Fig. 9. Circuits containing Resistance and Capacity. Impedance, Z,= VR2 -4- and e = Ve2 - or diagramatically, Fig. 10. Fig. 11. Circuits containing Resistance, Inductance, and Capacity. Impedance, Z,= Vw + (lw -£)'■ and e matically, _ 3 E2 + I (— M- Fig. 12. Fig. 13. IMPEDANCE AND REACTANCE. 115 Impedance factors and Multipliers. Frequency = 100. . /r)/ f— 25 1 1.842 2.182 2.535 2.904 2 7.7638 10.37 12.912 15.55 if 1.387 1.546 1.720 1.903 n 5.014 6.454 7.831 9.017 A 1.111 1.157 1.210 1.267 i 2.7i r>4 3.3826 4.012 4.642 0000 1.085 1.120 1.167 1.203 * 1.889 2.209 2.543 2.885 000 1.057 1.081 1.108 1.137 1.285 1.393 1.513 1.637 00 1.038 1.054 1.068 1.090 0 1.0264 1.036 1.048 1.061 0000 1.3068 1.3996 1.498 000 1.152 1.2104 1.2763 1.345 /— 2 0 00 1.1034 1.1422 1.1876 1.235 1 2.291 2.771 3.261 3.768 0 1.0710 1.0973 1.1277 1.160 * 1.624 1.863 2.116 2.378 1 1.0478 1.0676 1.0853 1.108 * 1.190 1.263 1.351 1.441 2 1 .0324 1.0443 1.0583 1.071 0000 1.146 1.206 1.271 1.341 3 1.0216 1.0293 1.03S4 1.048 000 1.100 1.139 1.184 1.233 4 1.0142 1.0191 1.0247 1.031 00 1.067 1.093 1.123 1.155 5 1.0094 1.0126 1.0162 1.0203 0 1.046 1.063 1.084 1.106 6 1.0063 1.0084 1.0107 1.0134 116 CONDUCTORS. Impedance factors. — Continued. * 8 8 8 8 * •!§ .- 8 • § • k 1 I Fig. 16. Convenient Arrangement of Three-phase Lines for 6000-10000 Volts. Balanced JLine, Three-Phase. — The following diagram shows an arrangement of the conductors of a three-phase circuit, which will be balanced in all its effects if there be but one circuit. The distances, 18 inches apart, are about standard for pressures as high as 12,000 volts. 8- 4 18 Fig. 17. Balanced Arrangement for Three-phase Lines. This arrangement is perhaps not so convenient for repairs, but is symmet- rical in all respects. If there be more than one circuit of this balanced arrangement, and the difference of phase is enough so that interference is found, then one or more of the circuits will have to be changed as shown in the following IMPEDANCE AND REACTANCE. 119 diagram (Fig. 18), the principle being to bring each of the three wires a circuit into the same relation with other circuits for an equal length or distance. Fig. 18. Arrangement of Three Three-phase Circuits, each Equilaterally Placed. In this Arrangement there is no Effect from One Circuit on Another. Three-phase Circuit in Same I*la«e. — It is sometimes advan- tageous to place all the conductors on one cross-arm on the same level as in the preceding diagram. In this case, if the load is heavy enough to cause interference between conductors, then two interchanges of wires should be made, dividing the circuit into three equal parts as shown. This will bring every wire into similar relations with all others, and tbe interfer- ence will therefore be the same on all. In order that this balancing effect should be correct along a line having branches, the reversals should be made between all branches; for instance, between the dynamo and the first branch there should be two reversals as shown, and between the first and second branches the reversals should be repeated, and so on. 120 CONDUCTORS. . If Wires of Three-phase Circuit are on same Plane, then they should be interchanged twice between Points when Branches are attached, as 2 Fzo.2, Another Arrangement ^Two-phase Circuit. No Keversal of mmmr^^^m^^m^m^ IMPEDANCE AND REACTANCE. 121 reversals of wires are needed, the inductive effects of the wires of one circuit on those of the other are neutralized. Two-Phase Circuits in Same Plane. — If the phases are treated as separate circuits, and carried well apart, the interference is trifling ; and should the loads carried be heavy enough to cause noticeable effect, the re- versal of one of the phases in the middle of its length will obviate it. The following diagram illustrates the meaning. J>C Fig. 23. Arrangement of Two-phase Four-wire Circuit with "Wires on same Plane. Wires of One Phase should be interchanged at the Middle Point of the Distance between Branches, and between its Origin and First Branch. Messrs. Scott and Mershon of the Westinghouse Electric and Manufactur- ing Co. have made special studies of the question of mutual induction of circuits, both in theory and practice ; and their papers can be found in the files of the technical journals, and supply full detail information. AI/EEltarATIXtt WIRING AIVI9 CODiOCVIO]V§. By General Electric Company. General Wiring- formulae. The following general formulae may be used to determine the size of con- ductors, volts lost in the line, and current per conductor for any system of electrical distribution. Area of conductor, Circular Mils = PXE2' P X E Volts loss in line = ■ x M. W Current in main conductors = — x T. Z)rr Distance of transmission (one way), in feet. W = Total watts delivered to consumer. P = Per cent loss in line of W. E = Voltage between main conductors at receiving or consumers' end of circuit. In using the above formulae and constants, it should be particularly observed that P stands for the per cent loss in the line of the delivered power, and not for the per cent loss in the line of the power at the generator. In continuous-current, three-wire systems, the neutral wire for feeders should be made of one-third the section obtained by the formula! for either of the outside wires. In both continuous and alternating current systems, the neutral conductor for secondary mains and house-wiring should be taken as large as the other conductors. When both motors and lights are used, on the Monocyclic System, the primary circuit should be figured as if all the power was transmitted over the outside wires, and the size of the power wire should be in the proportion to either outside wire, as the motor load in amperes is to the total load in am- peres. Secondary wires leading directly to induction motors on the Mono- cyclic System should all be of the same size as for a single-phase circuit of the same kilowatt capacity and power-factor. The three lines of three- phase circuits should be made of the same cross-section. 122 CONDUCTORS. < o "e3 Values of K. Values of T. System. Per cent power factor. Per cent power factor. 100 95 90 85 80 100 95 90 85 80 Single-phase .... Two-phase (four-wire) Three-phase (three-wire) 6.04 12.08 9.06 2160 11 ISO 11 ISO 2400 1200 1200 2601) i;;:;o i;j:;o :;ooo ir.00 lr.oo ;;:>xo 1090 1690 1.00 .50 .58 1.05 .53 .61 1.11 .55 .64 1.17 .59 .68 1.25 .62 .72 The value of K for any particular power factor is obtained by dividing 2160, the value for continuous current, by the square of that power factor for single-phase, and by twice the square of that power factor for three- wire three-phase, or four-wire two-phase. The value of M depends on the size of wire, frequency and power factor. It is equal to 1 for continuous current, and for alternating current with 100 per cent power factor and sizes of wire given in the following table of wiring constants. The figures given are for wires 18 inches apart, and are sufficiently accu- rate for all practical purposes, provided the displacement in phase between current and E.M.F. at the receiving end is not very much greater than that at the generator ; in other words, provided that the reactance of the line is not excessive, or the line loss unusually high. For example, the constants should not be applied at 125 cycles if the largest conductors are used, and the loss 20 % or more of the power delivered. At lower frequencies, how- ever, the constants are reasonably correct, even under such extreme con- ditions. They represent about the true values at 10 % line loss, are close enough at all losses less than 10 %, and often, at least for frequencies up to 40 cycles, close enough for even much larger losses. Where the conductors of a circuit are nearer each other than 18", the volts loss will be less than given by the formulae, and if close together, as with multiple conductor cable, the loss will be only that due to resistance. The value of T depends on the system and power factor. It is equal to 1 for continuous current, and for single-phase current of 100 per cent power factor. The value of A and the weights of the wires in the table are based on .O00J03O2 lb. as the weight of a foot of copper wire of one circular mil area. In using the above formulae and constants, it should be particularly observed that P stands for the per cent loss in the line of the delivered power, not for the percent loss in the line of the power at the generator ; and that E is the potential at the end of the line and not at the generator. When the power factor cannot be more accurately determined, it may be assumed to be as follows for any alternating system operating under aver- age conditions : Lighting with no motors, 95% ; lighting and motors to- gether, 85 % ; motors alone, 80 %. In continuous current three-wire systems, the neutral wire for feeders should be made of one-third the section obtained by the formulae for either of the outside wires. In both continuous and alternating current systems, the neutral conductor for secondary mains and house-wiring should be taken as large as the other conductors. When both motors and lights are used on the Monocyclic System, the pri- mary circuit should be figured as if all the power was transmitted over the outside wires, and the ^ize of the power wire should be in the proportion to either outside wire as the motor load in amperes is to the total load in am- peres. Secondary wires leading directly to induction motors on the Mono- cyclic system should all be of the same size as for a single-phase circuit of the same kilowatt capacity and power factor. The three wires of a three- phase circuit, and the four wires of a two-phase circuit should all be made the same size, and each conductor should be of the cross section given by the first formulae. WIRING CONSTANTS. 123 o o § 8 § o r-l CI co "* m co t- 00 © © 6 o p c3 c3 CO O CD O o CD O t?5 _^ o ^§ Ph 6 1 © ■* •# © CO CM in CM CM OS t~ in 3 2 2 ^ ^ q q s 3 & o CO cn CO O s s q £-< -r CS t-h q q 8 O os CO CO CO ■* oo ci CO t? § ® cS t- o -i< © § ^ © cn ei §g £ CO lO 1C ■* m co s tH CO y-i © © -5H © © # ># m s 2 q q q 8 8 8 o 55 CO 8 o s s § 3 q q § ^ § 1 S CO •# 3 CO '3 o 8 o 0) © Ph £ o Ph S s§ ^ CO s CO o q © s o q q 8 © © © © 8 8 IO 3 5i CO o S ^ o q q o o © © 8 8 8 8 o © .a s s 5 q q q CO q q q 8 8 © © © © OS S3 s -r p l- LO q q q q 8 8 8 8 1 1 9.1 •su IAV J« no *o oos ^« 39.1 0001 -I9n CO CO 00 CM S 8 co in © co" 1 1 CO © ■a 9JTA1 ;o •OR i i s ° © o o r-l CM co-* 124 CONDUCTORS. The venerators are rated by tlieir volt-ampere capacity and their apparent watts, and not their actual watts, so that the size has to be increased if the power-factor of the system is low. xnA^rsFomfTEiis. For lighting circuits using small transformers, the voltage at the prima- ries of the step-down transformers should be made about 3% higher than the secondary voltage multiplied by the ratio of transformation, to allow for the drop in transformers. In large lighting transformers this drop may be as low as 2%. Standard lighting transformers have a ratio of 10 to 1 or some mul- tiple thereof. For motor circuits, the voltage at the primaries of step-down transformers should be made about 5% higher than the secondary voltage multiplied by the ratio of transformation. Transformers used with 110 volt motors on any 60-cycle system should have a ratio of 4J to 1, 9 to 1, or 18 to 1 respectively for 1040, 2080, and 3120 volt generators. Transformers with a ratio of 10 or 20 to 1 should on no account be installed with motors operated from Mono- cyclic generators of standard voltage. The transformer capacity inkilotcatts should be the same as the motor rating in horse-power for medium-sized motors, and slightly larger for small motors, and where only two trans- formers are used. Capacities of Transformers to l»e used with ©©-Cycle Induction JfEotors. Kilowatts per Transformer. Size of Motor. Horse-Power. Two Transformers. Three Transformers. 1 .6 .6 2 1.5 1 3 2 1.5 5 3 2 n 4 3 10 5 4 15 7.5 5 20 10 7.5 30 15 10 50 25 15 75 25 onrcTiow motor§. The standard (General Electric) induction motors for three-phase and for monocyclic circuits are wound for 110 volts, 220 volts, and 550 volts ; motors of 50 H. P. and above are, in addition, wound for 1040 volts and 2080 volts. Motors for the two latter voltages are not built in sizes of less than 50 H. P. Where the four-wire three-phase distribution system is used, motors can also be wound for 200 volts. The output of an indtiction motor varies with the square of the voltage at the motor terminals. Thus, if the volts at the terminals happen to be 15% low, that is, only 85% of the rated voltage, a motor, which at the rated volt- age gives a maximum of 150% of its rated output, will be able to give at the 15% lower voltage, only (T8,fo)2 X 150= 108 % of its rated output, and at full load will have no margin left to carry over sudden fluctuations of load while running. INDUCTION" MOTORS. 125 Thus it is of the utmost importance to take care that the volts at the motor terminals are not below the rated volts, hut rather slightly above at no load, so as not to drop below rated voltage at full-load or over-load. The output of the motor may be increased by raising the potential ; in this case, nowever, the current taken is increased, especially at light loads. The direction of rotation of an induction motor on a three-phase or mono- cyclic circuit, can be reversed by changing any two of the leads to the field. Like all electrical apparatus, the induction motor works most efficiently at or near full load, and its efficiency decreases at light load. Besides this, when running at light load, or no load, the induction motor draws from the lines a current of about 30% to 35% of the full-load current. This current does not represent energy, and is not therefore measured by the recording watt-meter ; it constitutes no waste of power, being merely what is called an idle or "wattless" current. If, however, many induction motors are ope- rated at light loads from a generator, the combined wattless currents of the motors may represent a considerable part of the rated current of the gene- rator, and thus the generator will send a considerable current over the line. This current is wattless, and does not do any work, so that in an extreme case an alternator may run at apparently half -load or nearly full-load cur- rent, and still the engine driving it run light. While these idle currents are in general not objectionable, since they do not represent any waste of power, they are undesirable when excessive, by increasing the current-heat- ing of the generator. Therefore ft is desirable to keep the idle currents in the system as low as possible, by carefully choosing proper capacities of motors. These idle currents are a comparatively small per cent of the total current at or near full-load of the motor, but a larger per cent at light loads. Therefore care sbould be taken not to install larger motors than necessary to do the required work, since in this case the motors would have to work continuously at light loads, thereby producing a larger per cent of idle cur- rent in the system than would be produced by motors of proper capacity ; that is, motors running mostly between half-load and full load. Current taken l>y General Electric Co. Three-phase In- duction Motors at HO Volts. Starting Starting H. P. of Motor. Full-Load Current at Current Current. 150% of Full- at Full-Load Load Torque. Torque. 1 6.3 19 2 12 36 3 18 54 5 28 *42-84 28 10 54 70 54 15 81 120 81 20 112 167 112 30 168 252 168 50 268 400 268 75 390 585 390 100 550 825 550 150 780 1180 780 * The 5 H. P. motor is made with or without starting-switch. The current taken by motors of higher voltage than 110 will be proportion- ally less. The above are average current values, and in particular cases the values may vary slightly. 126 CONDUCTORS. Isolated motors running on the Monocyclic System are operated from two transformers, connected as shown in Fig. 24. Where there is no high-tension transmission line, the step-up and step-down transformers are not required, and only the two motor transformers shown at the right in the diagram are used. The connections of a Monocyclic circuit for the operation of a three-wire Fig. 24. Fig. 25. secondary lighting system and motors is shown in Fig. 25. The main trans- former has three terminals brought out from each winding, and a supple- mentary motor transformer is used and connected as shown. Where this connection is used for the operation of a single motor, the kilo- watt rating of the supplementary transformer should be about one-half of the motor rating in horse-power. This arrangement is primarily intended for secondary mains carrying lights and a number of motors. Judgment should be exercised in the use of this arrangement, since, if the motors con- nected are large as compared with the total capacity of the transformers, the fluctuations of load may effect the lights to an objectionable degree through variations of drop in the transformers. The motor load being in- ductive, it will cause wider variations of voltage in the transformers than would be experienced with the same current delivered to lights. The connections of three transformers, with their primaries, to the genera- tor and their secondaries to the induction motor, in a three-phase system, are shown in Fig. 26. The three transformers are connected with their pri- maries between the three lines leading from the generator, and the three secondaries are connected to the three lines leading to the motor, in what is called delta connection. The connection of two transformers for the supply of an induction motor from a three-phase generator is shown in Fig. 27. It is identical with the Fig. 26. Fig. 27. arrangement in Fig. 26, except that one of the transformers is left out, and the two other transformers are made correspondingly larger. The copper (•Muired in any three-wire, three-phase circuit for a given power and loss is "5%, as compared with the two-wire single-phase, or four-wire two-phase system having the same voltage between lines. 'The connections of three transformers for a low-tension distribution sys- tem by the four-wire three-phase system are shown in Fig. 28. The three 3H transformers have their primaries joined in delta connection, and their sec- ondaries in "Y" connection. The three upper lines are the three main three-phase lines, and the lowest line is the common neutral. The difference APPLICATIONS OF GENERAL WIRING FORMULA. 127 of potential between the main conductor is 200 volts, while that between either of them and the neutral is 115 volts. 200 volt-motors are joined to the mains, while 115 volt-lamps are connected between the mains and the neutral. The neutral is similar to the neutral wire in the Edison three-wire system, and only carries current when the lamp load is unbalanced. The potential between the main conductors should be used in theformuhe, and the section of neutral wire should be made in the proportion to each of the main conductors that the lighting load is to the total load. When lights only are used, the neutral should be of the same size as either of the three main conductors. The copper then required in a four- wire three-phase sys- tem of secondary distribution to transmit a given power at a given loss is about 3.3.3%, as compared with a two-wire single-phase system, or a four-wire ' vo-phase system having the same voltage across the lamps. The connections of two transformers for supplying motors on the four-wire vo-phase system are shown in Fig. 29. This system practically consists of two separate single-phase circuits, half the power being transmitted over each circuit when the load is balanced. The copper required, as compared with the three-phase system to transmit given power with given loss at the same voltage between lines, is 133£ % — that is, the same as with a single- phase system. APPIICATIOIS OF GEITERAL WIRING EORMEL^. Contirfuous Current. TWO-WIRE SYSTEM. Example : 500 half ampere, 110 volt-lamps. Distance to lights, 1000 ft.; loss in line — 10% of delivered power. 10 X HO2 : 490,900 CM. v 1+ i ' * i 10 x no X 1 „ .. Volts drop to lamp = —— = 11 volts. THREE-WIRE SYSTEM. Example : GOO half-ampere, 110 volt-lamps. Distance to distribution point, 1500 ft. Vrolts between outside lines at distributing point, 220. Loss in line = 8% of delivered power. Area of outside conductors = 2160 X 1500 X (600 X -5 X 110) _ „. 1ftn n ^ 8 X 220* - "'6'100 °-M- The area of the neutral feeder is 276,100 x \ = 92,030 CM. 8 X 220 X 1 Volts drop in circuit = — — — 17.6. 220-f-17.6=i237.6 volts at station between outside lines; and 118.8 volts between outside wires and neutral. Alternating- Currents. TWO-WIRE SIXGLE-PHASE SYSTEM. 125 CYCLES. Example: 1000, 16 c.p., 3.6 watt, 104 volt-lamps. 10 to 1 transformers Distance. 2000 ft. to generator. 2 volts less in secondary wiring. Drop in transformers for lighting is 3%. Loss in primary line to be equal to about "% of power delivered at transformers. Efficiency of transformers. 97%. Volts at transformer primaries = 106 X 10 X 1.03= 1091.8. 1000 X 16 X 3.6 = 57,600 watts. -^ — —== about 60,600 watts at transformer primaries. C.M. = TXfffS X 2400 =48,800 CM. 128 CONDUCTORS. No. 3 B. and S. = 52,633 CM. 2000 X 60,600 X 2400 = 4.64% loss of delivered power, in primary wiring. o2j ooo x loyi.o Volts loss in primary lines = 4.64 x 1091.8 X 1.35 co . m = 68A- 1091.8 + 68.4 = 1160.2 volts at generator. TWO-WIRE SYSTEM. 60 CYCLES. Example : The same load and losses as for the previous problem. Volts at transformor primaries = 106 x 10 X 1.03 = 1091.8. Load at transformer primaries = 60,600 watts. No. 3 B. and S. wire gives 4.64% loss in primary wiring. Volts loss in primary lines = 4.64 X 1091.8 X 1.11 _fi „ ioo =56-2- 1091.8 + 56.2 = 1148 volts at generator. TWO-WIKE SYSTEM, WITH THREE-WIRE SECONDARIES. 60 OR 125 CYCLES. The primary wiring is identical with that for the two-wire system. The secondary wiring is calculated, using the voltage between outside lines, and the three wires are made of the same cross-section. The drop in voltage on the secondary wiring as obtained by the formula is the drop between outside lines, and is twice the drop to each individual lamp. Monocyclic System. ©O Cycles. MOTOR AND LIGHTS ON SEPARATE TRANSFORMERS. (See Fig. 15.) Example : 1500 half-ampere, 104 volt-lamps. One 25 H.P. 110 volt-induc- tion motor ; efficiency, 85%. Distance from generator to transformers, 3000 ft. Distance from transformers to motor, 100 ft. Loss in motor circuit, 2|%. Loss of energy in transformers, 3%. Loss in primary circuit, 4%. Generator voltage, 1040 at no load. 25 X 746 Input at motor = - — — — = 21,940 watts. - 245,000. No. 0000 B. and S. wire = 211,600 2.5 x no2 CM.; but as two No. 0 B. and S. will give the same loss, and -— = 69.2% as great a drop in voltage, they are preferable. Making each motor lead of two No. 0 B. and S. wires in parallel, then P = ^|g^|^g = 2.9%. WU1 , . 2.9X110X1.28 . Volts lOSS tO motors = — = 4. Volts at primaries of transformers for motors = 1.05 X 9 X (110 + 4) = 1076. Volts on secondaries of lighting transformers = = 104.5 l.Uo X 1" Watts at primaries of motor transformers = 21,940 X 1-029 =23200 Watts at primaries of lighting transformers = 1500 X-5X104J =80)8Q(K Total watts delivered at transformers = 23,200 + 80,800=104,000. Power factor of load is 23,200 X -80 + 80,800 X .95 _ 104,000 ~~ K=2» = M10. APPLICATION OF GENERAL WIRING FORMULAE. 129 CM.: 4 X 10762 Taking No. 000 B. and S. wire = 167,805 CM., then P =_ 2610 = 4.19%. Drop in primary circuit = 4.19 X 1076 1.49 X 80.8 + 1.62 X 23.2 oa c 1t 100 X 104 = GS'5 V0lfcs- Voltage between outside lines at generator = 1076 -f 68.5 = 1144.5 volts. Current in main conductors = 777™ — t- = 106.1 amperes. Wto X -91 Primary teazer wire = "" '" - x 167,805 = 37,400 CM. required. 104,000 Use No. 4 B. and S., with a section of 41,742 CM. THEEE-WIEE SECONDARY FOR MOTORS AND LIGHTS. 60 CYCLES. (See Fig. 16.) Example : Distance from generator to transformers, 1000 ft. Ratio of main transformers, 9 to I. The load consists of 1000 half-ampere, 110 volt- lamps, and four 10-11. P. induction-motors. The distance from transformers to motors is 200 ft., and the length of three-wire lighting feeders is 150 ft. The drop in lighting feeders and motor circuits to be about 10 volts. Loss in primary circuit to be 3%. Lamp load = .5 X 110 X 1000 = 55,000 watts. P X E X M Assuming a per cent loss such that — — will be about 10 volts, then CM. = 159°5*5^° X 2400 = 163,600 CM. Taking No. 000 B. and S. wire with an area of 167,805 CM., we have P = 150X55,000 2100_gll 167,805 X 220^ X ^0 _ LA*. T7 .. , . v ... „ , 2.44X220X1.49 0 \ olts loss in lighting feeders = = 8. Voltage at transformers = 220 + 8 = 228. Size of neutral feeder = ~\ — = 55,935 CM., or about No. 2 B. and S. area, 66,373 CM. Input on each 10 H. P. motor at full-load with an efficiency of 84% is equal to 10X746__ .84 P X E X M Assuming a per cent loss such that — is about 8 volts, we have, = 3380 = 35,500 CM. ' ' — 3.5 X 2202 ' No. 5 B. and S. = 33,102 CM. taken for section of motor leads. „ 200 X 8881 X 3380 _ 33,102 X 2202 = 3.75. 17U1 ^ . 3.75 X 220 X 1 Volt loss to motors = — = 8.25. The motor load is 4 x 8881 X 1.0375 = 36,800 watts. The lighting load is 55,000 X 1.0244 = 56,340 watts. 56,340 + 36,800 = 93,140 watts. Assuming transformer efficiencies of 97%, — ^— = 96,000 watts load on transformers. The voltage at the transformer primaries, allowing 4% drop in trans- formers, is 228 X 9 X 1.04 = 2134. 130 CONDUCTORS. _ „ _ 1000 X 96,000 56,340 X 2400 + 36,800 X 3380 . . nnn„ __ °'M— 3X21342" X 967M) -19,000 CM. no. 7 B. and S. = 20,816 CM. 19,000 ._ ^- 20,816 X^--'- o 74 y 21 34 v 1 Volts loss in line = "^ — ' = 58.5. 2134 -f 58.5 == 2192.5 volts a ; 36.8 J 93.4 CM., but this is too small for outside work, hence we would use two No. 7 wires, and one No. 8 wire for the primary circuit. Three-Phase System. ©O Cycles. three-wire transmission. (See Figs. 17 and 18.) Example. — Required : the size of conductors and drop in line to transmit 5000 H.P. 3J miles, with a loss equal to about 10% of the delivered power. Voltage between lines at receiving end, 5000. Power factor of load, 85%. 10 X 50002 Two No. 0000 B. and S. wires per branch would answer ; but the drop in 1.32 voltage will be only — — , or 71.4% as great for the same loss of power, if we take four No. 0 B. and S. wires in parallel, or a line of twelve No. 0 B. and S. wires in all. The loss will be P = 528£ X ^ ^./ °° ^J46 X 1500 = 9.79% of delivered power, i.e., .0979 X 5000 = 489.5 H.P. lost in line. Tr 1+ , .. .. 9.79X5000X1.32 „„„ _ Volts lost in line = = 646 volts. Voltage at generator = 5000 + 646 = 5646 volts. ^1000 v 74fi Current in line =: — — — — x .659 = 506.5 amperes. FOUR-WIRE SECONDARY SYSTEM. (See FlG. 19.) Example. — Required: the size of conductors from transformers to the distributing centre of a four-wire secondary system for lights and motors. The load consists of four 15 H.P., 200 volt-induction motors, and 750 half- ampere, 16 c.p., 115 volt-limps. Length of secondary wiring from trans- formers to distribution centre, 600 ft. About 15 volts drop on lighting circuits from transformers to distributing centre. Efficiency of motors, 85%. 5 volts droi. on circuits from distributing centre to motors. Voltage at dis- tributing point between main lines is 205. Current in main lines for motors . 4 X 15 X 746 X .725 1(V1 1S— ^85x-200 = W1 amperes. Current from transformers for lamps is (750 X .5 X 115) X .607 i 2oo = amPeres. Total current from transformers is 131 -f- 191 = 322 amperes. W For motors, 191 = £- X -725. W— 54,000. 205 W For lamps, 131 = ^ X .607. JF= 44,240. Total watts = 98,240. jj Taking for trial two No. 0B. and S. wires in parallel for each of the main1 "li APPLICATION OF GENERAL WIRING FORMULAE. l3i conductors, as preferable to one No. 0000, then P r 9.75. X 105,592 X 2052 ' 1200 X 44,240 + 1690 X 54,000 _ 98,249 Volts loss in lines = 9.75X205X1.32 = 26.4. Volts at transformers between main lines = 231.4. Actual drop between main conductors and neutral to distributing point = 26.4x^=15.2 volts. 131 X 2 X 105 592 The section of the neutral conductor should be about — - — - — = 86,000 CM. We may use one No. 1 B. and S. wire, with a section of 83,694 CM. for the neutral. Two-I*hase System. O© Cycles. FOUR-WIRE TRANSMISSION. (See FlG. 20.) Example. — Required : the size of conductors and drop in line to transmit 5000 H.P. 3-J miles, with a loss equal to about 10% of the delivered power. Voltage between lines at receiving end, 5000. Power factor of load is 85%. Taking four No. 0 B. and S. wires in parallel, the line will consist of six- 5280 X 3 5 X 5000 X 746 teen No. 0 B. and S. wires in all. The loss will be P = —. ' ..AO ^ _nnn2 4 X 105.592 X 5000^ X 1500 = 9.79% of delivered power, or .0979 x 5000=489.5 H.P. lost in the line. Volts lost in line = PXJgX_jf=9.79x 5000X1.32 100 100 Volts at generating end of line = 5646. ,.. ,. 5000x746 „„„ Current in lme = — — — x -588 = 438.6 amperes. Alternating'-Current Arcs. Power factor is about .75. Calculate wire for apparent watts, not real watts. Chart and Table for calculating: Alternating--Current IJines. Ralph D. Mershon, in American Electrician. The accompanying table, and chart on page 137 include everything neces- sary for calculating the copper of alternating-current lines. The terms, resistance volts, resistance E.M.F., reactance volts, and react- ance E.M.F., refer to the voltages for overcoming the back E.M.F.'s due to resistance and reactance respectively. The following examples illustrate the use of the chart and table. Problem. — Power to be delivered, 250 k.w.; E.M.F. to be delivered, 2000 volts ; distance of transmission, 10,000 ft.; size of wire, No. 0; distance be- tween wires, 18 inches ; power factor of load, .8 ; alternations, 7200 per min- ute. Find the line loss and drop. The power factor is that fraction by which the apparent power or volt-am- peres must be multiplied to give the true power or watts. Therefore the 250 k w apparent power to be delivered is - — -^^ = 312.5 apparent k.w., or 312,500 volt-amperes, or apparent watts. The current, therefore, at 2000 volts will be 312 500 — -'— = 156.25 amperes. From the table of reactances, under the heading " 18 inches," and corresponding to No. 0 wire, is obtained the constant, .228. Bearing the instructions of the table in mind, the reactance volts of this 132 CONDUCTORS. line are 156.25 (amperes) x 10 (thousands of feet) x .228 = 356.3 volts, which are 17.8 per cent of the 2000 volts to be delivered. From the column headed " Resistance Volts," and corresponding to No. 0 wire, is obtained the constant .197. The resistance volts of the line are, therefore, 156.25 (amperes) X 10 (thousands of feet) X .197 = 307.8 volts, which are 15.4 per cent of the 2000 volts to be delivered. Starting, in accordance Avith the instructions of the sheet, from the point where the vertical line, which at the bottom of the sheet is marked " Load Power Factor .8," intersects the inner or smallest circle, lay off horizontally and to the right the resistance E.M.F. in per cent (15.4), and "from the point thus obtained," lay off vertically the reactance E.M.F. in per cent (17.8). The last point falls at about 23 per cent, as given by the circular arcs. This, then, is the drop in per cent of the E.M.F. delivered. The drop in per cent of the generator E.M.F. is, of course, " t 0 = 18.7 per cent. The resistance volts in this case being 307.8, and the current 156.25 am- peres, the energy loss is 307.8 x 156.25 = 48.1 k.w. The percentage lots is tr-x- ' = 16.1. Therefore, for the problem taken, the drop is 18.7 per cent, and the energy loss is 16.1 per cent. If the problem be to find the size of wire for a given drop, it must be solved by trial. Assume a size of wire, and calculate the drop in the manner above indicated; the result in connection with the table will show the direction and extent of the change necessary in the size of wire to give the required drop. The table is made out for 7200 alternations per minute, but will answer for any other number. For instance, for 16,000 alternations, multiply the reactances by 16000 -f 7200 = 2.22. As an illustration of the method of calculating the drop in a line and trans- former, and also of the use of the table and chart in calculating low-voltage mains, the following example is given : — Problem. — A single-phase, induction motor is to be supplied with 20 am- peres at 200 volts ; alternations, 7200 per minute ; power factor, .78. The distance from transformer to motor is 150 ft., and the line is No. 5 wire, 6 inches betAveen centres of conductors. The transformer reduces in the ratio 2000 : 200, and has a capacity of 25 amperes at 200 volts ; when delivering this current and voltage, its resistance E.M.F. is as 2.5 per cent, and its reactance E.M.F. 5 per cent, both of these constants being furnished by the makers. Find the drop. The reactance of 1000 ft. of circuit, consisting of two No. 5 wires, 6 inches apart, is .204. The reactance-volts, therefore, are .204 x j^ X 20= .61 volts. The resistance-volts are .627 x -^ X 20 = 1.88 volts. At 25 amperes, the re- sistance-volts of the transformers are 2.5 per cent of 200, or 5 volts. At 20 amperes they are ^ of this, or 4 volts. Similarly, the transformer reactance volts at 25 amperes are 10, and at 20 amperes are 8 volts. The combined re- actance-volts of transformer and line are 8 + .61 = 8.61, Avhich is 4.3 per cent of the 200 volts to be delivered. The combined resistance-volts are 1.88 + 4, or 5.88, Avhich is 2.94 per cent of the E.M.F. to be delivered. Combining the^e quantities on the chart Avith a power factor of .78, the drop is 5 per cent of the delivered E.M.F., or ~ = 4.8 per cent of the impressed E.M.F. The 105 transformer must therefore be supplied with 2000+ .952 = 2100 volts, in order that 200 volts shall be delivered to the motor. To calculate a four-Avire, two-phased transmission circuit, compute, as above, the single-phased circuit required to transmit one-half the poAver at the same voltage. The two-phase transmission will require tAVO such circuits. To calculate a three-phase transmission, compute, as above, a single-phase circuit to carry one-half the load at the same voltage. The three-phase transmission Avill require three Avires of the size obtained for the single-phase circuit, and with the same distance (triangular) betAveen centres. By means of the table calculate the Resistance- Volts and the Reactance- ± APPLICATION OF GENERAL WIRING FORMULAE. 133 Volts in the line, and find what per cent each is of the E.M.F. delivered at the end of the line. Starting from the point on the chart where the vertical line corresponding with power factor of the load intersects the smallest circle, lay off in per cent the resistance E.M.F. horizontally and to the right ; from the point thus obtained lay off upward in per cent the reactance E.M.F. The circle on which the last point falls gives the drop in per cent of the E.M.F. delivered at the end of the line. Every tenth circle-arc is marked Avith the per cent drop to which it corresponds. 3 3 eg §1 o ^ b Reactance-Volts in 1000 ft. of Line (= 2000 ft. of Wire) of ^ •S| a lor One Ampere (V Mean Square) at 7200 Alternations per Minute for the Distance given between Centres of Wire B.&S. - tf. 5-7 60 IE > Conductors. "7 £ 5 £ £.3 < 2 1" 2" 3" 6" 9" 12" IS" 24" 30" 36" 0000 639 .098 .046 .079 .111 .130 .161 .180 .193 .212 .225 .235 .244 000 507 .124 .052 .085 .116 .135 .167 .185 .199 .217 .230 .241 .249 00 402 .156 .057 .090 .121 .140 .172 .190 .204 .222 .236 .246 .254 0 319 .197 .033 .095 .127 .145 .177 .196 .209 .228 .241 .251 .259 1 253 .248 .063 .101 .132 .151 .183 .201 .214 .233 .246 .262 .265 2 201 .313 .074 .106 .138 .156 .188 .203 .220 .238 .252 .270 3 159 .394 .079 .112 .143 .162 .193 .212 .225 .244 .257 .267 .275 4 126 .497 .085 .117 .149 .167 .199 .217 .230 .249 .262 .272 .281 5 loo .627 .090 .121 .154 .172 .204 .223 .236 .254 .268 .278 .286 6 79 .791 .095 .127 .158 .178 .209 .228 .241 .260 .272 .283 .291 7 63 .997 .101 .132 .164 .183 .214 .233 .246 .265 .278 .288 .296 8 50 1.260 .103 .138 .169 .188 .220 .238 .252 .270 .284 .293 .302 134 CONDUCTORS. CHARGIiYG CURHEIT I»EM TIIIE OW CIRCUIT. Two Parallel Wires. Line E.M.F.= 10,000 Volts; Frequency =60 P.P.S; Sine Wave Assumed. Stanley Electric Manufacturing Co., Pittsfield, Mass. Charging Current in Amperes. 12 .0426 18 .0385 24 .0362 48 .0315 12 .0411 18 .0375 24 .0353 48 .0308 12 .0403 18 .0366 24 .0345 48 .0302 12 .0392 18 .0358 24 .0328 48 .0296 12 .0381 18 .0349 24 .0329 48 .02905 12 .0372 ' 18 .0342 24 .0322 48 .0284 12 .0365 18 .0333 24 .0315 48 .0279 03 a a © Charging Current in Amperes. 12 .0353 4 18 .0326 24 .0308 48 .0274 12 .0345 5 18 .0319 1 24 .0302 1 48 .0269 1 12 .0337 6 i 18 .0312 ! 24 .0296 48 .0264 12 .0329 7 18 .0305 24 .0290 48 .0259 12 .0322 8 18 .0295 24 .0284 4S .02545 12 .0315 9 18 .02925 24 .0278 48 .0250 12 .0308 10 IS .0285 24 .0273 48 .0246 Charging currents = - 7?-- Line E.M.F. N=. Frequency. k Capacity per mile of line in E.M.F. £5 CHARGING CURRENT PER MILE OF CIRCUIT. 135 It-h-* coloco©co t~i-iom© „ , W t-h CO t- CC o co co © © COClrl , w -f 01 © ~. — >- : : -r - X —. cc © © ^_j , ij; J5 ^ CO cc co tO-*tC2t-i OO O 3 lO CO rt N O iO O , . r»O00artl»L-ML".-t 1-0 © ooqeo<»qiqrH pg S3 ^cl i tOHOoiooioMoi-rco^ p CN It— i CM CI W* LO t~ C3 ^ lO O t- 1-1 1 _; ,_; ^ ,-i t-: -i t-h" t-h cn in co co ■BWPA ooo'oi -s"; x -" x 3 t- ci Lo ~ f 'co cc- ~i oV C co © ~ ci -t- x © ©- fi c ; c"i © :c :e co •# -# -*. 5 CO CO CO CO CO CO cc © co © © © cc ., .-. ~. r. r 38838 y%\\ -4 o i-o © © -j t^ l2 '- x co x 888888888888 cc ic ri :o -h i.o i - cc — x x cc - -: c i --. i - -r cc -f ~ :o r. co ~ co -* co co co_ q co t- -H co ■# " tH T-i -i CI CI CO © 2 o o o © CO -# © ip CI t- -h co © © © ■* CO CN CI t-i sssas 5 © ri (N CO rj< IC © t> 0 . © c; r ,-■ -, co io cc cc — t-h l < > 7467 73lT 7l95 1.169 1.109 .744 .709 .617 .567 0 .5328 12 .00147 1.23 1.15 .615 .554 .369 .231 1.340 1.267 .814 .769 .648 .581 IS .001 CO 1.34 1.26 m\ .603 .402 .251 1.442 1.368 .855 .805 .667 .589 24 .ook;;) 1.41 1.33 .707 .637 .425 .265 1.507 1.433 .887 .830 .682 .595 6 .00128 1.07 1.00 .535 7482 .322 .20T 17263 1.204 .858 .826 .744 .700 .6706 12 .00150 1.25 1.18 .627 .565 .377 .236 1.419 1.357 .918 .877 .770 .711 1 is .00163 1.36 1.28 .682 .614 .409 .256 1.5K 1.445 .956 .909 .785 .718 24 .00172 1.44 1.35 .719 .648 .432 .270 1.580 1.507 .983 .933 .798 .723 6 760130 l76 <^ - :.._; it_. .-.-.' :?t'--::;: -■- ;- ^ <: : ■:: ^^ \ '-. 'S: ::'- 1 X :^s Each Small Divis "" r! ■ \ N \ ~^ \- ~ X\ — ^ K S \ V .:■ \ V ^ vS N V •\ ^v- \ K^ ^ ^ x X Nr V V V \ k ■o \ ■ V \ ^V \ \ \ \ \ \ V V \ V N v| ■ • — : :\ N\ \ " ■ ■ \ v :. \ X \ k V. p V -X :A ~~.~: k; V ^ \ ■■■ ^ S '•■' \ V ^ -V— \ 1 \ ^\ • ■ ■ \ :A \ v.:-\ ~: \ \"' \ V \ V ■ Y . ■ ■ >v-, V \ A A A _\ \ V' V - Ah V A : . ■ : : \--. ■v. \ ^T \ ■ ■■■ \ - V- \ V V t ■" ' :\ . \ v \ r V v A" ... ■ 4 -V- 1 b 1 -4 P -4 p V Vrj \ \ \ £ k u h k V ,: ; ■ - X # A: ^ \ -v n '\ - ■\-:. A \ .A 5 \ A ■v; A A -" i — - \ — — t I \ \ \ -:- J . : .( :; ,- , ; - : :::. !!!! n:i fit] !!!l ItlJ i :!;! ;i! Load Power Factors 10 20 30 Drop in Percent of E.3I.E. Delivered BEII WIMIWCJ. The following diagrams show various methods of connecting up-call bells for different purposes, and will indicate ways in which incandescent lamps may also be'connected to accomplish different results. =£■ 6=6= Fig. 31. One Bell, operated by one Fig 32. One Bell, operated by Two Push, Pushes. 138 CONDUCTORS. Fig. 33. Two Bells, operated by One Push. FrG. 34. Two Bells, operated by Two Pushes. When two or more hells are required to ring from one push, the common practice is to connect them in series, i.e., wire from one directly to the next, and to make all but one single-stroke ends. Bells connected in multiple arc, as in diagram No. 24, give better satisfaction, although requiring more J- Fig. 35. Three-line Factory Call. A number of Bells operated by any number of pushes. All bells rung by each push. Fig. 36. Simple button, Three- line Return Call. One set of battery. \t -ft- FlG. 37. Simple Button, Two-Line and Ground Return Call. One set of Battery. Fig. 38. Two-Line Return Call. Illustrating use of Return Call Button. Bells ring separately. A Fig. 39. One-Line and Ground Return Call. Illustrating use of Return Call- Button. Bells ring separately. Fig. 40. Simple Button, Two- Line Return Call. Bells rinj together. Fig. 41. Simple-Button, One-Line and Ground Return Call. Bells ring together. The use of com- plete metallic circuit in place of ground connection is advised in all cases where expense of wire is not considerable. FlG. 42. Four Indication Annuncia- tor. Connections drawn for two buttons only. A burglar alarm cir- cuit is similar to the above, but with one extra Avire running from door or window-spring side of bat- tery to burglar alarm in order to operate continuous ringing attach- ment. BELL WIRING. 139 G i*L I Fig. 43. Four Indication Annuncia- tor, with extra Bell to ring from one Push only. Illustrating use of three-point hutton. Fig. 44. Acoustic Telephone with Magneto Bell Return Call. Ex- tension Bell at one end of line. In running lines between any two points, use care to place the battery, if possible, near the push-button end of the line, as a slight leakage in fheVir- cuit will not then weaken the battery. IP T MAT Fig. 45. Diagram of Burglar-Alarm Mat, two Bells, one Push and Automatic Drop ; all operated by one battery. Both bells ring from one push or mat, as desired, by changing the switch. When mat is to be used, throw it into the circuit by the switch, so that when the circuit is closed by a person stepping on the mat, the automatic drop will keep it closed, and both bells will continue to ring until the drop is hooked up again. (;as-li<-hi wraiNG. Fig. 46. Pendent and A utomatic Gas- Lighting Circuit, with Switch-board. Fig. 47. Pendent Gas-Lighting Cir- cuit, with Switch-board, Relay, and Tell-Tale Bell. Fig. 48. Diagram showing arrangement of circuits for Fire-Alarm or District- Messenger Service. Fig. 1 represents the engine-house or cen- tral station containing the local or open cir- cuit (8). 2 Represents the main or closed circuit on which is located the fire-alarm or messenger boxes (9). 3 Is the automatic register and winder. 4 Is the electro-mechanical gong. 5 Is the battery of open-circuit cells. 6 Is the battery of closed-circuit cells. 7 Is the relay and relay bell. Instead of, or in addition to, the gong (4), may be used a mechanical tower strike. PROPERTIES OP CONDUCTORS. Pure and Soft Copper. Specific gravity, pure annealed, at 60° F 8.89 lbs. Cubic foot Aveighs 555 lbs. Cubic inch weighs 32 lbs. 1,000 foot 1 inch square rod weighs 3,851 lbs. Tensile strength at 100° per square inch 23,366 lbs. Specific resistance 1 cubic centimeter 0° C 000001594 ohm. Resistance 1 cubic inch 15.5° C. or 60° F 0000006774 ohm. Resistance 1 foot of 1 square inch section 20° C 000008128 ohm. Resistance 1 mil-foot 0° C 9.59 ohms. Weight per mile of copper wire is (dia. in mils)2 Resistance per mile in ohms, of pure copper at 60° F., is 54,892 (dia. in mils)2 Specific conductivity of pure copper is 100, commercial copper runs from 96 to 102 per cent of the standard. Percentage of conductivity is found by measuring the resistance of a sample of the same length and weight as the standard, and at the same tem- perature, then if R = resistance of standard, and r = the resistance of sample, . = per cent conductivity. • * Percentage Conductivity of any Sample. The percafitK Je conductivity of any sample of a conductor, as referred to a standam, carPbe determined as follows : — Let R = resistance of a unit weight and length of the standard, at tempera- ture t, from tables. I = length of wire to be tested, w =z weight of wire to be tested, r = computed resistance of a pure standard copper wire of the same dimensions and temperature as the test sample. r1 = observed resistance at temperature t of the wire under test in ohms. Then as the resistance of a conductor is directly proportional to its length, and inversely proportional to its weight per unit of length (its cross-section), Rl* . r = ohms. By actual test, the resistance of the wire having been found to be r' at tem- perature t, then r1 : r : : 100 : x and the percentage of conductivity of the wire is _100r r' Rise of Resistance witn Temperature. The resistance of conductors is not a linear function of the temperature, and hence its variation with the temperature must, for very precise work, be represented in the ordinary formula : — 140 risp: of resistance. 141 R = r (1 + a t ± b t2) Where R = resistance at the temperature t, r = resistance at 0° C, t = temperature in degrees C, a and b = numerical constants from table below. The following values of the constants have been found, but they are really applicable to the original samples under test only : — Metals (very pure) Mercury German silver (Cu 60- - Zn 26— Ni 14) . . Platinum silver (Pt 67 — Ag 33) . . . . Platinoid (Cu 59 — Zn 25.5 — Ni 14 — W 55) Silver gold .00382 +.00000126 000882 —.000000362 .000443 +.000000152 .00031 " .00021 " 0006999 —.000000062 For ordinary calculations the formula may be written and used as fol- lows : — R = r (1 + at) the values of a being given in the following table : — Metal. a Silver .00377 .00388 Gold .00365 .00390 .00247 .00453 Tin .00365 .00387 .00389 .00354 .00088 .00028 to 00044 The following table gives the value of the principal practical units of resis- tance which existed previous to the establishment of the International Units. UXIT LXTERXATIOKAL OHM. B.A. OHM. Legal ohm 1884. SlEMEKS'S OHM. International ohm B. A. ohm . . . Legal ohm . . . Siemms'sohm . . 1 0.9866 0.9972 0.9407 1.0136 1. 1.0107 0.9535 1.0028 0.9894 1. 0.9434 1.0630 1.0488 1.0600 1. Thus to reduce British Association ohms to international ohms we divide by 1.0136, or multiply by 0.9866 ; and to reduce legal ohms to international ohms we divide by 1.0028, or multiply by 0.9972, etc. 142 PROPERTIES OF CONDUCTORS. HARD-DRAWIV COPPER TELEGRAPH WIRE. (J. A. Roebling's Sons Co.) Furnished in half-mile coils, either bare or insulated. Approximate SizeB. &S. Gauge. Resistance in Ohms per Mile. Breaking Strength. Weight per Mile. Size of E. B. B. Iron Wire equal to Copper. 9 4.30 625 209 2 t? 10 5.40 525 166 3 g 11 6.90 420 131 4 I 12 8.70 330 104 6 $. 13 10.90 270 83 6§3 14 13.70 213 66 Is 15 17.40 170 52 16 22.10 130 41 10 | In handling this wire the greatest care should be observed to avoid kinks, binds, scratches, or cuts. Joints should be made only with Mclntire Con- nectors. On account of its conductivity being about five times that of Ex. B. B. Iron Wire, and its breaking strength over three times its weight per mile, copper may be used of which the section is smaller and the weigbt less than an equivalent iron wire, allowing a greater number of wires to be strung on the poles. Besides this advantage, the reduction of section materially decreases the electrostatic capacity, while its non-magnetic character lessens the self-in- duction of the line, both of which features tend to increase the possible speed of signalling in telegraphing, and to give greater clearness of enuncia- tion over telephone lines, especially those of great length. IE.4I»-E]|TCASED AKTI.IIBITCTI©1¥ TELEPHOUE A\ll> TELEGRAPH CABLE!. (Roebling's.) Plaix Cables, Lead For Metallic For Telegraph Encased. Circuit. Circuits. No. of Size Wire No. of Size Wire No. of Size Wire Wires. B.&S. Gauge. Pairs. B.&S. Gauge. Wires. B. &S. Gauge. 4 IS 5 18 3 14 7 18 15 18 4 14 10 18 25 18 7 14 50 18 50 18 10 14 100 18 75 18 20 50 100 14 14 14 ■■^^■^^■^H COPPER WIRE TABLE. 143 d oseoTjjoeooiooq o OOONOOt-OOMCO-tOOrHO (N OS "* © 1* r-i OS i— © X CO tll-COXf-MKMOf O © 01 L.0 O LO iq ;:::m':---:i -o_ © © :jq»oa^co(NiNrt S © ©' ::' ©' x ©* i~ -? yi oi - f r-Tr-Ti-T .a o Q Ph ®30ioNOiOMrt w OOOWLOOOOMNOOt>(M^Mt't-drt»l> © i-h © 01 " to i" -T — '- © 01 co © — © -v lo X oi I- 7i i - i -y r~ 7 j t - i.q © x o i x -i_ — _ x t - i--. -f :o oi oi i-i oc -)T t-T cT t-^ loT •*" cd" of of r-T i—" t-T r-1 ® d CO l>- rH CO LO ** CO ooooMo-*riNLOHioaP3Lodd6^fflt-.' ■jihiOsx- x CO © co — © © © © : t © © r- -t © -f 01 -C ",— , -TO ©, — __'-0_ © l- .71 © L-© LOCO MNrt (N © Cl3fi"ocO «D lO TP"« 3 £ d Ol Ol -* 1Q O o (oootOHt-mooct-o tmt:N«ni;-i;oqncicioc3 O b- :l-.0 a ® © os •# ■* co co id LidddrHTco^HN ■* oi i-i r-i : i -.; "7 — : " i " 7 > .7 /. / - L- -f 71 M M t^qqqx^rit- i_. •_ r ©©©©©©©©©©©©©©©©©© 0- q <5 o M 0 ~ ©©i-!Oico-PLO©t^ao©©i-ioico-*fL.o©c-- 1©© ^^^^^^^^ H D\AiV s^'a 144 PROPERTIES OF CONDUCTORS. (SOt-OOH(SOt-« co o> 1-1 ** ?j .:; q oi ; -: -r r r. — © o lo -* o tfflOIMfflt-W'JMMNHHHOO OiOWM; cq CO LO in © © © co oi t- © iq to C-; O; © OI Ol © -r © t~ H © CO 10 © CO 01 © © i -p © m' o" ' - - n r: - Q r- & cq © qi ~ - -ffl^WMINHHH n ■# Lq q ■* t- OiMiOCl too -- iq L~ Ol 1 t- LO © X L- i)< 00 01 © 01 © -t © -t< -i © © © © to © 0 i X 0 I -i -i< © © 0 1 10 © lO 01 © t- © LO ■_ 0 I © X © LO -f CO 01 Ol n ri ri oo © © in m co io n i.o © © © © m t- © © in -+i © t -. : i i o f. - — i.o © to -p CO CO © -P CO 10 ^< © © 01 © 01 © © © © to © © -f ~ 10 — © in h x jq oi © i - ro -r o_i — ~ r i- r- © io io — to to to © © © © © ©©©©©©©© © c ©©©©©© ©5 ©"_©_©_©' ©©©"©©'©_©"©©©'©'©'©©*©'©'©'©© CO © © n 01 -o -1-10 © l- X © © i 0 1 to -f LO © I- X © © HHMCNOKIC) 010 1 Old 0 1 CO to to CO tO CO tO CO CO CO -# w q ^ QrZ co ^ 3 m S rt co m-S"^03 B 9 iS § Od ©■ ©??Si © -^02 1 ^! § *$%* 2 6-2 S "21 0 'g©^'J Is P ; ^ „• r3 © *43 o © ^ 2 ~v •" — ^ ■ "? ia -^-5 © i 3) i bD ^ > - '; •Io-?© ft S^g.5 -g 3 i •S-3"-SS 2 -°-=§^ ©- S ' an ra to cs *-• Cv>s4H'a N© i ^•|2ii S J^S.S' &= ^ -selts ■"- §J*s 2§o ^ ^•^^ +3 s, c^ o 73 CO git, the com ^presenting •< W. G. wires st fourth sig as follows : Resistance 1 per = 0.1469 B 14365 B. A. U. 141729 interna and 80° C, 1. significant di half a unit, r B. &S. or A. ch, the neare ^©&qH '43 cpoS -SSINIP SS^in © II j5^ „ S «P 5^111 CO fto.73 d c« fl 3n com of cop 3f hard or 20° ried to correc diamet No. 3( 05 T3 "1 © 2 J3 © 0 ^'tSgOOcSt^cj ^bV2 Jg^s^-© ^'S ©«^ -silSS-S &* S^^ . |j] 1^ I 11-3 *» a i^l""Hiiflilll S«„ Cv_„cDt-lrt^H-g

o to hot- © CO CO —i -+ -t- © i- ~. x -h —i ci m io © © h c-l ro i- x c- — . -t- i- x © c-1 •& co © © © C © - — : i :-. i - x -* O O O O O n iH Cl o . j a § c-o 5 H g g ^ O N » H h a i- -t :i ■; i-i i-o -t co ci i-i ^ o oo o ci t- lo i- S l-~ L; -' ^ ° >~ 9 p — — ?-' v oooooopooqoo q "q q o o d *h i-t = = O — i-i — •• O O O O O O ( ) O O O O O c lO l> OJ rj doooo'oooo'odddi-H cocinciciLOrfiao x. o ci i- o l: i- n -f co ci ci n h lo co s o lo c-. lo i-i :o o co o '-1 i- — co oo co i- c: i- t- o r, a o lo -f ti ci lo •# .1 ii ci co -c lo co x o oi o co i-o oi ~ ii -r — co - c: co o oo -r ci oo o t- o o o o o o o o ti ii ii ci ci co -r lo oo x c oi co c oo ci h ci lo co -h oooooooooqoqopoopOiiniicioico-c*Loqcop ■ n Cl =t< O I 00 CO i lo lo to oi -t w a d dddoo n«'*»6o co cr. oit«t-t-oooi(NHco Ci-*C5i-lOt--#Tt* 3 ^ ^ - X CO | 2 r. ?! © " © o -* co eo U5 ci © eo g d ©' in X 01 :o © so •& X s © U0 CO UO ■* - o © eo lO ■* © -c CO ** rH" X X _ X -f lO 05 X X co -r 05 © -+ q CO so q os X IO is 8 d -# o 01 X X so m -' ^" i S3 s g co 3 co eo CO ^ so <* eo so q in © © LO m -J coo co X 01 01 © © * se X © CN SO £ CN o s r^ IN 8 rt 1-1 r"1 " * c 1 i0 © to -J CI ~ © © © 2 © © 2 eo 2 so bO CO 3 © 3 q © CO © © q © 1 --2 CO r' 9 ■* oi d d d d d d d d d d d d © © © © © © ^ X © <# t- X m © © ■y. ^ _ X m r* | ts r\ Zj! ""£ Lr 5 5J £_! CO 2 x- 2 2 © © 1 © S? 00 X ■_* ?! S 3 ?: ~ ^ 5 s 2 © © 1 5 2 § © © © 3 5 3 CO " in d d d © d d © © d © d © © ©' © X © X •-CO © ^ * x~ o-f h h th ssss^ qqqqqqqqq o©o ©o©©© o©©©p©o©©©©o©©©©©©©©x©©©© oi © eh x i- X >- *(5 ^h co co oi ft ci ci h h h « -; h o ~ -i a § S3 2 ? n 2 s s 0 Sci^^cicieocJ OOOOHHKisq ©©'©©©©©©''-- SS3 © © CO — 3 S H ?; w i s 1 ft ft 3 3,000 12.5 54.78 .009084 2.597 5 5,000 18.3 70.72 .015139 7.214 8 8,000 26.0 89.55 .024220 18.464 12 12,000 35.2 109.55 .036328 41.538 15 15,000 41.6 122.48 .045410 64.9.92 20 20,000 51.6 141.43 .060548 115.372 25 25,000 61.0 158.12 .075682 180.278 30 30,000 70.0 173.21 .090817 259.722 35 35,000 73.6 187.09 .105955 353.340 40 40,000 86.8 200.00 .121082 461.440 45 45,000 94.9 212.14 .136227 584.098 50 50,000 102.7 223.61 .151357 721.026 55 55,000 110.3 234.53 .166501 872.547 60 60,000 117.7 244.95 .181625 1,038.258 65 65,000 125.0 254.98 .196772 1,218.5S6 70 70,000 132.1 264.58 .211901 1,413.264 75 75,000 139.1 273.87 .227043 1,622.457 80 80,000 146.0 282.85 .242176 1,845.952 85 85,000 152.8 292.55 .257303 2,083.759 90 90,000 159.5 300.00 .272434 2,336.405 95 95,000 165.1 308.23 .287587 2,603.046 100 100,000 172.6 316.23 .302709 2,884.082 110 110,000 185.4 331.67 .332991 3,489.958 120 120,000 198.0 346.42 .363267 4,153.433 130 130,000 210.2 360.56 .393527 4,874.226 140 140,000 222.2 374.17 .423797 5,652.899 150 150,000 234.0 387.30 .454061 6,484.573 160 160,000 245.6 400.00 .484328 7,383.042 170 170,000 257.0 412.32 .514622 8,835.525 180 180,000 268.3 424.27 .544884 9,344.686 190 190,000 279.4 435.89 .575140 10,411.241 200 200,000 290.4 447.22 .605427 11,536.681 220 220,000 312.0 469.05 .665975 13,959.567 240 240,000 333.0 489.90 .726498 16,612.114 260 260,000 353.5 509.91 .787058 19,496.997 280 280,000 373.7 529.16 .847605 22,612.233 300 300,000 393.6 547.73 .908140 25,957.464 320 320,000 413.1 565.69 .968672 29,533.696 340 340,000 432.3 583.10 1.029214 33,340.181 360 360,000 4 51.3 600.00 1.089738 37,376.652 1 Mil Foot = 9.718 B. A. Units @ 0° C. (Dr. Matthiessen.) 152 PROPERTIES OF CONDUCTORS. TABLE OF MHUEWSTOarS, WEIGHT, AKD RESIS- TANCE OF PURE COPPER WIRE — Continued. (Edison or Circular Mil Craug-e.) Length. Resistance. Legal ohms at 75° Fahr. 110.087 66.054 41.288 27.527 22.022 16.516 13.213 11.011 9.4381 8.2589 7.3407 6.6069 6.0060 5.5059 5.0820 4.7192 4.4044 4.1292 3.8865 3.6706 3.4773 3.3035 3.0031 2.7528 2.5411 2.3596 2.2023 2.0647 1.9432 1.8353 1.7387 1.6517 1.5016 1.3765 1.2706 1.1798 1.1012 1.0323 .9716 .9177 285.9 476.5 762.3 1,143.4 1,429.2 1,905.7 2,382.0 2,859.9 3,334.9 3,811.0 4,287.7 4,763.8 5,240.5 5,716.5 6,192.9 6,669.4 7,146.0 7,622.3 8,098.4 8,574.7 9,C51.6 9,527.6 10,480.6 11,433.6 12,386.0 13,338.7 14,291.3 15,243.9 16,197.4 17,149.9 18,102.1 19,055.4 20,961.1 22,866.0 24,772.1 26,677.8 28,583.1 30,488.3 32,393.8 34,298.7 .0651602 .0240743 .0154178 .0055470 .0038522 .0028301 .0021671 .0017120 .0013868 .0011467 .00096315 .00082057 .00070758 .00061635 .00054172 .00047990 .00042807 .00038415 .00034673 .00028656 .00024070 .00020514 .00017690 .00015409 •00013544 .00011995 .00010701 .00009604 .00008667 .00007163 .00006019 .00005129 .00004422 .00003852 .00003386 .00002099 .00002675 .003497600 3 .002098640 5 .001311780 8 .000874578 12 .000699663 15 .000524745 20 .000419807 25 .000349840 30 .000299863 35 .000262400 40 .000233227 45 .000209914 50 .000190821 55 .000174931 60 .000161465 65 .000149937 70 .009139938 75 .000131193 80 .000123480 85 .000116622 90 .000110477 95 .000104960 100 .000095410 110 .000084460 120 .000080730 130 .000074970 140 .000069997 150 .000065600 160 .000061735 170 .000058309 180 .000055242 190 .000052478 200 .000047707 220 .000043733 240 .000040368 260 .000037484 280 .000034986 300 .000032799 320 .000030870 1 340 .000029155 360 1 Mil Foot = 9.718 B. A. Units @ 0° C. (Dr. Matthiessen.) CAPACITY OF COPPER WIRES. 153 SAFE CARRYING CAPACITY OF COPPER WIRE§. Below will be found the formulae of Forbes and Kennelly for safe carrying capacity of copper conductors. The results, which would be obtained by using these formulae, have been somewhat modified in practice, and the reader is referred to the tables in the "National Code" for capacities recommended by the underwriters. Size of Conductors. (Prof. G. Forbes.) Bare Overhead Wires. — The relation between the diameter of a conductor and the current it can safely carry without over-heating is 2 H 1— DH- Ui X .24 "Where 1= Current in amperes. D — Diameter of wire in centimeters. t z= Excess of temperature C. of wire over the air. H=z Coefficient of radiation and convection = .0003. R = Specific electrical resistance of material per b. cm. at the lim- iting temp. .24 = Calories in a Joule. Insulated Overhead Wires. — For gutta-percha and india-rubber insulation, L V t V tL. \ 1= k I \ .48 It X * X 10 + 3Z>3 log. e D1 ) Where Dx = Diameter of conductor. IX = Diameter of insulated cable. t = Excess of temperature of conductor over air. k = Heat conductivity of insulator ; for G.-P. = .00048 ; for I.-R. = .00041. Kennelly's Mule of the Sate Diameter of an Insulated Panelled Wire. If the limiting safe diameter of an insulated panelled wire be such that twice the proposed full load upon it shall only raise its temperature 40° C, then the best formula is d = .0147 1\ d being in inches and I in amperes ; or approximately f/ = 70 Heating of Bare Conductors S»y a Current. The temperature to which a bare copper wire freely suspended in still air will be raised when traversed by a current is approximately "tfs X 90,000 + t°, T° = temperature of wire in F°. t° = temperature of air in F°. 1 = current in amperes. d = diameter of wire in mils. For a given presumable maximum elevation of temperature the requisite diameter is approximately 154 PROPERTIES OF CONDUCTORS. 1 11 OX WIRE. Iron. Specific gravity . 7.7 Cubic foot weighs 480 lbs. Cubic inch weighs 2779 lb. Tensile strength per square inch ...... 50,000 to 60.000 lbs. Specific resistance 1 cubic centimeter at 0° C. . .0000005 ohms. Resistance per mil foot 58 ohms. Steel. Specific gravity . 7.932 Cubic foot weighs 490 lbs. Cubic inch weiglis 2834 lb. Tensile strength per square inch 55,000 to 80,000 lbs. Specific resistance 1 cubic centimeter at 0° C. . .000013 ohms. Resistance per mil foot 82 ohms The above items are for the metals as metals, and not when in wire. Re- sistance of iron wire varies so much, by reason of drawing and hardening, that it is not practicable to state specific resistances, weights, and strengths. The following tables give approximate averages. OALYAAIZED 1 HOX WIKffi FOB TEJLBCiltAPH: AXJL* TEiEPHOSE LOES. (Trenton Iron Co.) Weight per HKile-Ohm. — This term is to be understood as distin- guishing the rrs'nt.mce of material only, and means the weight of such material required per mile to give the resistance of one ohm. To ascertain the mileage resistance of any wire, divide the "weight per mile-ohm" by the weight of the wire per mile. Thus in a grade of Extra Best Best, of which the weight per mile-ohm is 5,000, the mileage resistance of No. 6 (weight per mile 525 lbs.) would be about 9§ ohms ; and No. 14 steel wire, 8500 lbs weight per mile-ohm (95 lbs. weight per mile), would show about 69 ohms. Sizes of Wire used in Telegraph and Telephone lines. No. 4. Has not been much used until recently ; is now used on important lines where the multiplex systems are applied. No. 5. Little used in the United States. No. 6. Used for important circuits between cities. No. 8. Medium size for circuits of 400 miles or less. No. 9. For similar locations to No. 8, but on somewhat shorter circuits ; until lately was the size most largely used in this countrv. Nos. 10, 11. For shorter circuits, railway telegraphs, private lines, police and fire alarm lines, etc. No. 12. For telephone lines, police and fire alarm lines, etc. Nos. 13, 14. For telephone lines, and short private lines ; steel wire is used most generally in these sizes. The coating of telegraph wire with zinc as a protection against oxidation is now generally admitted to be the most efficacious method. The grades of line wire are generally known to the trade as " Extra Best Best" (E. B. B.), " Best Best" (B. B.)', and " Steel." " Extra Best Best " is made of the very best iron, as nearly pure as any commercial iron, soft, tough, uniform, and of very high conductivity, its weight per mile-ohm being about 5,000 lbs. The " Best Best" is of iron, showing in mechanical tests almost as good results as the E. B. B., but not quite as soft, and being somewhat lower in conductivity ; weight per mile-ohm about 5.700 lbs. The Trenton " Steel" wire is well suited for telephone or short telegraph lines, and the weight per mile-ohm is about 6,500 lbs. TESTS OF TELEGRAPH WIRES. 155 The following are (approximately) tlie Aveights per mile of various sizes of galvanized telegraph wire, drawn by Trenton Iron Co.'s gauge : Lbs. 720. 610. 525. 450. 375. 310 . 250. 200. 160. 125. TESTS ©E TEIEGRAPH WIRE. The following data are taken from a tab^e given by Mr. Prescott relating to tests of E. B. B. galvanized wire furnished the Western Union Telegraph Co. : Wei ght. Resistance. Temp. 75.8° Fahr. "i ^ ^ 6 % *s - O Ph^ a o z "So^ 3 s 3d I 8 a ID 1 +2 "5 1 £ 3 o fo o 4 .238 1,043.2 886.6 6.00 958 5.51 5 .220 891.3 673.5 7.85 727 7.26 6 .203 758.9 572.2 9.20 618 8.54 3.05 7 .180 596.0 449.9 11.70 578 10.86 3.40 8 .165 501.4 378.1 14.00 409 12.92 3.07 9 .148 403.4 304.2 17.4 328 16.10 3.38 10 .134 330.7 249.4 21.2 269 19.60 3.37 11 .120 265.2 200.0 26.4 216 24.42 2.97 12 .109 218.8 165.0 32.0 179 29.60 3.43 14 .083 126.9 95.7 55.2 104 51.00 3.05 .Joints in Telesrrapn Wires. — The fewer the joints in a line the better. All joints should be carefully made and well soldered over, for a bad joint may cause as much resistance to the electric current as several miles of wire. WEIGHT ANR RESISTANCE OE GALVANIZED IRON WIRE PER TSXMJE. (Roebling.) Gauge. B. &S. Weight per Mile. Resistance. Ohms. Gauge. B.&S. Weight per Mile. Resistance. Ohms. 6 7 8 9 10 550 470 385 330 * 268 10 12.1 14.1 16.4 20 11 12 14 16 216 170 100 62 20 32.7 52.8 91.6 156 PROPERTIES OF CONDUCTORS. SIMS, WEIOHT, I.BH o s |! coal Iron Wire in Pounds. So S| o'rt.5 2 < -^ o ft _bc-3 Bright. Annealed. 00000 .450 .15904 1.863 2.833.248 12,598 9,449 0000 .400 .12566 2.358 2,238.878 9,955 7,466 000 .360 .10179 2.911 1,813.574 8,124 6,091 00 .330 .08553 3.465 1,523.861 6,880 5,160 0 .305 .07306 4.057 1,301.678 5,926 4,445 1 .285 .06379 4.645 1,136.678 5,226 3,920 2 .265 ■ .05515 5.374 982.555 4,570 3,425 3 .245 .04714 6.286 839.942 3,948 2,960 4 .225 .03976 7.454 708.365 3,374 2,530 5 .205 .03301 8.976 588.139 2,839 2,130 6 .190 .02835 10.453 505.084 2,476 1,860 7 .175 .02405 12.322 428.472 2,136 1,600 8 .160 .02011 14.736 358.3008 1,813 1,360 9 .145 .01651 17.950 294.1488 1,507 1,130 10 .130 .01327 22.333 236.4384 1,233 925 11 .1175 .01084 27.340 193.1424 1,010 758 12 .105 .00866 34.219 154.2816 810 607 13 .0925 .00672 44.092 119.7504 631 473 14 .080 .00503 58.916 89.6016 474 356 15 .070 .00385 76.984 68.5872 372 280 16 .061 .00292 101.488 52.0080 292 220 17 .0525 .00216 137.174 38.4912 222 165 18 .045 .00159 186.335 28.3378 169 127 19 .040 .0012566 235.084 22.3872 137 103 20 .035 .0009621 308.079 17.1389 107 80 .031 .028 .0007547 .0006157 392.772 481.234 13.4429 10.9718 22 0 "* 3D CO 6 23 .025 .0004909 603.863 8.7437 $?§§ | $' Oj - 24 .0225 .0003976 745.710 7.0805 fslllof* 25 .020 .0003142 943.396 5.5968 26 .018 .0002545 1,164.689 4.5334 27 .017 .0002270 1,305.670 4.0439 III! f-J =11 28 .016 .0002011 1,476.869 3.5819 29 .015 .0001767 1,676.989 3.1485 *JI^lo'S'dIB'"' fl 30 .014 .0001539 1,925.321 2.7424 §il^3fc|i I 31 .013 .0001327 2,232.653 2.3649 32 .012 .0001131 2,620.607 2.0148 ga><*H^-~~ga>_ 3^ 4,107 l 14 .06408 .0953 7 9 11 56.9 $60.00 9,016 7 19 .10767 .1389 10 12 14 103 110.00 11,368 7 18 .12090 .1522 10 12 14 10S.5 110.00 14,336 7 17 .13578 .1670 10 12 14 115.5 110.00 18,081 7 16 .15225 .1837 11 13 15 140 130.00 22.799 7 15 .17121 .2025 12 14 16 165| 150.00 30,856 19 18 .20150 .2328 12 14 16 184 165.00 33,912 19 17 .22630 .2576 13 15 17 218 190.00 49,077 19 16 .25410 .2854 14 16 18 260| 210.00 60,0S8 37 18 .28210 .3134 15 17 19 314 260.00 75,776 37 17 .31682 .3481 16 18 20 371 290.00 99,064 61 18 .36270 .3940 18 20 22 463 385.00 124,928 61 17 .40734 .4386 19 21 23 557 415.00 157,563 61 16 .45738 .4S85 20 22 24 647 460.00 198,677 61 15 .51363 .5449 22 24 26 794 535.00 250,527 61 14 .57672 .6080 24 26 28 970 615.00 296,387 91 15 .62777 .6590 26 28 30 1,138 750.00 373,737 91 14 .70488 .7361 29 31 33 1,420 900.00 413,639 127 15 .74191 .7732 30 32 34 1,553 1,000.00 160 PROPERTIES OF CONDUCTORS. IPECIAI CAHLEft FOR §TR£ET-€AR WIRIHT6. Car wiring cables have a wrapping between the wire and rubber to facili- tate stripping for soldering. The 7-14 single braid is adapted for ordinary car wiring for two 25 h.p. motors. The triple braid is recommended for taps to motors, as it will stand abrasion and is more durable than rubber tubing. The 75-25 braided to .500" diameter is standard for field leads of the GE-800 motor, and fits the -rubber bushings in the motor frame. The 49-22 braided to .025" diameter is standard for armature leads of the GE-800, and for all leads of the GE-1000 motors. These cables are also well adapted for leads for suspending arc lamps. (General Electric Company.) I a P u p'3 %u List price. =3^ 1"? a> j uopBinsuj •eaqouj •pgpreig; la^ureid •suiq;o§8j\[ ■QIIH •paprejg; ja^ureiQ •surqo§8j\[ ■»nw jod aoura^sis •saqoui •pspreag •surqo§8j\[ ■onre J9d ooire^sis -a^j noi^'Binsiri •seqonj •poprejg •sumoSaj\[ jod eorre^sis -9^ nop'Bxug'ni oooot-cot~oc5'#co'#o ooooooooooooo ooooooooooooo w 10 uo 10 10 oioooinoei O X OS l-j co eg 000000 -11014 JOJ lA-B9q 00 j; OO^NOOOOlO^MNrtO 162 PROPERTIES OF CONDUCTORS. 222S2S2Soooooooooooooooooo<=>=> - = - ~ ^ - ~ - - ~ ~ _ — « = = ^« ' v <~o_ lo co :o c-. co co co co ~i ol oi ?i el 5 o o o o i- S o NNNrlrtHHHH t-T t-Tm r-Tr-TT-TTH iH l-Tr-Tr-Tr* t-T r-Tr-Tr-T ^^icirfjo»d«o t~ l- gc o o io c m !- ol CO '4 ^ 2? g £2 CO o SSiSSissnn liiiia§3883888888gSg88S8888§ CO - fie 3 — 0 CO in -+ o -f 10 ICO 572 625 672 722 771 875 945 988 048 133 §::::::: HH 1-1 000000000000000000000 nolens o o c o c o o o o io ico ci-i-.ooooooo nl sa rAT ci ci ci ci c : c c c t- t- n i- l- L-o lo uo ooioo L1 .l // e •100Ci O^Nia © HHrt"N 163 s ■M 000' I jad eoud ^sn •WOOO'l J8 •u OOO'I jad aoiad ^srq; '« 000' I -T95 qj rOHOMocoqjj^t-riO CS COOl t^t^t^tOirOi-icOicOcoaO'*'* Sr-l?Hr-ll-n-l © id ci o^ t^ t^ s^i Tt<-#0©t-COi-liN-*OCO.-llO OS ffllON HHHNINMK oaociiot^T- o-* eo«ot-«o t^co"oc4co-*mi-ic-5i-icot-o NW-*K3OON<0fflMI>C0H HrlriCJCNCO'* oooooo- qioiqqqq h h c) « co io CO CM i-H OS 00 ^# os so ci t-J t-" o m id t^ lq irj o t^ l- >q 'i e 6 in d d 6 -©IHIOC* ©rtC0ON^00H«0?)iHOt0l00DI>C0OT)iO(N »<*010»0f - - - O idc5oco©©©©idid©©©c5©©c5 HNMiot'C ■! ::.::".-....:..::"-::■.: ■QCO'#-TlW©©l.'~00CS CO OS 05 Ifl ©s 9g?2l5P!202S$2S,^SS^ciS?,-|ocoi-i«5coiM-*©©r>oom i-li-lr-lT-l JS CO © © «Ot^THiH00idl>00©ci»d©© coco-*int-©©OT*ic-©-i-- HHtHiHNIMIMCO'* SNiOHHte r)5 oi oo id © as c-i in -;'.■.- i ".: • i ■■ ~ • ' .:■■.■■:■■• \ HHrtNC-lCOM'tlOa © © © in © CN COCOf © cicS^ ©^^©oo©io^co^TH©©©©©©©©ogg©o©©o© © © © © © c: © ©. — .. ©©_©_©© <= "'ooooooooooqoo 10©10©0©10©©©10©© c^coco^io©t-co©©i>) in ©_ SPECIAL FINISHES. 165 Below is a table of prices at wliicli special finishes for any of the fore- going wires and cables can be furnished. C. L. Plain lead cover over the rubber. C. L. A. Lead cover with jute and asphalt over the lead. C. L. A. I. Lead cover, jute and asphalt and band iron armored. To obtain the price of the cable desired, add to the list price of the rubber covered cable braided, the list price of the finish desired for the diameter nearest to that of the braided cable. A cable having a lead cover, jute and asphalt over the lead, and wire armored (C. L. A. W.), in addition to the above special finishes can also be furnished. Prices on application. To obtain approximate weight of cable having special finish, add to the weight of the cable the weight of the special finish as given below. SPECIAL FO'ISHES, (General Electric Company.) Diameter C Approx. Weight per 1000 feet. . L. List price per 1000 feet. C. L. A. C. L . A. I. of Braided Cable. Inches. Approx. Weight per 1000 feet. List price per 1000 feet. Approx. Weight per 1000 feet. List price per 1000 feet. .200 157 $30.00 252 $60.00 , . . .225 170 31.50 268 62.50 . .250 191 34.00 297 66.50 .275 214 37.00 327 70.50 .300 227 38.50 345 73.00 .325 345 53.00 475 89.50 .350 376 57.00 514 94.50 1,131 $193.50 .375 391 59.00 534 97.00 1,162 197.50 .400 424 63.00 574 102.00 1,229 206.50 .425 438 65.00 590 105.50 1,254 212.00 .450 473 69.00 634 111.50 1,325 222.00 .475 498 72.50 665 115.00 1,370 227.50 .500 519 75.00 691 117,00 1,417 230.00 .550 567 79.00 751 125.00 1,506 241.50 .600 620 85.50 816 133.00 1,616 255.50 .650 656 90.00 864 139.00 1,901 294.00 .700 1,118 144.50 1,352 199.00 2,498 369.00 .750 1,194 153.00 1,442 209.50 2,632 384.50 .800 1,194 153.00 1,442 209.50 2,632 384.50 166 PROPERTIES OF CONDUCTORS. SJPECIAE FlUfliHES- Cow^wwerf. Diameter C . L. c. L. A. C. L . A. I. of Braided Cable. Inches. Approx. Weight per 1000 feet. List price per 1000 feet. Approx. Weight per 1000 feet. List price per 1000 feet. Approx. Weight per 1000 feet. List price per 1000 feet. .850 1,258 160.50 1,516 218.00 2,742 398.50 .900 1,317 167.00 1,583 226.50 2,847 411.50 .950 1,423 179.50 1,707 241.50 3,022 433.50 1.000 1,482 186.50 1,773 249.00 3,132 447.00 1.05 1,556 190.00 1,859 257.50 3,263 461.00 1.1 1,631 201.00 1,946 267.50 3,397 477.00 1.15 1,705 210.00 2,030 277.50 3,820 533.50 1.2 1,795 220.00 2,131 291.50 3,987 559.00 1.25 1,854 225.50 2,201 298.50 4,098 572.50 1.3 1,959 237.50 2,322 313.00 4,294 595.50 1.35 2,018 240.00 2,393 317.50 4,409 607.00 1.4 2,851 330.00 3,257 415.00 5,419 724.00 1.45 2,989 348.00 3,410 432.50 5,639 750.50 1.5 3,008 350.00 3,432 434.50 5,681 755.00 1.6 3,362 378.00 3,717 470.00 6,097 810.00 1.7 3,400 392.50 3,872 488.00 6,335 827.50 1.8 3,615 416.50 4,113 515.50 6,694 882.00 1.9 3,792 436.00 4,309 538.00 6,987 905.50 2 3,988 457.50 4,529 563.00 7,315 945.00 In leading cables a tape is used over the rubber in place of the regular braid. For thickness of lead used on above finishes, see page h21. If other thick- nesses than these are desired, special prices will be quoted upon application. PAPER OilLATED AAI» LEADED WJTtES AND CABLEI. There will be found on the following pages data and prices of a full line of paper insulated and lead covered wires and cables. All cables insulated with the fibrous covering depend for their successful operation and mainte- nance upon tbe exclusion of moisture by the lead sheath; and this fact should constantly be borne in mind in handling this class of cables, conse- quently the lead on these cables is ext^a heavy. The use of jute and asphalt covering over the lead is strongly recommended on all this class of cables, inasmuch as the life of the cable is absolutely dependent upon that of the lead. Paper insulated cables cannot be furnished without the lead covering. WIRES AND CABLES. ijfBqds y pire e^ap .19 ao •saqoui •pi39T[ J9A0 .I919UTeiQ "S8t[0UJ *P138T[ ss9U5[0iqx ■smqoSai\[ 90U'B^SIS9JI uoi^upisui tH tJH eo t- cm li ' - ~ :-.'..• - <~. - — - ~ i - — '-•" '2- Z- :~ t '-■ ao)00o>oOr-i-HHrt:iM-ti-tL;c3 vr i-c;q?i co iq ' t-I r-5 " ' ri ri H H rH rH rt rt H rt ri H TH tH ri iM' IN N N C -i< - -1 I- OOOrtHrtf ^«H5^«H£»h«He oooooooooooooooooooooooog o o g o o — . ■ . '','-? — '■ ~ Ht-niOH(Ortin*3'-sooa«iNHrjiffi't'*HO'Offl iOi>t-HS- : r. ~. M t- go cn ooooaooocoooOHHH?iM::t li o^ccoooi-iNiq " ^ ' ' 'rtHrHriHHHrHHr-iH^rHHoiNji« CHOHMO-fClM«^TPNt-NOOafU-?l?IC5CC-tff K S 'O O T< M I- :1 'J ~ 1 1 I- X CI X M -iT.XTjrillt-H Li Li O t- LI Li -O L~ I- 1- GO X Ci O O ^h T-i "M II CI Li t" CO OS N •S9tpni ^"tfqdsy puu 9;np J9A0 ■i9;9ttreig "S9r[0UI PB9T J9AQ .I9^9ttret(J •sgqoui ■p^l ss9tnpii|x •saiqo§9i\[ ^^«K^H2"£o£«S^~^:3 oooooooooooooooooogoooooo oooiooc' jiooooooioog LI Li 11 11 LI LI ". l i OiJ^IJIClClMHHO^lOU •sgqoui 9^nf J9AQ .T9}9UrBI(I l"CJM.H'3"tl>Nt-fflOClNSflflO«0^>* i mxht t- ii ci x ct go -h — . -h :c ci -oo^to O Cf.CCOH 11 11 CI CI -H -H Li cscihnt? •S9qOUJ •p^l J9A0 J9^9Xni?T(I OflHt-Onir •S9qouj •pi39rj SS9U2[0pqX £««^5««B«-* •smqo§9j\[ •9HM -^J 90U'BJSIS9'JJ CM be® Is =3. ooosri •188^ 0001 9syeunxo.iddy oooooc ~ >- c lo = 5ciSoqqcqq8oq82 22S°3:s>St5; ~Sol « S -3 P Lo - f^ -¥ - ? 3 £ r :" -■ '--■ •* ^ omwioncb oo t> co -* x os co .q so l~_ x 3 £-__ os ol -* x 5 co. x £ co co lo co rtrt t-i n rl i-l llcTof ci CO Co'cO •* ■* rti~«Tt-^(X cTr-T i s^O^ 0001 -99.i oooi jod ^qSiaAV 8syeunxo.iddy -h so co CO X L- i-i ~ Ct-l-OONO-f coos r' O t- rtWO L- i-^CO -# iO (O I - -T CC O^ — I •+ L- — LO L- <5M«MT)aiH ^9£ 0001 J9o — CO l- C -f 01 10 CO n CO X -* t- t- hcon x c-j -t; ioc'cz '-I1-. — ~l—,r-/irt ^ci.T^'~l0i,'l r-Ti-T r-^-H rH i-Tilil of efcO CO CO CO •# -#~LO (> CO Gfi-T 199jI 000T -9J 9DU«I "SII ~ t£~~ -------- - ~ ~ ~ ~ ~ ~ - — — © c © o qqqqqqqqqqqqqiaqqqiqqqqqqS >q © .1 CO i-i ri -,-' x — - co' so — ' ::' o ! x' i - co' oi x OS oi -h CO i-i o-i co" - ". CO CS -f t- l> CN CO CO i-i CO -+ os co 10 I - co -+ -V o. It 01 ~ HHHNHHi-HNflCKCCOt-tiO-l- rl rt N N CO •?90£ ooox jsd ?q;§i8AV e~«raixo.iddy qncjOOONonaHrtMcnoo-foON-coffl 0 OS LO 1-0 O OS OS t- tt»H-t I- CO O! X OS 0 I SO so X X 1 LO CD © Ol & CO ># KC] 1.0 L- (OS CM L0 1> CT. lOtDOl-CC -T r-Tii i-Tr-TiTii oi olci ei' co co co ■* lo co i^coo W*£ oooi -^a 8^88 o o o o c o i" ~ -o "c — ~ c c ~ ;ooooo lo o o lo q o o —_ o c lo_ © ic uq uq qoioqaq " O (s^ oi i-O iO o i i - t- i.o v: lo \~ io '-:' t! -'■ -- — i c i - 2 ' — ) O 58888888888 ooooooooooo "" 9 S 9 2 2" o" o" o" o" o" cci-.coi- — io — r; CM CO CO ■* LO t- O OI LO o telephone cables. 169 TELEPHONE CABLEi. (By John A. Roebling's Son's Co.) Lead-encased for "Undei'g'rouiid or Aerial Use. The insulation of these cables is dry paper. The company manufac- tures several styles of 19 B. & S. G., 20 B. & S., G., and 22 B. & S. G., ac- cording to the use for which they are intended. The most common size is 19 B. & S. G. They also supply terminals and hangers. Specifications for Telennone Cafoles. l. Conductors. Each conductor shall be .03589 inches in diameter (19 B. & S. G.), and have a conductivity of 98 per cent, of that of pure soft copper. 2. Coke. The conductor shall be insulated, twisted in pairs the length of the twist not to exceed three inches, and formed into a core arranged in reverse layers. 3. Sheath. The core shall be enclosed in a pipe composed of lead and tin, the amount of the tin shall be not less than 2^ per cent. The pipe shall be formed around the core, and shall be free from holes or other defects, and of uni- form thickness and composition. 4. Electrostatic Capacity. The average electrostatic capacity shall not exceed .080 of a microfarad per mile, each wire being measured against all the rest, and a sheath grounded ; the electrostatic capacity of any wires so measured shall not exceed .085 of a microfarad per mile. 5. Insulation Resistance. Each Avire shall show an insulation of not less than 500 megohms per mile, at 60° F., when laid, spliced, and connected to terminal ready for use ; each wire being measured against all the rest and sheath grounded. 6. Conductor Resistance. Each conductor shall have a resistance of not more than 47 B. A. ohms, at 60° F., for each mile of cable, after the cable is laid, and connected to the terminals. IEIEPHOXE CABLES, By John A. Roebling's Son's Co. Number pairs. Outside diameters. Inches. Weights 1000 feet. Pounds. 1 & 214 2 I 302 3 515 4 629 5 S 747 6 U 877 7 B 912 10 U 1,214 12 if 1,375 15 l l,5o6 170 PROPERTIES OF CONDUCTORS. TELEPHONE CABIES-OMrfiiiiKd. Number Pairs. Outside Diameters. Inches. Weights 1000 feet. Pounds. 18 It1* 1,758 20 H 1,940 25 h% 2,332 30 h7E 2,748 35 H 2,985 40 h\ 3,176 45 if 3,365 50 if 3,678 55 ill 3,867 60 if 4,055 65 lit 4,241 70 2 4,430 80 2» 4,804 90 2i 5,180 100 2§ 5,505 TELEGRAPH CABLEi. By John A. Roebling's Son's Co. Eead-enca«ed for Underground "Use. These cables are made of either rubber, cotton, or paper insulation. The sizes and weights are approximately correct for rubber and cotton insula- tion. Both sizes and weights are slightly reduced for paper insulation. In all cases the cables are lead-encased. Specifications for Teleg-rapn Cables. 1. Conductors. Each conductor shall be .064 inches in diameter (14 B. & S. G.), and have a conductivity of 98 per cent of that of pure copper. 2. Coke. The conductors shall be insulated to 3% with cotton, and formed into a core arranged in reverse layers. This core shall be dried, and saturated with approved insulating compound. 3. Sheath. ' The core shall be enclosed in a pipe composed of lead and tin. The amount of tin shall not be less than 2.9 per cent. The pipe shall be formed around the core, and shall be free from holes or other defects, and of uni- form thickness and composition. 4. Insulation Besistance. The wire shall show an insulation of not less than 300 megohms per mile, at 60° P., when laid, spliced, and connected to terminals ready for use, each wire being measured against all the rest and the sbeath grounded. 5. Conductor Resistance. Each conductor shall have a resistance of not more than 28 International ohms, at 60° F., for each mile of cable, after the cable is laid, and connected up to the terminals. TKLKGRAPH CABLES. 171 TELEGRAPH CABLES. By John A. Roebling's Son's Co. 14 B. & S. G. Insulated to 362. 16 B. &S.C. Insulated to g52. 18 B. & S. G. Insulated to &. S is SB d«« 2 &2 20 li 714 Ji 593 §1 ' 452 25 If 863 1t3s 708 11 541 30 1& 1,008 H 824 l 633 35 H 1,147 ll55 938 iiV 723 40 1» 1,268 If 1,053 H 813 45 If 1,431 U 1,182 h% 903 50 If 1,577 IS 1,311 H 994 §UBMA»OE CABLES. By John A. Roebling's Son's Co. £ Armor wires. Total weights. Pounds. uB Outside diameters. & % si *8 Number of wires. Numbers, B. W. G. 1,000 feet. Mile. i 12 8 1,250 6,600 2 15 8 1,722 9,092 3 U 14 6 2,363 12,477 4 1ft 16 6 2,794 14,752 5 1ft 16 6 4 2,968 15,671 6 1| 16 4 3 822 20,180 7 11 16 3 3,972 20,972 10 n 18 5,404 28,533 The core consists of 7 X 22 B. & S. tinned copper wires, insulated with rubber to 385 of an inch, laid up with proper jute bedding. Telegraph cables can be supplied with gutta-percha insulation. This is the best insulation for submarine work, and its reliability and durability more than make up the difference in co6t between it and any other insula- tion. 174 PROPERTIES OF CONDUCTORS. AirMIIfUM. (From paper by Alfred E. Hunt, S. B., and book published by tbe Pitts- burg Reduction Company.) Specific gravity 2.68 Cubic foot weighs, cast 159.6 lbs. Cubic foot weighs, rolled 167.1 " Cubic inch weighs, cast .0924 " Cubic inch weighs, rolled .0967 " Tensile strength in pure soft wire, per square inch . . 26,000 Tensile strength in pure hard-drawn rods, per square inch, 40,000 Conductivity as related to 100% cond. copper: 99|%pure 63.09% 99% pure 62.17% 98% pure 56.17% "Weight per mile of aluminum wire is .004817 (diameter in mils). Aluminum for Electrical Conductors. (From paper by Alfred E. Hunt, S. B.) 1. Any given volume of copper is^= or 3.332 times heavier than an equal volume of aluminum. 2.68 2. The equivalent price of fourteen cents per pound for copper for any length of any equivalent section of aluminum wire or bar would be 14 cents times the factor 3.332, or 46.65 cents per pound. That is, one thousand feet of wire of, say, one-tenth inch diameter, would cost equally as much if bought of copper at 14 cents per pound or aluminum at 46.65 cents per pound. Aluminum, therefore, at 29 cents per pound is only 62% of the cost of copper at 14 cents per pound, section for section. 3. Reckoning the copper conductor to have its maximum of 100 per cent conductivity, and the aluminum to have a conductivity of 63 per cent (which the Pittsburg Reduction Company are ready to guarantee for their special pure aluminum metal for electrical conductors), then for an equivalent electrical conductivity a given section of copper that can be placed at 100 should be increased in area in round numbers to 160 to give an equal con- ductivity. 4. Due to their relative specific gravities, the weight of the given equal length of the aluminum conductor with 160 sectional area will be only forty- eight per cent of the weight of the copper conductor with sectional area of 100, having the same electrical conductivity. 100 y 8.93 = 893, weight of the copper. 160 x 2.68 =428.8, weight of the aluminum. |||-8 =48 per cent. 5. As to their relative cost for electrical conductors of equal conductiv- ity, aluminum at twenty-nine cents per pound is the most economical con- ductor, as compared with copper at fourteen cents per pound. Taking as an illustration, an aluminum conductor to replace a copper wire of No. 10 B. & S. gauge (about one-tenth of an inch diameter), the aluminum wire of equal, in fact somewhat superior, electrical conductivity would be of No. 8 B. & S. gauge ( slightly over one-eighth of an inch diameter). The weight of a mile of No. 10 copper wire is 162.32 pounds ; and its cost at 14 cents per pound would be equal to $22.72. The weight of a mile of No. 8 aluminum wire would be 79.46 pounds, and at twenty-nine cents per pound would cost $23.04. Forty-eight per cent of the weight of No. 10 copper wire, which will give equal electrical conductivity in aluminum wire, would only weigh 77.91 pounds; so that, more accurately, $22.59 would be the cost of a mile of aluminum wire at 29 cents per pound to replace a mile of No. 10 copper wire at 14 cents per pound, costing $22.72. 6. The Continental requirements in tensile strength for soft copper wire, rods, and bars used as electrical conductors is twenty-two kilograms per square millimeter; the English requirement being similarly fourteen tons per square inch; and our American requirement is about its equivalent of 32,000 pounds per square inch. ALUMINUM. 175 08 "53 ?$ « i II o,-< 1 176 PROPERTIES OF CONDUCTORS. 40,000 42,000 44,000 46,000 48,000 50 000 51,000 53,000 55,000 o 8fc P. .4605 .5818 .7325 .9235 1.187 1.468 1.852 2.335 3.084 Comparative weight of given lengths of equal Conduc- tivity, Copper at 100. t^ o 0 < P3 3 S3 £ 33,000 34,000 35,000 36,000 37,000 39,000 40,000 41,000 42,000 Comparative section of equal Conductivity, Copper at 100. ID § o .4288 .5408 .6820 .8600 1.105 1.367 1.724 2.173 2.741 (A a o o CO lO d ft ®5 27,000 27,000 28,000 29.000 30,000 32,000 33,000 35,000 39,000 6 d in 3 Oft .4012 .5058 .6380 .8044 1.034 1.278 1.613 2.033 2.565 is s 204.31 181.94 162.02 144.28 128.49 114.43 101.89 90.74 80.81 .9 I? 3 . a ■*lflSDt>OOOiO'~l 12.8 approximately. Zinc, 1 part ) Gold-Silver — Infer, IpS?} ll:6 approximately. Platinoid — German Silver, with \ p. c. of Tungsten . . . 19.2 approximately. RELATIVE COKRUCTIVITIEi ©E METAL§ AND ALLOTS. (Weiller.) 1. Pure silver 100 2. Pure copper 100 3. Refined and crystallized copper 99.9 4. Telegraphic silicious bronze 98 5. Alloy of copper and silver (50 per cent) 86.65 6. Pure" gold ° • 78 7. Silicide of copper, with 4 per cent of Silicium 75 8. Silicide of copper, with 12 per cent of silicium 54.7 9. Aluminum, 99- « 63.09 10. Tin with 12 per cent of sodium 46.9 11. Telephonic silicious bronze 35 12. Copper with 10 per cent of lead 30 13. Pure zinc 29.9 14. Telephonic phosphor-bronze 29 15. Silicious brass with 25 per cent of zinc 26.49 16. Brass with 35 per cent of zinc 21.5 17. Phosphor tin 17.7 182 PROPERTIES OP CONDUCTORS. 18. 19. 20. Alloy of gold and silver (50 per cent) 16.12 Swedish iron „ 16 Pure Banca tin ,'.'.. 15.45 Antinionial copper 12.7 Aluminum bronze (10 per cent) 12.6 Siemens's steel , 12 Pure platinum 10.6 Copper with 10 per cent of nickel 10.6 Cadmium amalgam (15 per cent) ............ 10.2 Dronier mercurial bronze 10.14 Arsenical copper (10 per cent) 9.1 Pure lead 8.88 Bronze with 20 per cent of tin ............. 8.4 Pure nickel 7.89 Phosphor-bronze with 10 per cent of tin ......... 6.5 Phosphor-copper with 9 per cent of phosphorus 4.9 Antimony 3.88 TEOTPEHATUJUB OF C©ari>XJCXO«S WIIH COBFJFICIFH-TiS. (From Kempe.) Por metals the resistance increases as the temperature increases. The formula which represents the effect of temperature may be written lit = Bo (1 + o>t + (B^2) where lit is the resistance at the final temperature, Ho is the resistance at the standard temperature, t is the increase in temperature, and oo and (£> are coefficients. Por most purposes the following approximate formula may be used : Rt = Jio (1 + oo t). The value of oo for use in the approximate formula is given in the follow- ing table, ooe being the value per centigrade degree, and oo/per Fahrenheit degree. Metal. 00c oo/ Silver 0.00377 0.00210 Copper 0.00388 0.00215 Gold 0.00365 0.00203 Aluminum 0.00390 0.00217 Platinum 0.00247 0.00137 Iron 0.00453 0.00252 Tin 0.00365 0.00203 Lead 0.00385 0.00214 Mercury 0.00088 0.00049 Alloy, 2Pt + l Ag . . 0.00022 to 0.00031 0.00012 to 0.00017 2 Au + 1 Ag . 0.00065 0.00036 8 Pt + 1 Ir . . 0.0013 0.00072 German Silver . . . 0.00028 to 0.00044 0.00016 to 0.00024 TEMPERATURE. 183 Dividing- Coefficients for Correcting- the observed Resist- ance of Crutta-Percha at any Temperature to 1 .VJ JP. Temp. F.° Coeff. Temp. F. Coeff. Temp. F.° Coeff. Temp. F.° Coeff. 90 .3197 77.5 .8269 65 2.139 52.5 5.533 89.5 .3320 77 .8589 64.5 2.222 52 5.748 89 .3449 76.5 .8922 64 2.308 51.5 5.970 88.5 .3583 76 .9267 63.5 2.397 51 6.202 88 .3722 75.5 .9627 63 2.490 50.5 6.442 87.5 .3866 75 1.000 62.5 2.587 50 6.692 87 .4016 74.5 1.039 62 2.687 49.5 6.951 86.5 .4171 74 1.079 61.5 2.792 49 7.220 86 .4343 73.5 1.121 61 2.899 48.5 7.500 85.5 .4501 73 1.164 60.5 3.012 43 7.791 85 .4675 72.5 1.209 60 3.128 47.5 8.093 84.5 .4856 72 1.256 59.5 3.250 47 8.406 84 .5044 71.5 1.305 59 3.376 46.5 8.732 83.5 .5240 71 1.355 58.5 3.506 46 9.070 83 .5443 70.5 1.408 58 3.642 45.5 9.422 82.5 .5654 70 1.463 57.5 3.783 45 9.787 82 .5873 69.5 1.519 57 3.930 44.5 10.17 81.5 .6100 69 1.578 56.5 4.082 44 10.56 81 .6337 68.5 1.639 56 4.240 43.5 10.97 80.5 .6582 68 1.703 55.5 4.405 43 11.39 80 .6837 67.5 1.769 55 4.575 42.5 11.84 79.5 .7102 67 1.837 54.5 4.753 42 12.29 79 .7378 66.5 1.908 54 4.937 41.5 12.77 78.5 .7663 66 1.982 53.5 5.128 41 13.27 78 .7960 65.5 2.059 53 5.327 40.5 13.78 Example : The insulation resistance at 62° F. of a wire insulated with Gutta-percha is 500 meghoms ; what is the resistance at 75° F. ? Resistance =: 500 -f- 2.687 = 186.1 megohms. Dividing Coefficients for Correcting the observed Resist- ance of Hooper's India-It ubber at any Temperature to ¥5° E*. Temp. F.° Coeff. Temp. F.= Coeff. Temp. F.° Coeff. Temp. F.° Coeff. 90 .680 80.5 .868 71 1.108 61.5 1.414 89.5 .691 80 .880 70.5 1.122 61 1.433 89 .698 79.5 .891 70 1.137 60.5 1.451 88.5 .708 79 .902 69.5 1.152 60 1.470 88 .716 78.5 .914 69 1.167 59.5 1.489 87.5 .726 78 .926 68.5 1.182 59 1.508 87 .735 77.5 .938 68 1.197 58.5 1.527 86.5 .745 77 .950 67.5 1.212 58 1.547 86 .754 76.5 .963 67 1.228 57.5 1.567 85.5 .764 76 .975 66.5 1.244 57 1.587 85 .774 75.5 .987 66 1.260 56.5 1.608 84.5 .784 75 1.000 65.5 1.276 56 1.629 84 .794 74.5 1.013 65 1.293 55.5 1.650 83.5 .804 74 1.026 64.5 1.309 55 1.671 83 .814 73.5 1.039 64 1.326 54.5 1.693 82.5 .825 73 1.053 63.5 1.343 54 1.715 82 .836 72.5 1.068 63 1.361 53.5 1.737 81.5 .846 72 1.080 62.5 1.378 53 1.759 81 .857 71.5 1.094 62 1.396 52.5 1.782 184 PROPERTIES OF CONDUCTORS. Dividing- Coefficients — Continued. Temp. F.° Coeff. Temp. F.° Coeff. Temp. F.° Coeff. Temp. F.° Coeff. 52 1.805 49 1.949 46 2.106 43 2.274 51.5 1.828 48.5 1.975 45.5 2.133 42.5 2.303 51 1.852 48 2.000 45 2.160 42 2.333 50.5 1.876 47.5 2.026 44.5 2.188 41.5 2.363 50 1.900 47 2.052 44 2.216 41 2.394 49.5 1.925 4G.5 2.079 43.5 2.245 40.5 2.424 Mean Temperature. A piece of wire or cable whose length is I, and temperature t°, when con- nected to another wire or cable whose length is lv and temperature tx°, has a mean temperature It + LU linear H\pan§ioii of Metals due to Change of Temperature. A rod or wire I feet long will, by an increase of temperature of t°, increase its length to I (1 + at°) feet, where a has the following values : — Value of a for Metal. Zinc Lead Brass Copper Iron Steel Platinum Glass < Specific Heat, F.° .000016540 .000015830 .000010500 .000009560 .000006830 .000006381 .000004910 .000004870 c.° .00002976 .00002848 .00001890 .00001720 .00001229 =00001145 .00000884 .00000876 Specific heat Specific heat Element. of equal Element. of equal Weights. Weights. Water 1.0000 Rhodium .0580 Lithium . . .9408 Silver .0570 Sodium • . . .2934 Cadmium .0567 Magnesium . Aluminum . .2499 .2143 Tin Iodine .0562 .0541 Sulphur . . .1776 Antimony .... .0508 Potassium .1696 Tellurium .... .0474 Manganese . .1140 Thallium .0336 Iron .... .1138 Tungsten .0334 Nickel . . . .1091 Iridium ..... .0325 Cobalt . . . .1070 Platinum .0324 Zinc .... .0955 Gold .0324 Copper . . . .0951 Mercury (liquid) . . .0333 Bromine (solid) .0843 Lead .0314 Arsenic . . .0814 Bismuth .0308 Palladium . .0593 Osmium .0306 ^m^^m^^^^^m COEFFICIENTS. 185 If W = "weight of one substance whose temperature is T and specific heat S, w = weight of another substance whose temperature is t and specific heat s. Temperature of mixture = -Trri , z=L, WS -f- ivs 1 w (h — t) s-sw (T - t)' Temperature Coefficients of the .Resistivity of Pure Copper. , Temp. a o 5 2 Temp. a o .2Mi c3 .£ bJO^ «6 O O o o Matth Meter- Stand* Intern Ohms. o O o o i-hQ 0°. F°. 0°. F°. Matt Mete Stan Intel Ohm 0 32.0 1. 0. 0.14173 20 08.0 1.07968 .033294 0.15302 1 :w.s 1.003S8 .00168C 0.14228 21 o:t.s 1.08378 .034939 0.15360 2 35.0 1.00776 .003358 0.142S3 22 71.6 1.08788 .036581 0.15418 3 87.4 1.01166 .005036 0.14338 23 73.4 1.09200 .038222 0.15477 4 80.2 42.0 1.01558 .006712 0.14394 24 75.2 1.09612 .039859 0.15535 5 1.01950 .008386 0.14449 25 77.0 1.10026 .041494 0.15594 0 42. S 1.02343 .010059 0.14505 26 7S..S 1.10440 .043127 0.15653 7 44.0 1.02738 .01173C 0.14561 27 so.o 1.10856 .044758 0.15711 8 46.4 1.03134 .013400 0.14617 28 82.4 1.11272 .046385 0.15770 9 48. 2 1.03531 .015068 0.14673 29 84.2 1.11689 .048( »11 333 0.15830 10 50.0 1.03929 .016734 0.14730 30 80.0 1.12107 .049 0.15889 11 51.8 1.04328 .018399 0.14786 40 104 1.16332 .065699 0.164S8 12 ->>;.() 1.04728 .020062 0.14843 50 122 1.20625 .081436 0.17095 13 5,-.. 4 1.05129 .021723 0.14900 60 140 1 .24965 .096787 0.17711 11 15 57.2 1.05532 .023382 0.14957 70 158 1.29327 .111 0.18329 50.0 1.05935 .02503S 0.15014 80 176 1.33681 .126 0.18946 16 Cil.S 1.08339 .02669- 0.15071 90 104 1.37995 .139863 0.1'. 558 17 02.0 1.06745 .02834b 0.15129 100 212 1.42231 .152995 0.20158 18 04.4 1.07152 .02999£ 0.15186 19 00.2 1.07559 .031646 0.15244 I 1 ] Heat Conducting »• Power of JfEetals. Relative hea t Relative heat Metal. conducting power. Metal. conducting power. Silver 100 43.6 Go Cot d. . . . 98.1 8*.5 Tin ... . Steel . . . 42.2 >per (rolled) 39.7 Copper (cast) 81.1 Platinum 38.0 Aluminum . . 66.5 Sodium . . 36.5 Zinc .... 64.1 Iron (cast) . 35.9 Bismuth . . 61.0 Lead . . . 28.7 Cadmium . . 57.7 Antimony . 21.5 186 PROPERTIES OF CONDUCTORS. RE8ISTAICE METALS. Following are data on modern resistance metals, supplied by Hermann Boker & Co., of 101-103 Duane Street, New York. The resistance data are from tests by Helmlioltz and tbe German Impe- rial Physical and Technical Institute of Charlottenburg, Germany. Dimensions, Resistances, and Weijrlits of Resistance Wires. i Ohms per 1000 feet. Feet per Lb. Approxi- 6 SZ3 A 1 mately. S s < Xfl eg % Cm ® I © S u ■5 Jo 3 a? 3 £ z > 14 .0641 4107. 125.9 73.5 63.7 49.7 56.6 85. 79.2 10 .0508 2583. 200.3 116.9 101.4 78.9 90.1 135.3 125.9 17 .0453 2048. 252.6 147.4 127.8 99.6 113.9 170.6 158.7 18 .0403 1624. 318.6 185.9 161.2 125.6 143.4 215.5 200.5 19 .0359 1289. 401.4 234.3 203.1 158.2 181.1 271.0 252. 20 .0320 1024. 506.5 295.6 256.3 199.7 227.9 342.3 318.4 21 .0285 812.3 641.5 374.4 324.6 252.9 288.7 433. 402.6 22 .0253 640.1 805.7 470.1 407.7 317.5 362.6 543.5 505.5 23 .0225 506.25 1022.1 596.6 517.2 402.8 459.9 689.6 641.4 24 .0201 404. 1280.7 747.6 648. 504 9 576.3 870. 809.1 25 .0179 320.4 1620. 945.6 819.7 638.9 729. 1098. 1021.2 26 .0159 252.8 2036.5 1192.9 1030.5 802.8 916.4 1370. 1274.1 27 .0142 201.6 2566.2 1497.8 1298.5 1011.5 1154.8 1724. 1604. 28 .0126 158.8 3238.1 1890.1 1638.5 1276.4 1457.1 2174. 2022. 29 .0113 127.7 4125. 2407.8 2087.2 1626. 1856.2 2777. 2583. 30 .0100 100. 5148.7 3005.3 2605.2 2029.5 2316.9 3448. 3207. 31 .0089 79.2 6491.6 3789.2 32S4.7 2558.8 2921.2 4347. 4043. 32 .0080 64. 8187.5 4779.1 4142.8 3227.3 3684.3 5555. 5167. 33 .0071 50.4 10322. 6025.1 5222.9 4068.9 4644.9 7142. 6600. 34 .0063 39.69 13020. 7600.4 6588.1 5132.6 5659. 9090. 8354. 35 .0056 31.56 16416. 9582.7 8308.5 6471.1 7387.2 11100. 10323. 36 .005 25. 20698. 12081. 10473. 8158-8 9314.1 14286. 13280. 37 .0044 19.83 26094. 15229. 13203. 10285. 11743. 17543. 16315. 38 .004 16. 32916. 19213. 16655. 12975. 14712. 22220. 20665. 39 .0035 12.25 41495. 24218. 20996. 16357. 18672. 27700. 25761. 40 .0031 9.61 52373. 30570. 26500. 20644. 23567. 35714. 33215. RESISTANCE METALS. 187 maximum Amperes for Safe Constant XiOad with Free Radiation. Nickeline I. B. &S. Gauge No. Superior. la la. and German Silver. Nickeline II. 18 11.8 15.75 17.2 18.2 19 10.25 13.6 14.4 15.6 20 8.5 11.5 12.1 13.0 21 7.2 9.7 10.0 11.0 22 6.0 8.0 8.4 9.1 23 5.2 6.8 7.1 7.8 24 4.5 5.8 6.0 6.5 25 4.0 4.9 4.8 5.5 26 3.5 4.1 4.1 4.6 27 3.0 3.6 3.6 4.0 28 2.7 3.1 3.1 3.5 29 2.5 2.9 2.9 3.2 30 2.3 2.7 2.7 2.9 32 2.0 2.5 2.5 2.63 34 1.7 2.2 2.2 2.3 36 1.5 2.0 2.0 2.0 Resistance Ribbon. " Superior " Crrade. ■3© .2 s Ohms per 100 feet <% it, &$ H Jin. Jin. fin. I in. fin. fin. fin. lin. 8 .128 25.36 12.68 8.45 6.34 5.07 4.22 3.62 3.17 9 .114 28.59 14.29 9.53 7.14 5.71 4.76 4.08 3.57 10 .101 32.22 16.11 10.74 8.05 6.44 5.37 4.60 4.02 11 .0907 35.93 17.96 11.98 8.98 7.18 5.99 5.13 5.49 12 .0808 40.19 20.09 13.39 10.04 8.04 6.69 5.74 5.02 13 .0719 45.61 22 80 15.20 11.40 9.12 7.60 6.51 5.70 14 .0641 50.72 25 36 16.90 12.68 10.14 8.45 7.24 6.34 15 .0571 57.18 28.59 19.06 14.29 11.43 9.53 8.16 7.14 16 .0508 64.44 32.22 21.48 16.11 12.89 10.74 9.20 8.05 17 .0452 71.86 35.93 23.95 17.96 14.37 11.97 10.28 8.98 18 .0403 80.38 39.19 26.79 20.09 16.07 13.39 11.48 10.04 19 .0359 91.22 45.61 30.40 22.80 18.24 15.20 13.03 11.40 20 .0320 101.44 50.72 33.81 25.36 20.29 16.90 14.50 12.68 21 .0284 114.36 57.18 38.12 28.59 22.87 19.06 16.33 14.29 22 .0253 128.88 64.44 42.96 32.22 25.77 21.46 18.41 16.11 23 .0225 143.72 71.86 47.90 35.93 28.74 23.95 20.53 17.96 24 .0201 160.76 80.38 53.59 40.19 32.15 26.79 22.96 20.09 25 .0179 182.44 91.22 60.81 45.16 36.49 30.40 26.06 22.80 26 .0159 202.88 101.44 67.62 50.72 40.57 33.81 28.98 25.36 27 .0142 228.72 114.36 76.24 57.18 45.74 38.12 32.67 28.59 28 .0126 257.76 128.8S S5.92 64.44 51.55 42.96 36.82 32.22 29 .0112 287.44 143.72 95.81 71.86 57.49 57.90 41.06 35.93 30 .0100 321.52 160.76 107.17 80.38 64.30 53.59 45.93 40.19 31 .0089 364.88 182.44 121.62 91.22 72.97 60.81 52.12 45.16 32 .0079 405.76 202.88 135.25 101.44 81.15 67.62 57.96 50.72 33 .0071 457.44 228.72 152.48 114.36 91.49 76.24 65.33 57.18 34 .0063 515.52 257.76 171.84 128.88 103.10 85.92 73.64 64.44 35 .0056 574.88 287.44 191.62 143.72 114.97 95.81 82.12 71.86 36 .005 643.04 321.52 214.34 160.76 128.60 107.17 91.86 80.38 37 .0044 729.76 364.88 243.25 182.44 145.95 121.62 104.25 91.22 38 .0039 811.52 405.76 270.50 202.88 162.30 135.25 115.93 101.44 The number of feet to the pound of any size of the above ribbon can be found by dividing the constant 0.26 by the cross sectional area in square inches. 188 PROPERTIES OF CONDUCTORS. Resistance Ribbon. la la Quality. 6 <^> . Ohms per 1000 feet <3& 5o PQJS H g m. tin. § in- 5 in. f in. f in. fin. lin. 8 .128 14.81 7.40 4.93 3.70 2.96 2.'46 2.11 1.85 9 .114 16.69 8.34 5.56 4.17 3.34 2.78 2.38 2.08 10 .101 18.80 9.40 6.26 4.70 3.76 3.13 2.70 2.35 11 .0907 20.97 10.48 6.99 5.24 4.19 3.49 2.99 2.62 12 .0b08 23.46 11.73 7.82 5.86 4.69 3.91 3.35 2.93 13 .07i9 26.63 13.31 8.87 6.65 5.32 4.43 3.80 3.32 14 .06*1 29.62 14.81 9.87 7.40 5.92 4.93 4.22 3.70 15 .0o71 33.38 16.69 11.12 8.34 6.68 5.56 4.77 4.17 16 .0508 37.60 18.80 12.53 9.40 7.52 6.26 5.37 4.70 17 .0*52 41.94 20.97 13.98 10.48 8.38 6.99 5.99 5.24 18 .0403 46.92 23.46 15.64 11.73 9.38 7.82 6.70 5.86 19 .0359 53.26 26.63 17.78 13.31 10.64 8.87 7.60 6.65 20 .0320 59.24 29.62 19.75 14.81 11.84 9.87 8.46 7.40 21 .0284 66.76 33.38 22.25 16.69 13.55 11.12 9.53 8.34 22 .0253 75.20 37.60 25.07 18.80 15.04 12.53 10.74 9.40 23 .0225 83.88 41.94 27.96 20.97 16.77 13.98 11.98 10.48 24 .0W1 93.84 46.92 31.28 23.46 18.77 15.64 13.40 11.73 25 .0179 106.52 53.26 35.50 26.63 21.30 17.78 15.21 13.31 26 .0159 118.48 59.24 39.49 29.62 23.69 19.75 16.91 14.81 27 .01*2 133.52 66.76 44.50 33.38 26.70 22.25 19.07 16.69 28 .0126 150.40 75.20 50.13 37.60 30.08 25.07 21.50 18.80 29 .0112 167.76 83.88 55.92 41.94 33.55 27.96 23.96 20.97 30 .0100 187.68 93.84 62.56 46.92 37.53 31.28 26.81 23.46 31 .00&9 213.04 106.52 71.01 53.26 42.60 35.50 30.43 26.63 32 .0079 236.96 li8.*8 78.98 59.24 47.40 39.49 33.82 29.62 33 .0071 267.04 133.o2 89.01 66.76 53.40 44.50 38.15 33.38 34 .0063 300.80 150.40 100.26 75.20 60.16 50.13 42.97 37.60 35 .0056 335.52 167.76 111.84 83.88 67.10 55.92 47.93 41.94 36 .005 375.36 187.68 125.12 93.84 75.07 62.56 53.62 46.92 37 .0044 426.08 213.04 142.02 106.52 85.21 71.01 60.87 53.26 88 .004 473.92 236.96 157.97 118.48 94.78 78.98 67.64 59.24 Specific Resistance and Temperature Coefficient. Specific Resistance at 20° C, Coefficient for 1°C. Superior la la, hard la la, soft Nickeline No. II., hard Nickeline No. II., soft . Nickeline No. I., hard . Nickeline No I., soft . German Silver, average Manganin Constantin 85.4 to 86.5 50.2 47.1 33.9 32.3 43.6 40.7 31.5 47.5 50. — 55 .00067 to .00073 — .000011 + .00u005 ■ .000168 -- .000181 - .000076 - .000077 - .00025 : .00001 : .00001 "SUPERIOR" WIRE. Specific gravity, 8.4. Specific resistance at 20° C, 86 microhms. Coefficient of temperature, mean value, for l°C.,-f 0.00065. BOKER S WIRES. 189 Resistance of one circular mil foot of " Superior " wire f ohms. 20° C, 517.5 This resistance material does not rust, nor show any sign of oxidation at ordinary temperature, and it shows no sign of deterioration after being sub- mitted to a temperature just below a visible red heat as a permanent load. Prices of JBare Wire per Pound. B. &S. Gauge. Inch. Superior. la la. Nickeline I. Nickeline 11. 15 and heavier .057 $1.07 $.078 $0.66 $.61 16 .05082 1.09 .80 .69 .63 17 .04525 1.09 .80 .69 .63 18 .0403 1.09 .80 .69 .63 19 .0358 1.09 .80 .69 .63 20 .0319 1.09 .80 .69 .63 21 .0284 1.121 .85 .72 .66 22 .0253 1.16 .88 .75 ,70 23 .0225 1.16 .88 .75 .70 24 .0201 1.24 .94 .78 .74 25 .0179 1.26 .96 .83 .77 26 .0159 1.28 .96 .85 .79 27 .01419 1.33 1.04 .90 .84 28 .01264 1.37 1.09 .94 .88 29 .01125 1.40 1.12 .97 .91 30 .010 1.45 1.17 1.02 .96 31 .00892 1.52 1.24 1.09 1.03 32 .00795 1.60 1.33 1.16 1.10 33 .00708 1.69 1.45 1.26 1.20 34 .0063 1.81 1.55 1.38 1.33 35 .0056 1.98 1.75 1.55 1.49 36 .005 2.56 2.20 2.13 2.07 37 .00445 4.21 3.85 3.72 3.72 38 .00396 6.36 6.00 5.93 5.87 39 .00353 8.11 7.75 7.68 7.62 40 .00314 10.36 10.00 9.93 9.87 .00196 15.60 15.25 15.18 15.12 190 PROPERTIES OF CONDUCTORS. Prices of Silk Covered IFire per Pound. 9 tjj S3 a o Superior. la la. Nickeline I. Nickeline II. O 6 6 / r K 7 19 20 30 40 50 60 70 SO 90 100 110 120 130 HO 150 KILOVOLTS EFFECTIVE Fig. 0. Points, Smooth Core Alternator, 125 Cycles. and over again, and independent of the atmospheric condition, the frequency, etc., to such an extent that the striking distance between needle points offers the most reliable means to determine very high voltages. For this reason, it is used in this manner as final check in all high potential insula- tion tests of the General Electric Company. DISRUPTIVE VALUE OF DIELECTRICS. 195 2d. No physical law has been found to represent satisfactorily all the observations. Some point to the existence of a constant dielectric strength of air, analogous to the tensile strength of mechanics. Others point to the existence of a spurious counter E.M.E. of the spark or transition resistance from electrode to air. 3d. Constant dielectric strength. Cylinders of 1.11 in. diameter give an average disruptive strength of air of 60 kilovolts per inch. Cylinders of .315 in. diameter, an average dielectric strength of 77. Spheres at very small distance point toward the latter value. As a disturbing factor in this case, enters the electrostatic brush discharge, which by a partial breakdown of the air surrounding the electrodes changes and increases the size and decreases the distance of the effective terminals. z.t> I' I "0 // 1 / / / |l.5 <* (*■€■ *y // 1 ^ 'S ,*> /$/ .5 /I * 11 10 9 1 0 20 30 i ECT 0 VE- 5 0 ''/ // / ■v .< ■ 2 7 I / 5 4 3 2 1 /// /$/ ^ ^/ W ,« £y ''/ ,\' <*2 &&£ 10 20 30 40 50 60 70 80 90 100 110 120130140 KILOVOLTS EFFECTIVE Fig. 0. Comparison of Points and Spheres. Smooth Core Alternator, 125 Cycles. 4th. Counter E.M.F. of the sparks. The tests with sharp points give 22 kilovolts, or 11 kilovolts for a single transition from terminal to air. Spheres give curves pointing to a similar phenomenon. Electric conductors in- serted at right angles into or parallel with the discharge, point to the exis- tence of a counter E.M.F. of the same magnitude. The beginning of the electrostatic brush discharge is at a potential of this magnitude also. 196 PROPERTIES OF CONDUCTORS. TABLE. — POIXTS. 2\" needles. 125 cycles. Smooth Core Alternator : Ironclad Alternator : A-10-30-1500. A -10-60-1500. of . Kilovolts: effective. of . Kilovolts : effective. d a> C3r3 3-3 ** 2 p'rt d. ?T3 ^'3 c3 3 5'" d. 3' • it fee . p 3 d 3 °° tL " < ti'" ^.3 < 1-3 .3 •n S .25 4.25 .25 4.13 .5 10.0 .5 9.0 10.0 9.5 11.0 14.5 1.0 20.4 1.0 16.0 1S.5 17.7 22.0 25.3 1.5 29.3 1.5 24.3 26.0 25.1 31.0 35.5 2.0 35.2 2.0 30.5 30.5 30.5 38.0 43.0 2.5 40.4 2.5 33.9 35.0 34.4 43.5 50.3 3.0 45.6 3.0 36.3 38.0 37.1 48.0 54.5 3.5 49.4 3.5 42.2 41.7 42.0 51.0 63.0 4.0 52.5 4.0 41.3 45.0 43.2 55.5 4.5 59.6 4.5 45.5 48.0 46.7 61.0 5.0 61.0 5.0 48.4 50.5. 55.0 49.5 5.5 65.7 5.5 53.0 54.0 6.0 69.8 69.5 69.65 6.0 56.1 58.8 57.4 6.5 73.4 74.7 74.05 6.5 59.8 62.0 60.9 7.0 77.5 79.2 78.35 7.0 63.3 64.7 64.0 7.5 83.8 83.0 83.4 7.5 67.5 69.0 68.3 8.0 86.8 87.3 87.05 8.0 70.9 73.4 72.1 8.5 90.5 90.2 90.35 8.5 75.8 76.0 75.9 9.0 95.0 93.7 94.35 9.0 79.S 79.2 79.5 9.5 97.7 96.3 97.0 9.5 84.8 82.5 83.6 10.0 101.5 99.0 100.25 10.0 88.8 86.4 87.6 10.5 107.0 103.0 105.0 10.5 93.5 89.5 91.5 11.0 111.5 107.5 109.5 11.0 97.7 93.0 95.4 11.5 114.0 110.5 112.5 11.5 102.0 12.0 121.0 116.0 118.5 12.0 107.7 12.5 125.5 120.0 122.75 12.5 111.0 13.0 133.0 123.0 128.0 13.0 117.5 13.5 135.0 127.0 131.0 13.5 122.5 14.0 140.0 129.0 134.5 14.0 128.0 14.5 144.0 136.0 140.0 14.5 134.4 15.0 150.0 15.0 138.3 15.5 155.0 16.0 159.5§ * 85° F. Weather sultry. t 75°-80° P. Weather clear and hright. j 70° F. Weather cool and cloudy. § Internal discharges in intermediary transformers F' F". VALUES OF VARIOUS DIELECTRICS. 197 / // / /' ^ / / /,-> ♦ / / « « -^ s£ - 10 20 30 40 50 60 70 80 90 100 110 120 130 KILOVOLTS EFFECTIVE Points in Air. Fog and Steam at Atmospheric Pressure. Ironclad Armature, 125 Cycles. Values of Various Dielectrics. Steinmetz, February, 1893. A. I. E. E. Electrostatic gradient at Formula for Calculating tbe Material. 0 5 I 25 Sparking Distance. Kilovolts, in Kilovolts per Centimeter. D E = P. D.in Kilovolts. Air 139 4170 130 16.7 3200 52 130 339 81 80 64 30 16 36 43 10.1 11.9 1660.0 15.3 .86 D = . 24 ^ + .0145^z D = 7.66 E -\- 2.3 E* D = 7.66 E D = 3E D = 12.4 E D = 12.5 E D = 15.7 E D = 30E D = 60E & = 28E D = 23E D = 55(E — 2)2 Vulcanized fiber, red . . . Dry wood fiber Paraffined paper .... Melted paraffine .... Boiled linseed oil ... . Turpentine oil Copal varnisb Crude lubricating oil (min- eral oil) Vulcabeston Asbestos paper Creeping discbarge . . . 198 PROPERTIES OF CONDUCTORS. Tests of Vulcanized India-Rubber. Lieutenant L. Vladomiroff , a Russian naval officer, has recently carried out a series of tests at the St. Petersburg Technical Institute with a view to establishing rules for estimating the quality of vulcanized India-rubber. The following, in brief, are the conclusions arrived at, recourse being had to physical properties, since chemical analysis did not give any reliable result : 1. India-rubber should not give the least sign of superficial crack- ing when bent to an angle of 180 degrees after five hours of exposure in a closed air-bath to a temperature of 125° C. The test-pieces should be 2.4 inches thick. 2. Rubber that does not contain more than half its weight of metallic oxides should stretch to five times its length without breaking. 3. Rubber free from all foreign matter, except the sulphur used in vulcan- izing it, should stretch to at least seven times its length without rupture. 4. The extension measured immediately after rupture should not exceed 12% of the original length, with given dimensions. 5. Suppleness may be deter- mined by measuring the percentage of ash formed in incineration. This may form the basis for deciding between different grades of rubber for certain purposes. 6. Vulcanized rubber should not harden under cold. These rules have been adopted for the Russian navy. — Iron Age, June 15, 1893. CJUTTA-PERCHA. Specific gravity, 0.9693 to 0.981. Weight per cubic foot, 60.56 to 61.32 pounds. Weight per cubic inch, 0.560 to 0.567 oz. Softens at 115 degrees F. Becomes plastic at 120 degrees F. Melts at 212 degrees F. Oxidizes and becomes brittle, shrinks and cracks when exposed to the air, especially at temperatures between 70 and 90 degrees F. Oxidation is hastened by exposure to light. Oxidation may be delayed by covering the gutta-percha insulation with a tape which has been soaked in prepared Stockholm tar. Where gutta-percha is kept continually under water there is no notice- able deterioration, and the same applies where gutta-percha leads are cov- ered with lead tubing. Stretched gutta-percha, such as is used for insulating cables, will stand a strain of 1,000 pounds per square inch before any elongation. The breaking strain is about 3,500 pounds per square inch. The tenacity of gutta-percha is increased by stretching it. Where Z)= diameter of gutta-percha insulation, and d = diameter of con- ductor of copper (both dimensions in mils) the weight of gutta-percha per knot is -_. When w — Aveight of stranded copper conductor per knot in pounds, and W = weight of gutta-percha per knot in pounds, then outer diameter = V70.4 w + 491 W mils. If the conductor is solid, then, outer diameter — V55 w + 491 W mils. After one minute's electrification, the insulation resistance per knot of best quality gutta-percha insulated cable will be, = 750 (log D — log d.) megohms at 75° F. Resistance of Gutta-JPercha under Pressure. — The resistance of gutta-percha under pressure increases according to the following formula, j when R — the resistance at the pressure of the atmosphere, and r the resis- tance at p pounds per square inch. r = R(1 + 0.00023 p). GUTTA-PERCHA. 199 Resistance of Griitta-Pcrclia decreases with Rise of Tem- perature.- The resistance of gutta-percha decreases, as per the follow- ing form'-" as the temperature rises, where R = resistance at the low temperature, r ■=. resistance at the high temperature, t = difference in temperature, degrees F. ; then log R = log r — t log 0.9399, and log r-logR + t log 0.9399. Capacity and Resistance of Ccutta-Percha. The resistance of a plate of gutta-percha one foot square and .001 inch thick = 1.066 megohms at 75° F. The electrostatic capacity of the same piece at the same temperature is .1356 microfarads. The product of the resistance in megohms by the electrostatic capacity in microfarads, both taken at 75° F., after one minute's electrification = 144.4. Ratio of D -)- d for strand and solid conductors. For stranded conductor insulated with gutta-percha, 5=^1 + 6.97^. d " it) For solid conductor insulated with gutta-percha, S 1 + 8.93—- ' w In which D = outer diameter of cable, and d =z diameter of conductor, and W and to = weight of gutta-percha and of conductor respectively in pounds. The approximate electrostatic capacity of a gutta-percha insulated cable per knot is °-1877 microfarads. log D — log d The electrostatic capacity of a gutta-percha insulated cable compared Avith one of the same size insulated with india rubber is about as 120 is to 100. Jointingr Cfutta Percha Covered Hrire. First remove the gutta-percha for about two inches from the ends of the wires which are to be jointed. Fig. 4. Next cross the wires midway from the gutta-percha, and grasp with the pliers. Fig. 5. 200 PROPERTIES OF CONDUCTORS. Then twist the wires, the overlapping right-hand wire first, and then, reversing the grip of the pliers, twist the left-hand wire over the right. Cut off the superfluous ends of the wires and solder the twist, leaving it as shown in Fig. G. Next warm up the gutta percha for about two inches on each side of the wist. Then, first drawdown the insulation from one side, half way over Fig. 7. the twisted wires, Fig. 7, and then from the other side in the same way, Fig. 8. Then tool the raised end down evenly over the under half with a heated iron. Then warm up the whole and work the " drawdown " with the thumb and forefinger until it resembles Fig. 9. Now allow the joint to cool and set. Fig. 9. Next roughen the drawdown with a knife, and place over it a thin coating of Chatterton's compound for one inch, in the center of the drawdown, which is also allowed to set. Next cut a thick strip of gutta-percha, about an inch wide and six inches long, and wrap this, after it has been well warmed by the lamp, evenly over the center of the drawdown. Fig. 10. Fig. 10. The strip is then worked in each direction by the thumb and forefinger over the drawdown until it extends about 2 inches from center of draw- down. Then tool over carefully where the new insulation joins the old, after which the joint should be again warmed up and worked with the fore- finger and thumb as before. Then wet and soap the hand, and smooth and round out the joint as shoAvn in Fig. 11. Between, and at every operation, the utmost care must be exercised to remove every particle of foreign matter, resin, etc. JOINTS IN CABLES. 201 Joints in Jtubner Insulated Cables. Preparation of Ends. — Remove the outside protecting braid or tape, and bare the conductor of its rubber insulation for two or three inches back from the end. Clean the metal carefully by scraping with a knife or with sandpaper. Tfletal Joint. — If solid conductor, scarf tbe ends with a file so as to give a good long contact surface for soldering. If conductor is stranded, carefully spread apart the strands, cutting out the centres so conductors can be butted together, the loose ends interlacing as in Fig. 9, and bind wires down tight "as in Fig. 10. with gas or other pliers. Solder carefully, using no acid ; resin is the best, although jointers often use a spermaceti candle as being handy to use and easy to procure. Large cables are easiest soldered by dipping the joint into a pot of molten solder, or by pouring the molten metal over the joint. The insulation of all kinds of joints is done in the same manner, the only difference in the joint being the manner in which the conductors are joined together. Following are some of the styles of joining conductors, which are afterward insulated with rubber, and covered with lead when necessary. Seeley's Cable Connectors. —Tbe cuts below show a style of cop- per connectors very handy in joining cables. They are copper tinned over, and after putting in place can be " sweated" on with solder ; when dry can be insulated as previously described. 202 PROPERTIES OF CONDUCTORS. Insulating- the Joint. — Jointers must have absolutely dry and clean hands, and all tools must be kept in the best possible condition of cleanliness. Clean the joint carefully of all dux and solder ; scarf back the rubber insulation like a lead-pencil for an inch or more with a sharp knife. Carefully wind the joint with three layers of pure unvulcanized rubber, taking care not to touch the strip with the hands any more than neces- sary ; over this wind red rubber strip ready for vulcanizing. Lap the tape upon the taper ends of the insulation, and make the covering of the same diameter as the rubber insulation on the conductor, winding even and round. Cover the rubber strip with two or three layers of rubber-saturated lead covering-. — 11 the insulation is covered and protected by lead, a loose sleeve is slipped over one end before jointing, and slipped back over the joint when the insulation is finished, a plumber's wiped joint being made at the ends. Fig. 17. Joints in Waring Cahles. — This cable is covered with cotton, thoroughly impregnated with a composition of hydro-carbon oils applied at high temperature, the whole being covered with lead to protect the insula- tion. The insulating properties of this covering are very high if the lead is kept intact. Metal joints are made as usual, and a textile tape may be used for cover- ing the bare copper. A large lead-sleeve is then drawn over the joint, and wiped onto the lead covering at either end ; then the interior space is filled with a compound similar to that with which the insulation is im- pregnated. Joints in Paper Insulated Cahles. — This cable is covered or insulated with narrow strips of thin manila paper wound on spirally, after which the Avhole is put into an oven and thoroughly dried, then plunged into a hot bath of resin oil, which thoroughly impregnates the paper. This insulation is not the highest in measurement, but the electrostatic capacity is low and the breakdown properties high. When used for telephone pur- poses the paper is left dry, and is wound on the conductor very loosely, thus leaving large air spaces and giving very low electrostatic capacity. Joints are made as in the Waring cable by covering the conductor with paper tape of the same kind as the insulation, then pulling over the lead sleeve, which is finally filled with paraftine wax. Hundreds of miles of such cables being thus employed at pressures ran- ging from 500 to 10,000 volts — notably in the Metropolitan district of New York. Cost of Straight or Sleeve Joints Insulated with Ruhhei On rubber-insulated, lead covered cable. Plumber 1 hour .25 Insulator i hour .15 Helper 1 hour .15 Red rubber 1 oz. @ $1.00 per lb. .07 Pure rubber . . . , 1 oz. @ $2.00 per lb. .15 Grimshaw tape . 1 oz. @ .50 per lb. .03 UNDERGROUND CONSTRUCTION. 203 Copper sleeve .035 Lead sleeve .06 Solder 1$ lbs. @ .20 .30 Pasters . . . . 2 .005 Coal • .10 Candle (for flux) .01 Total . $1.31 Cost of T Joint on Rubber Insulated Cable. T on rubber-insulated lead-covered cable. Plumber H hour $.375 Insulator § hour .225 Helper 1£ hour .225 Red rubber li oz. @ $1.00 per lb. .11 Pure rubber 1£ oz. @ 2.00 per lb. .23 Grimshaw tape 1J oz. @ .50 per lb. .05 Solder 2 lbs. @ .20 per lb. .40 Lead T .26 Copper T .075 Pasters .0075 Candle .01J Coal .10 Total $2.07 UHTDEItOieOUHTJ* EJLECTKICAI, COISTRrCflOI. Mr. Louis A. Fei'guson, in paper before the National Electric Light Asso- ciation in May, 1899, gives the results of his observations as to the cost of laying and maintaining underground conductors. Labor, fittings, paving, and laying one length "of Edison main tube costs from $5.45 in unimproved streets', with no paving, to $29.81 in asphalt. The annual cost of supervision and maintenance amounts to 1.9% per annum of the original investment. The total cost per duct foot of laid conduit of various types is given in the following table, where the higher price is for asphalt pavement, and the lower one for no pavement. National conduit $16.74 to $57.24 Francis conduit 14.66 to 55.16 Lithocite conduit 15.18 to 55.6S Camp tile 14.14 to 54.64 Three-inch iron pipe 22.50 to 66.00 Manholes as used in Chicago cost for size 2/ + 2' x 3' from $32.18 to $38.63 ; for size 8'x8'x 8' $194.65 to $224.72. I,AW OF B. & S. CflTCIE. The absence of a wire table may often be compensated for by remember- ing the following approximate facts concerning the B. & S. gauge. Diameter of No. 10 wire = .1 inch. Resistance of No. 10 per 1000 feet =r 1 ohm. Weight of No. 10 per 1000 feet = 31.37 lbs. Diameters are halved for every six units increase in gauge No.; i.e., No. 16 has half the diameter of No. 10, and No. 4 has twice the diameter of No. 10. Accordingly cross-sectional areas double at every decrease of three in the gauge number. The gauge numbers correspond to cross-sections and conductivities which vary as an inverse geometrical progression having a ratio of 1 .26. 204 PROPERTIES OF CONDUCTORS. FISIlfG EITECT8 O*' E^ECTMC CIRREITI. By W. H. Preece, F. R. S. See " Proc. Roy. Soc," vol. xliv., March 15, 1888. M The Law — I = a d §, where /, current ; a, constant ; and d, diameter — is strictly followed; and the following are the final values of the constant "a," for the different metals as determined by Mr. Preece : — Inches. Centimeters. Millimeters. Copper 10,244 2,530 80.0 Aluminum .... 7,585 1,873 59.2 Platinum 5,172 1,277 40.4 German Silver. . . . 5,230 1,292 0.8 Platinoid 4,750 1,173 37.1 Iron 3,148 777.4 24.6 Tin 1,642 405.5 12.8 Alloy (lead and tin 2 to 1) 1,318 325.5 10.3 Lead 1,379 340.6 10.8 Table Chiving- the J&iameters of Wires of Various IVlateri als Which Will lie ITuse«l by a Current of Given Strength. — W. H. Preece, F. R. S. d= ( (I\*P Diameter in Inches. .5 Tj5 ~ C g s 0ni .5 »2 £K jgg Jo i to 111 ai Suo CO »=t + l >=>b+*m (4) X = 3S~*W- (5) x— /3yl — 3j y- w In these formulae - one-half span. : one-half length of wire in span. = deflection at centre in same units as y. w = weight per foot of wire. Suppose Ave have a span of 200 feet of hard-drawn copper wire weighing one pound to 10 feet, and a deflection of two feet or .01 of the span. « •=(£)" + •• == 2500.33 +. (3) 1 = 100 [1 + 1(^)1. = 100.026 6 +. 21 = 200.053 +. (4) x = 7501 — V56,205,001 — 30,000. (4) T — /30,0( "->J- In calculating the table, page h65, the deflection of the line was determined at — 10° F. by formula 4, the value of S being 30,000 divided by 3.85 or 7,792. For the other temperatures the length of the wire was calculated from the following formula : Length = I (1 -}- .000009 3 t ) Here t is the difference in temperature in degrees Fahrenheit. By formula 5 the deflection corresponding to the new length was found. The coefficients of linear expansion for each degree Fahrenheit are follows : Copper, .000 009 3. Iron, .000 006 8. Lead, .000 016. TEMPERATURE EFFECTS. 207 TEMPEBATURE EFFECTS IEf §PAWS. i Temperature in degrees Fahrenheit. Ph _g —10° 30° 40° 50° 60° 70° 80° 90° 100° Deflections in inches. 50 .5 6 8 9 9 10 11 11 12 60 .7 8 10 11 11 12 13 13 14 70 1. 10 11 12 13 14 15 15 17 80 1.2 11 13 14 15 16 17 IS 19 90 1.6 13 14 16 17 18 19 20 21 100 1.9 14 16 17 19 20 21 23 £4 110 2.3 16 18 19 21 22 24 25 26 120 2.8 17 19 21 22 24 26 ' 27 28 130 3.2 19 21 23 25 26 28 29 31 140 3.7 20 23 25 27 28 30 32 33 150 4.3 oo 24 26 28 30 32 34 36 160 4.9 23 26 28 30 32 34 36 38 170 5.5 25 28 30 32 35 37 38 40 180 6.2 26 29 32 34 37 39 41 43 190 7. 28 31 34 36 39 41 43 45 200 7.7 31 33 36 38 41 43 45 48 Hard-drawn copper wire, 60,000 pounds strength per square inch. Strain at — 10° F., 30,000 pounds per square inch. The following tables give the dip in feet and inches of No. 0 B. & S. cop- per trolley wire between spans 125' apart, and the strain in pounds for vari- ous temperatures : Initial Maximum Strain SOOO I £ N f oi ci i:o lo x oi i- oo ~ - -: : i c co co — — o ; /: oo t:i;.-.[. -.-t::;!- ~ x i - - i.o i o -os ht o i . : -)-_ co. q :~. ci to q x t-i l- -# i-j © Jdolrtt-eooLi d ci — ' cd d 1 1- io o i © xj d 3 CO ITS I LO -+ T CO CO CO 0 1 0 1 0 I 0 I >-i ■- — — _ O CM <-H -* CO 00 CO 0 j 0 I --i ■ ~ (0 ?' ? f1 - - : 100a CO CO 'X uo -f x ;-o co — co HM50CS CM - o: co ? i Lc i oo o o oo > P i-i s ;2 oj c o, 4 I; g 4 8 s oo ^ fi* co co in co CO CO L.0 ed|i^sd oj d oo co CO OS — I «* oj cs d co d © OS O CN JO c L0 CO X X CO " io od ci ~ ^ -I- ~o ': ?^8 "*° CM S CS o CM ' teas ~ oo /: x f- i~ COtrtOOl- -* oi d l- id gnSoa 01 01 01 01 01 t — i co o io> ^^'88S Z. ~ OJ x - r. i - .-:' : . ~ ~ '- ' ~ '. ON-f oco -h co co co oi HfflOOOH 23 5 id Si 5 01 CO CO T*< oi t- -#" oi d t^ io ssi^a 10 01 CO CO CO SgS8!5 CO t- IO CO CM fe te t- ■> t- - t>TjH(NOI>lO o-tqi? CO CO 01 O X ^8SSc| CO UO CO 01 H O CO O O CO p p & s a UO -* CO CM i-l CO CO CO CO CO OCONOOO 'o^lcolSJ ^322Sdo M O 01 I- • = '-.' z •fHodo Ol Ol 01 01 ? oi oi oi oi 5COOO CO •5 P 12 S :2 10 i.O -H CO CO CO CO CM 00 id iS ■* -+i co co :-;v,. 7jh ^?^a « i> n o i> co t-^oidt^id CO 1.0 -+ 01 01 -1 01 01 Ol 01 ClOOOO --j co io x oi c^cm322 ^OCOiOiC-* OC0OOC0 J* rt< co eo cm oocqtHco^ o eg c-i oo" id J2 -*1 -* CO CO o oo t- co q CMCMC^ MONO OOC5C0 q^ 388 oi oi oi oi oi = =3^ OTtHOUOO -*' T-i d os d ol^olo^ i§8^8 :8Sc! COCO 10- 00 CO OH M CM CN CM CM CO CO < l?0% U0> CO t- 00 OS illll' US"" ©cot- oo© nidrid00 co^c?8^8 t^ -* CM © t^ g 210 PROPERTIES OF CONDUCTORS. ft fi-N Sg sa 0 ft 3 g ft a, ui sxredg O rl Ol Ol CO 60066 O -H MfflO t> OS O r-H CO g^S p-® 00 CO op 60066 So'SSo 1-1 CI CO tH m SSggS CIOICOCO^ S2922 NLOMOM OOO rH T-i tH (M CO "# IO ■■#0001 coo OOOOH 66666 fflMOOM OOOOO Sg8§g O ^ Ol 01 CO qqooo tH 01 CO ■<* 10 co oi co -h o O-IHCl CO qqoqo 66 666 N LO 00 o n qqq q q 66666 1-1 (N co -# m 00c o^oooo o co co o to 00006 g|tLOgg qqoqn co t- co o o KQOrtW n ci n 5 S t- co co q q dp'odri C0-+001-1 LO CO t- t- 00 6 6666 -# o o UO C£ 2S22S CO 01 -H !>• O UO CO 01 L-j CO CO CO -# -# -* 32222 Th CO 0-1 CO o 01 01 CO CO CO - :i ri ?'] : §§Sii 00000 CO l^ CO CO O S383S oqqoo LENGTH OF WIRE AND DEFLECTION. 211 tocootoco 3§3838 OHNNM t-*H80 t- lo m o oo d^d coco i-IOI C0<# 13 31 I3 00OM d th oi oi co 1-131 C0rt< 13 1-H31 O0-# 13 lOOLlOO i-l 3) CO -# 13 13 3 l- O 13 Tf Oi CO 00 » d h q S i> ffl CO C- CO COO CO CO r-J OS L- CO t~ CO C5 O CO 13 X O 01 i-i co t-j t- oq CO CO tH -# >3 §1111 M3gO« i-i oi 3! oi eg TH CO C5 i-H ■* cot-oacid OI 13 t>; CO H CO t- 00 00 O OOOOO Ills 0 ^ 13 CO ^ 01 -+ 10 CD bhi- jo x» i-i 31 CO -* 13 i-i OI CO -* 13 13 O 13 O CO COrHCO CO t- *j5 13 13. d co i-i oi ot -r 13 ;! 3 ~. $§ ^co^ CO O CO 00 O OS 3J ■* !>; O oi co co oo •* CO CO i-J CO CO 01 31 00 CO CO t- CO OC1-C CO 13 00 © 35 Ah 31 CO *# 13 GO 31 ^ t- O £hco°> CD 00 888.SS. • ~:. '- i. t2cOC5oS i-t rH i-l c-1 31 cocdSt??? oo ■* 13 d t> 2||3S 3113C0O00 C3CDC^C0CD IS§1S §1111 31 13 00 O 00 CD CO 3 CO 13 00 00 Tf CO -* oo oo d d d S522S 13 13 13 d d 0 O CD 00 CO 31 00 -# O ^cqoqSoo ooooo 212 PROPERTIES OF CONDUCTORS. H to M : to - s s SB is A H g «4 ft o H H n coeioi-H i-h Iddososci 38S32 doooocot^ x oo oo oo t- rlOCOO COCO CO CS Tt< rHOOlOOl tJ i>: d «o CO cococ- o-# rHCS t> COtF CO id id lO id 222gg^ COCOOCSO id id -* ■<# ■* n q a i> w £833 CO CO CO CO CO cioioioioi t- -jj CO CO CO HOO-.OSN COCOoioioi ^ si <*> :* «> «£ oi co © io -h co c« oo ■# HOOOOt- CD CO LO LO -f -t1 CO CO CM Ol m'NHi-"h HHrtrtrt rt'r-irHTHr-? 33S3S 83S38|8S3g88 Klt^Nltnto Ico co co co co id id id id id in t^t>t-cococococococoUoiomioioio cS-jfci °< cn S3 8 o ^-*tJ- iq •* c-j t-j o os oo ) CO OS CO-* iid>*n5^ i N a n h > CS 30 c- t- > oi oi (N oi Ht-n»io I N (M N N O-. I > © © © © © °-«>8gg odood q rjj o iq o ^ridoiq OICNOI O HOOSMOO t-t-OOLO 2% 0^ W -H lO CO — < CO CO q o o cs os HHHOO OS t^ LO •# OI odood • mcoooocc ■* oi o oo co I P.cqqqqcc © © © © © C COOIOOSOO t- CD lO -cK CO i-OiO 1-0 -H ■# -f -cM -t< -V -f i ^^Jh^oc ■■^■MHMI^I^^ DEFLECTIONS OF WIRE. 213 o ©©©© CO CO § CN © Tfi id © t> 00 QOOOU1 'rHCN COCO ooooo © o © © o z> ©©©© = © © © © D CN 00 •+ © ' rH i-i CM' CO ©©©C© ' rH t-i CN # id © c- © © © © © co •* id id © 11111 CO -# •*' id © ©© ©© © cd co ■* -#' id 33838 © W © LO © i I ,- ? HNNNM ©©©©© ©© © ©© © © © © © © © S l> CN ©©t>t^c6 ©©©©© S © S © lo © ©©©© i- © i- © ■-. oi-^oqcot^ ■* id id © © ©©©©© 1~ © 1.0 © 1- 00 CN iq © CN CO "#■*'"* id OO©©© © 1.0 © i- © Ifl t>; © CN iO A HNfifi S3g§S CN «# © 00 © HrtrtHN ©o©o © i-H rH i-l CN CN i-l CN CO -# lO ©©© ©© c © © © © 00 © -V CN © ©©©©© © t- »q h c4 6 os os © r-j © ©© ©© ~ ~ ~ ~ § CO 00 OS OS © ©©©© © Iffl UO © © t- ©©o©c -*ididid« ©© ©©© © i.O © 1 0 © ocMist^q ■* A ■* <# id © © © ©© co©© ©©©©© 214 PROPERTIES OF CONDUCTORS. # id co oo oi ooooo S3888 h e4 co •* co ISSS t-OOOOd ooooo SP92S tMCN CC#m t-i o CO r-t O 00 L- co co co on CO 0.1 0J LO c © ci oi x' x cj p :i I- :: x -t © :; l^oco quoco © t> ■* co >o 3 -s q: :•:- u- CN OI OI OI 01 co r ". :o © 01 -+1 Ci . Ci £ 10 "+ ■* CO 00 00 00 § joi "? ^ if' -*■ — co : ir '•"■ -" [ ONOtHCO ~ 'C- 2 ,x ,'i co t- o CO CO o-pqoq SSSScS 01 © L-; cn in § h o o ci MHOl CO © ci CO t~ *3 2 a te I1 .- ^OOOiOCl >-; co lo i> to onoioK lilll oi o lo m o ~. ™ L" '-'-. - (0 0 1 X 1.0 0 io" 01 CO CC 01 X CO X CO © ■# ■* LO 00 Ci O^COOCO P S te S £3 - i' ," '- sS^coi O — 1 CD © 00 © OI 00 © LO LO LO LO LO TH ©00©©00 O0 00 00 CO CO goiolSS L0 LO © L0 LO © •+ £.-r c CO CO © CO CO : j b" x co ~ ©lOtH ©© © 00 © 00 © 00 C^ CO © '"= = ="i r- ~1 : ■■ •: i ~ oi 00©COCi zi ^ co go © ©CiCOCp © t- © I> © © -+t t^ r- ItH © 00 ■* t- CO co© t-o<# CO CO CO CO CO ©©©©C0 0( i3gc*S i o; -r oi oi ©-H©C0© oi •* co lo co c. c. x Ol g LO CO © © 00 CO CO © 10 X — i0 CO ■^olol COLO CO t-^© ggocgco ^-cH 00 00 00 -1 01 01 c : ^ L- op Ci © . 1 cs © S £J ^ ©' CO 00 LO © ©' ' CO © © © LO © OJ C_l -+ X 0( • oo io © o in 5 ©" oi r-5 w I ©CO©© ^ ^ •* LO CO © © CO © CO © i : io ^- co '-. ) 00 CO CO © 1-B co^ £322©-!= 216 PROPERTIES OF CONDUCTORS. .2 0) O US © rH s q § CO 8 1 CO o q « s q q 1 8 •^98, i ui Sl reclg C co 13 :■' " oi :■: OOOOC ©O-H rHrH go t- co <3 -t --. i- — i- - thOItJhuoc: ©31 CO lOH 00 CD -* CO rH 13 i- t>; CO © CO L- 13 ■* oi o co o 13 co 83mSS CO t- rH 13 CO -# CO 31 r- © CO 01 i-i © © £2 £■ ^ « 55 © © 01 -t- 13 3^8 ^3 CO CO 13 CO rH OH^CO* qqq q© dddod 88£88 13 O 13 q 13 C-i <3 t^ © 35 CO CD © 01 13 CO CO O CO CO OOHrtH 13 d i3 d gi o o o o o rH 31 CO 13 CD qqqoo de'eidd SK88S 33S q t- CO 13 CO OCO CO O CO oi oi oi co c- OO ooo difloia'd 10 l_- © 01 o CO-* 13 13 CO O rH CO 13 CO O OOOO dodoo gjg§§§ ooooo tOt-MOO rHOcOCDiO C5COCOOCO © © go j> © 5=C« 0*1 CO O rH 00 13 13 O 13 d 13 t- q oi 13 l- - l~ ~ l~ ■'■ •v oi rid CO co -r oi o l- 3. H CO 10 CO rH 31 31 31 31 co oi co q 13 rH O CO CO 13 os q o h oi 6l 31 31 31 CO rH © 00 CD 13 OO O rH31 .: 0 CO CO o- OOOrtMlt wbhi;q loo3».. r* 13 13 13 CO ©©CO© _ qqqqq t- O 31 13 ^ rSgSSel qooq^ odd d CO 13 CO t^ CO TABLE OF STRAINS AT CENTRE OF SPANS. 217 rtCOQtSOO CO 00 l~ © UO L~ IQ CO rH Ci HWLOOOO CO CD © CO © d 06 oo t~ © •h co ia t> o CO © 8 CO § 00 © © CO © id d d d d (M IC t- O CM N O 00 rt N LOO«HO oq i>- lo >* oi CM O 00 rH CO i- CO 10 — X cooo:;o rH ci co d cd co co o ci lo ?5^5? o q S S oq d © d d d 10 c l0 c lO co© © co© ct co co r: to oi lo t~^ d oi CO 01 X lO — 1-CTHtNOO ills? © d •# ■* co i- CO L0 l- c/ CI t^lOCOr- d d oi io y ^ •- ^ ~. o CO CO OI ■* f ~ = '/ '.- ", co .-. - ' i o t^ i-l th CM CI O -i oi ci oi co ©■* ,-tc5 t- tj5 d d i-i t> CM 01 OI CO CO ©©co©© 0 g CO IO CO d oi co d t-^ 10 ~ CO L- i-i 01 CM CO CO TJH ~ co co co H 01 t- OI CO CO f-MOOCl CO "# LO IO CD CO " CO 10 CO ©COCO t-CD £©:§£©■ oi oi rn o d CO oi OI CM CM CD O CO CD O © 0 1 -t- CO 00 COCO CD^CM t-h co d d ci C1 lo ■* c4 HMLO I- © rH CO W CO CO os oo t^ jo jo ?..' j';?',? "oi^ 55 -2 2 r-5 ■; ©oo©©co oi in co ©' co © Ol CO -)■ rH TH:ot --2 co CO © OS Ol LO CO © OS CO © O0t-©LOrH WgHOO ti5 od s4 © © I- 10 0.1 CO t- 3335 oo©ioco-* lOHt^mo; i£thi>coq6 CO CO © CO SO lo - © 'ol x OS © oi ©' id 00 © -f © © L- © © © O CO CM r-H © OS ^ CO i.O oi CO © rH LO © •* © rH LO © -H 00 © I- id Ol rH O O © O rn o =s : i © 92 ?? 3! ^ S h © 00 L- LO LO © © I- CO rH CO LO I- © LO © L- CO © '-'- t ; © t- CO © © ©© CO ~ © SSoo'S©' (OOOTHTjHtd LO CM © t- LO 11111 -ji © © rH t^ OS rH Ol rH lO — © X © 10 0OSLO©© 3HCOLOH rH t- 01 00 rH 10 •+ -J CO CO lo oi os ©* oo x co i-oi ..: x co i - o i - © © T-H •<* « © © ■* rH Ot ■H 00 ■* © ? -J os t^ lo oi © © 00 © © rH © CO © LO ■* © m th t- -#COrH©00 CO © L0. © © t- LO rf 00 rH 00 tH LO © t- © LO rH g 00 "I S3 gig © rH OI © rH OI •# © 00 rH - ir © © © ©CO©©© CS ©' rH r- © LO rH © Ol 1^ CO © rH O0' -H© t-OS© CO © © rH © §8S8g 0 © © CO © - io © !- co © co © © eo oi lo co © oo Ol 01 Ol 01 O -H © t- 0- o©oo© o V&$'< Ol — I © CO t- - i •- - co i ■ : © i - i - i ■ .5 x> oi © c 01 CO 10 © -/ "' Ol Ol Ol c< © CO LO © < © 00 i0 C l L0 -H Ol rH C rH © 00 © L0 -H © LO -H -- 00 rH GO © © COCSTtH©©" HHCICICM LO © 00 CJ1 rH ■IBQjl m streds © t~C0©© ©©©oo BREAKING WEIGHTS. 219 jtfotes. Comparative Resistance of Woods (Addenbrooke). The measurements were made along the grain by inserting terminals two inches apart in sound, dry, well-seasoned pieces of the woods, each piece being 3" x §" X I". Other tests across the grain gave results from 50 to 100 per cent higher. Wood. Megohms. Wood. Megohms. Mahogany ... 48 Lignum Vitae . . . 397 Pine . ... 214 Walnut . . . .478 Kosewood . . . 291 Teak . . . ' 734 BREAKI9TG WElGfHTI COPPER AND SIJLICOI¥ BROAXE WIRES. Breaking weight hard-drawn copper wire per 1000 C. M. = 47.12 lbs. Breaking weight soft-drawn or annealed copper wire per 1000 C. M. = 26.69 lbs. Breaking weight No. 0 B. & S. hard-drawn copper wire = 4973 lbs. Breaking weight No. 0B. & S. soft-drawn or annealed copper wire = 2817 lbs. Breaking weight silicon bronze wire per square inch, 80920 lbs. Breaking weight No. 4 B. W. G. silicon bronze wire, 3600 lbs. Horse-power lost in copper conductor at a density of 1000 amperes per square inch cross section, is equal to the number of thousands of cubic inches of copper -+- 10%. — By Prof. G. Forbes rzheensioxs of cross arm§. Regular size, 3J inches X 4J inches, IJ-incTi holes. Special size, 4 inches X 5 inches, U-inch holes. 2-pin, 3 feet long ; 4-pin, 4 or 5 feet long ; 6-pin, 6 feet long. CABLE TESTING. Cables — I nderg-iound ami Submarine. The majority of the methods of tests and measurements given herein are applicable to both aerial, underground, and submarine cables. I nsulation Resistance, Direct Deflection iVKethod, with Minor Galvanometer.— This method, Fig. 1, is generally used in this country in underground and submarine work. ^XXXXXXXXXXXXXXXsS CABLE Fig. l. a and b = leads. G = galvanometer, Thomson or D'Arsonval, mirror type. £zr shunts for G, usually i, TJ-5, TCfJn0. B = battery, 20, 50, or 100 chloride silver cells. R = resistance box of 100,000 ohms. BK=. battery reversing key. SK=z short-circuit key for G. First connect a to lower contact point of SK. and take constant of G, using iooo shunt, and small number of cells, say 5 (depending upon the sen- sitiveness of G), with standard resistance R only in circuit, b being discon- nected as shown. If 5 cells are used in taking constant, and 100 cells are to be used for test, Constant : G deflec. X shunt X R X 20 = megohms. 1,000,000 After obtaining the constant, measure insulation resistance of lead b, by joining it to SK instead of a, disconnecting the far end of b from the cells. The result should be infinity ; but if not, deduct this deflection from the deflection to be obtained in testing the cable proper. Now connect the far end of b to the conductor of the cable, the far terminal of latter being free. Then open S K carefully, and observe if there are any earth currents from the cable. If any, note deflection due to the same, and deduct from bat- tery reading if in the same direction, or add to it if in opposite direction. Short-circuit G with SK, and close one knob of B K, using, say, the j%n shunt. After a few seconds open SK; if spot goes off the scale, use'a higher shunt. If deflection is low, use a lower shunt. After one minute's electrification, note the deflection. The result may be worked out from this reading, but the current should be kept on for three or five minutes longer, and readings taken at end of each minute. The deflection should decrease gradually. At the end of the last minute of test, open BK, and allow the cable to 220 CABLE TESTING. 221 discharge fully. Then close £A~ and press the other knob of BK, revers- ing the battery. After a few moments, open SK, and take readings of deflec- tions as before. The insulation resistance where d is the deflection at a gi constant ren time, and S is the shunt used. If no shunt is used, Note that in the above constant, the ordinary constant is multiplied by 20 for the reason that the battery is increased '20-fold, or 5 :: 100, In case the same battery is vised for testing as for obtaining the constant, then constant : G detlec. X S X R 1,000,000 If there be no earth currents, the readings with opposite poles of battery to the cable should not vary appreciably at any given minute. Pronounced variation between the readings at given times and unsteady deflection indi- cate defective cable. Insulation Itesistance l»y Uletliod. of 1a»mm of Cliarg-**. The insulation resistance of a cable or other conductor having considera- ble capacity may be measured by its loss of charge. Let one end of the conductor be insulated, and the other end attached to an electrometer, in the manner shown in Fig. 2. ELECTROMETER Fig. 2. Let It =. Insulation resistance in megohms per mile. (,'= Capacity in microfarads per mile. E = potential of cable as charged. e =z potential of cable after a certain time. Depress one knob of key K, and throw key K' to the right, and charge the cable for one minute; then throw key K/ to the left, thus connecting the cable to the electrometer. Note the deflection E. Noting the movement of the spot for one minute, take reading e at end of minute, then 26.06 4 ~~ C log E If an electrometer is not conveniently at hand, use a reflecting galvanom- eter, and after charging cable as before, take an instantaneous discharge, noting deflection E due thereto. Recharge cable as before, then open K' and at end of one minute, the galvanometer having been disconnected from cable in the meantime, take another discharge-reading of cables, and apply the same formula as before. If a condenser of low capacity be inserted be- tween K' and the galvanometer, the latter need not be disconnected. The advantage of the use of the electrometer is that "the actual loss of potential of the cable may be observed as it progresses. 222 Testing- Joints of Cables by Clark's Method. In the figure (Fig. 3) the letters refer to the parts as follows : Fig. 3. G is a high-resistance mirror galvanometer. S is the shunt. /.", is the short-circuit key. It may he on the shunt hox or separate. K,/ is a reversing key. KN/ is a discharge key. B the hattery, usually 100 cells chloride of silver. C is a I microfarad standard condenser. The joint to he tested is placed in a well-insulated trough, nearly filled with salt water. A copper plate attached to the lead wire is placed in the water to ensure a good connection with the liquid. The connections are made as shown in the figure, one end of the cable being free. To make test close K /n for a half minute; then release it (first depressing one knob of key K„), thereby discharging the condenser C, through the galvanometer, and note the deflection, if any. A perfect piece of cable of the same length as the joint is then placed in the vessel, and if the results with the joint are practically equal to those obtained with the perfect cable, the joint is passed. When the deflection is very low, it is evident that the joint is sound, and it may then be considered unnecessary to compare it with the piece of cable. It is very important that the trough and apparatus be thoroughly insulated. Electrometer Method. — This method possesses the advantage that it dispenses with a condenser, and thereby avoids possible misleading re- sults due to elective absorption by that instrument. The connections for the electrometer test are shown in the accompanying figure (Fig. 4). ELECTROMETER Fig. 4. B is a battery of about 10 cells. B, is a battery of 100 or more cells. As in the preceding test, it is here highly essential that the insulation of the trough should be practically perfect, or at least known, so that if not perfect, proper deductions may be made for deflections due to it alone. To test the insulation of the trough, depress Kt, and close switch S. This CABLE TESTING. 223 charges the quadrants of the electrometer, and produces a steady deflection of its needle, and shows the potential due to the small battery B. Now open switch S, still keeping K, closed, and watch the deflection of needle for about two minutes. If the insulation of the trough is not perfect, there will be a circuit, so to speak, from the earth at the trough to the earth shown in the figure, and a fall in the deflection will be the result. If, how- ever, the drop of potential is not more than is indicated by a fall of two or three divisions, the insulation of the trough will suffice. The electrometer is discharged by closing switch S, which short-circuits the quadrants, K, being open at this time. The joint is now connected as in the figure. Switch .5 is opened, and key Ku depressed, thus charging the joint with the large battery Br This produces a quick throw of the needle, due to the charging of the joint. Next, keeping Ktl closed, discharge the electrometer by closing switch S for a moment. The switch is then opened, and if the joint is imperfect as to its insulation, the deflection will rise as the elec- tricity accumulates in the trough. The deflections are recorded after one and two minutes, and are compared, as in the previous test, with a piece of perfect cable. The results obtained with the joint should not greatly ex- ceed those of the cable proper. IMrect Reflection Iffethod. — The insulation resistance of joints - may also be tested by the direct deflection method already described, and when great accuracy is not required, is preferable, owing to its simplicity. Capacity. Capacity tests are usually made by the aid of standard condensers. Con- densers, or sections of the plates of condensers, may be arranged in parallel or in series (cascade). Arrang-ement of Condensers — Parallel. — Join like terminals of the condensers together, as in the figure ; then the joint capacity of the condensers is equal to the sum of the respective capacities. Capacity, C= C + C, + C„ + Ctll. mi mi iiu liu Fig. 5. Condensers in Series or Cascade. — Join the terminals, as in Fig. 6. The total capacity of the condensers as thus arranged is equal to the reciprocal of the sum of the reciprocals of the several capacities, or Capacity in series = 1 Fig. 6. Condensers are now constructed so that these two methods of arranging the plates of a condenser may conveniently be combined in pne condenser, thereby obtaining a much wider range of capacities. 224 Testing Capacity l>y Direct Discharge. — It is frequently de- sirable to know the capacity of a condenser, a wire, or a cable. This may be ascertained by the aid of a standard condenser, a trigger key, and an astatic or ballistic galvanometer. First, obtain a constant. This is done by noting the deflection / = flux from one pole, then in a Series wound multipolar dynamo ; „ rev. X p X n x $/ E- w * _ E x io8 rev. x p X n In a Multiple wound multipolar dynamo, rev. X n X $>, E~ W AETEIllSfATiarO CURRENT MACHINES. For alternating or periodically varying currents there are three values of the E.M.F. used, or of which the value is required : a, The maximum value, or the top of the wave, 6, The instantaneous value of a point in the wave, c, the virtual E.M.F., or Vmean2 value of the full wave. 230 alternating' current machines. 231 In addition to the symbols used for continuous currents, let k= a constant varying from 1.1 to 2.5 depending on the relative widths of the armature coils and pole-pieces, usually taken as 2.22. 0 = angle through which the armature coil is turned at the instant taken. Then, for single-phase alternators, maximum E = 2tt x n x 3y X rev. X p 108 In this case n = number turns in series, and $ = maximum flux enclosed per turn, , E max. x 108 and $, = 2tt x»X rev. X p -r, „ 2?r x » X $/ X rev. x p X sin 9 Instantaneous E ■— — ' = maximum flux enclosed per turn, &/ = 7\ E inst. X 108 Virtual E = 2tt x n x rev. x p X sin 9 rev. x p X k X */ X n In this case n = number of conductors joined in series with one another around the armature, _ E vir. X 10s ' rev. x p X k x n For multiphase alternators n = the number of conductors in series in a phase, and in two-phase ma- chines the E.M.F.'s of each phase would be the same as in a single-phase dynamo. In three-phase alternators the E.M.F. between terminals will depend upon the method of connecting the armature conductors. The two most common methods are called the delta connection and the Y or star connection, both shown in the following diagrams. DELTA CONNECTION Y OR STAR CONNECTION Figs. 1 and 2. , Values of E.M.F. in three-phase connections when x=y = z. In the delta- connected armature the E.M.F.'s between terminals are those generated in each coil, as shown in the diagram. In the Y-connected armature the E.M.F. between any two terminals is the E.M.F. generated by one of the coils in that phase multiplied by the V3 or 1.732. Two-phase circuits are sometimes connected as a three-phase circuit ; that is, both phases have a common return wire. In this case the pressure, be- tween the two outgoing wires is V2 x E, and the current in the common return will be / V2, both conditions are on the assumption that E and / in each phase is the same. 232 DYNAMOS. V=z the E.M.F. at machine terminals where E = total E.M.F. generated. Then, in alternators the E.M.F.'s are shown in the following diagram, the load of the alternator heing non-inductive, and the armature reaction being neglected, 2n n LI X Ir = V at machine terminals. V=^(2n?iLI)2-{-(l7')2 when L =z coefficient of self-induction, r = re- sistance of armature + external circuit. Continuous Current Machines. The current in a dynamo depends upon a. Its E.M.F. b. The resistance of its internal circuit + the resistance of the external circuit on which it is working. c. Any counter or opposing E.M.F. in circuit, such as storage batteries being charged or motors being run. Then let sE — total E.M.F. of the dynamo, e = counter E.M.F. of the circuit, R = internal resistance of the dynamo, r — resistance of external circuit, 1=. current in amperes flowing. Then if the external circuit have no counter E.M.F., as when supplying cur- rent for incandescent lamps, E /=^nr7 = amPeres or, if a storage battery is being charged and its opposing E.M.F. — e r E~€ then I = -g—. — If E, — external E.M.F. of dynamo as measured by voltmeter at brushes at the load in question fhor. r— ~L Alternating- Current Dynamos. In alternating-current machines another factor in addition to the resist- ance of the circuits, internal and external, tends to retard or reduce the current, viz., the reactance of the circuits (see index for reactance and impedance). Let L = coefficient of self-induction of armature, L' = coefficient of self-induction of external circuit, n = number of cycles, ~~ to rr 27T h, EQ = open-circuit voltage of alternator, other symbols same as for d.c. machine, then, reactance = w L ohms, impedance = Vi?2 +(&> L )2 ohms, In A.C. dynamos E = Vmean2 = E 233 OTE, = l{r*-{- aI/>) —y B + rz-{- o>I/ + +<^f *& X 7t tuvn^ x" 10— ^ The inductance L of a circuit in henrys is the ratio — ? rr L I (c.y.s.) max. $V}(X l(V-8 or if 1 is expressed in virtual amperes then L = — ■ n $ = L I V2 108 and the E.M.F. of self-inductance is E = V2 it n * v 10-8 where v =. cycles per second, or E = 2?r v L I volts. If to =1 2ttv, to!/ = reactance of the circuit in ohms, and the E.M.F. of self- inductance of the circuit is = M=. ImL = reactance voltage. Energ-y in Balanced Three-phase Circuit. In the following diagram of a Y connected multiphase generator and cir- cuits, let ey = E.M.F. of any phase in the armature, iy = current of any phase in the armature, E = E.M.F. between mains, /= current in any main, w, = energy of one phase of the armature, JV= total energy, Wy = C, 1/ but E = e, V3 I=i. W—2>w,- V3 In the following diagram of a delta connected multiphase generator and circuits, let e2 = E 7=i2V3 w% = e2 i2 W=3w* = ^|/= 1.732 E I V3 ~ 1.732 E Therefore for any balanced three-phase system, the energy is equal to the product of the E.M.F. pIG 5 between any pair of mains and the current in one main, divided by V3 ; the result being multiplied by the cosine of the angle of lag ; i.e., the power factor. 234 DYNAMOS. then, It •— resistance per leg of Y-connected armature, r =z resistance per phase of A connected armature, I-R loss in Y-connected armature = 3 I2R I2Rl 3 in A connected armature = Energy in Three-phase Circuits. Jj ■=. current in any one of the three -wires of external circuit, i =z current in one phase of the armature for delta connection, W — watts output of a balanced three-phase generator, 1.732 = V3 .577 = 1 ~ V3 E = volts between terminals (or lines) on either delta or Y system, v = volts of one phase of the armature if connected in " Y," R — resistance per leg, of Y connected armature, r = resistance per phase of A connected armature, W— 3 I, v = — '.—— I' E 1.732 (either with Y or A armature. V3 For A = »^- v3 for A v, = E — - — - = 1.732 E /, which shows statement in brackets to be true. V3 ir- W W- v W- ' E X 1.732 V, = 1.732 i in delta system. I2R loss in Y connected armature = 3 I/Ji. I2R loss in A connected armature = 3 ( — /- ) r W3/ : I,*r. ■ : 5 e,j E, E, E b E Fig. 6. ik i, E E=E, E=^E/= 1.732 E,. I AMPERES =1. 732 XZW / AMPERES = 1.732X2 or y /AMPERES = 1. 732 X V or X I AMPERES = Z 3i Z AMPS. <^ V AMPS. / AMPERES =V i2T I AMPERES = 05 Delta Connection. Star or Y Connection. FIGS. 8 and 9. Values of current in three-phase connections, where x = y=.z. CURRENTS. 235 Direction of Current in a Conductor. To determine in which direction the current in a conductor is flowing, place a compass underneath it. If the north pole of the needle points to the left, the current is flowing forward or away from the observer. With the compass above the conductor, if the north pole of the needle points to the right the current is still flowing away from the observer. These results are often shown as in the accompanying cuts. Direction of Current about an Electromag-net, and location of its Poles. If the direction of the current flowing in the wire of the coil is not known, then with a magnet find the north pole of the magnet, by approaching the compass to one of the poles ; the north-pointing pole will be repelled by the north pole of the magnet, but attracted by the sonth pole. Then by placing the right hand on the coil, with the thumb extended at right angles and pointing in the same direction as the north pole of the core, current will be flowing in the direction pointed by the fingers. Of course, if we know the direction of the current, and wish to find the north pole of the magnet, placing the hand on top of the coil, as above, with the fingers extended in the direction in which the current is flowing, the north pole of the core is in the direction in which the thumb is extended. Another way is to look at pole of magnet. If current is going round right- handed you have a south pole ; if left-handed, a north. See " Corkscrew " Rule. ""fcCTfON DYNAMO •. RIGHT HAND. Fig. 12. Direction of Current in a Dynamo Armature. A simple rule is : facing the commutator of the dynamo, speaking now especially of the bipolar type, and assuming the left pole to be north or -f-. and the armature to be revolving counter clock-wise, then the current is flowing to the right across the face of the armature, or the left brush is positive, or the terminal from which current will flow, returning by the negative or right-hand brush. Reversing the direction of rotation will reverse the polarity of the terminals. The accompanying figure illustrates a graphic method, called Fleming's Right- hand Rule. 236 DYNAMOS. Direction of Rotation in a motor. Knowing the direction of current in the circuit, or which is the posi- tive and which the negative terminals of the circuit, the direction of rota- tion of the armature can easiest be determined by use of the accompanying diagram (Fig. 12), which is called Fleming's " left-hand rule." field magnets. In the paragraph on the E.M.F. of dynamos, preceding, the symbol $ is used to indicate the total flux or quantity of magnetic lines forced through the core of the armature by the field magnets. This value of course depends upon the degree of excitation, i.e., the amount of current and number of turns of wire on the field magnets. To determine this value in an existing ma- chine, run it at a proper speed, and measure the E.M.F. with a voltmeter. Then E x 10s MOTOR LEFT HANO.v — for continuous current machines. C rev. X n Fig. 13. and $rr — r for alternating current dynamos. rev. xn xp Xk and if (& = magnetic induction, or Gauss = lines of force per square centi- meter, and A =z area of cross-section of armature core in square centimeters. Then density of lines in armature = (B = — . Magnetic Circuit of a Dynamo. The path over which lines of force flow, be it iron or air, is called the magnetic circuit, and is subject to laws analogous to those for electric con- ductors. It has its magnetio resistance, which is directly proportional to the length of the circuit, and inversely proportional to its cross-section and permeability, the latter being somewhat analogous to conductivity in an electric conductor. In a dynamo the path through field-magnet cores, pole-pieces, field-yoke, air-gaps, and armature core, forms the magnetic circuit of tbat machine. The calculation of its value follows well-known laws, and is as easily car- ried out as the calculation of the resistance or conductance value of an electric conductor or path. In any piece of iron Let I = length of the piece. s = cross-section of the same, ju. — permeability rr (B -f- 30,, In the magnetic circuit of a dynamo let Aa = area of cross-section of armature core, Ag = area of cross-section of air-gap under the full pole-piece + a per- centage for fringe. Am = area of cross-section of magnet core, AP — area of cross section of pole-piece, Ay — area of cross-section of yoke, I = length of any part, $ — total flux, 237 Total reluctance = (-^-) + (-2L) + (J*-\ + (J*-) + (JlU) call this total reluctance 7»',n, Then ._1.257X»X / i?m total flux through magnetic circuit, and ?i = number of turns of wire, and /—current in amperes. Application of Magnetic Circuit to Dynamo Design. Let (B = flux per square centimeter, then in any part of tne magnetic cir- cuit of a dynamo, /T> $ KiD=z~, and after it is decided at what induction it is best to work the A iron of the circuit the cross-section (B The armature core is invariably of laminated soft annealed wrought iron or steel, while the magnet cores and yokes are often of cast iron, although most generally to-day some part, if not all, of the core is of mild cast steel. If cast iron is used, it is only necessary to increase the cross-section to satisfy the equation Experience has shown that there is a very considerable leakage of lines of force in an electro magnet ; some cutting across without going through the armature path, others leaking across corners, etc. This leakage, amounting to 30 to 50 per cent of the the total flux, has to be made up by increasing the ampere turns of the magnets beyond that necessary to furnish the requisite flux for the armature part of the circuit, by a percentage or amount repre- sented by the leakage. This leakage has been determined for different types of field magnets by Edison and others, and a table of such values follows. In dynamo calcula- tion the leakage value may be represented by v. Stray Field in Dynamos. Name of Dynamo. Field. Arma- ture. Remarks. Value of v. Edison-Hopkinson Bipolar Drum Poles next to bed-plate 1.32 Edison (American) Bipolar Drum Poles next to bed-plate 1.40 General Electric Co. Multipolar Drum Direct driven 1.25 Kapp Bipolar Drum Yoke next to bed-plate 1.30 Siemens .... Bipolar Drum Yoke next to bed-plate 1.30 Manchester . . . Double magnet Long Bed and one pole cast 1.49 2 pole ring together Ferranti .... Double magnet core- Ordinary pattern alter- 2.00 Multipolar lessdisk nating. The following formulae are useful in calculating approximately the mag- netic leakage in a dynamo : 238 DYNAMOS. 1. The permeance, or reciprocal of magnetic reluctance, between two parallel opposed surfaces is A±A2 where d is the distance between the surfaces in centimeters, and where At and A2 are the areas of the surfaces in square centimeters (see Fig. 14). Ai -dHA2 Fig. 14, Fig. 15. 2. The permeance between two equal rectangular areas situated in the same plane, having corresponding sides parallel and a common axis of sym- metry, is if — is large (see Fig. 15), T(D- d)) f% — d — f^, Fig. 16. where L = length of each rectangle, measured perpendicularly to common axis of symmetry (i.e., to the plane of the paper in the figure) in centimeters. d = distance between adjacent parallel sides in centimeters. D = distance between remote parallel sides in centimeters. 3. The permeance between two equal rectangular areas at right angles to one another, having one pair of sides in the one parallel to the correspond- ing pair in the other, is 2D + d (n - 2) j where d, Z), £,, and L2 are the lengths in centimeters of the dimensions shown in Fig. 17. If d r= _D, the permeance in this case becomes As the resistance of the two air-gaps in any dynamo Fig. 17. is usually more than 80% of the total resistance of the magnetic circuit, the length of the iron part of the circuit is of little conse- quence excepting in cost of material, and is determined largely by the amount and style of winding necessary for the field magnet coils. Other considerations govern the length of air-gap, such as sparking at the brushes, heating of pole-tips, heating of teeth in Paccinotti ring, regula- tion of voltage, current, etc., thus compelling the use of more magnetizing force to overcome that part of the circuit than all other parts combined. If then Jim = total reluctance of the magnetic circuit of a dynamo, ampere turns : <*> Rm '' L257 CURRENTS. 239 and, as it is necessary to know the ampere-turns required for each part of the circuit, the items may be tabulated as follows : — Formula' for Different parts of the Mag-netic Circuit of a Dynamo. Square centimetre units. Armature core ; ampere-turns = $ x —. — * — -~ 1.257 Magnet cores ; ampere-turns = Pole-pieces ; ampere-turns = Yoke ; ampere-turns z Am X ' Ap X hp h For square inch units the divisor will be 1.257 X 2.54 = 3.193, or better, Armature core ; ampere-turns z ■ Aa The two air-gaps ; ampere-turns = x -/ X -3132 Magnet cores ; ampere-turns = $ x —. ■ X .3132 An X H™ h Ap X HP ly Pole-pieces ; ampere-turns = $ X —. X .3132 Ap X HP Yoke ; ampere-turns = $ X -; — X .3132 Ay X f-y Types of Dynamos as Determined \>y tlieir Connections. There are five types of dynamo connections in common use in the United States, viz.: — 1. Magneto machines. 2. Separately excited machines. 3. Series machines. . 4. Shunt machines. 5. Compound wound machines ; this last having two classes, i.e., long shunt and short shunt. The above types apply especially to continuous current dynamos, but alter- nating current machines are usually made separately excited as per No. 2, and are sometimes made self -excited , from separate coils on the armature, connected to a commutator on the shaft adjacent to the collecting-rings. Other alternating current dynamos, in fact nearly all those used in the United States to-day for lighting, or for lighting and power purposes, that have been constructed since 1891, are of the type known as composite wound, in which the fields are separately excited from an outside source, and in addition to this a heavy wire series winding is also wound on the field coils, and a portion of the current from tbe main circuit is shunted through them, being passed through a commutator on the armature shaft first to be rectified. This current is of course in proportion to that flowing in the main circuit, and adds excitation in proportion to the load, thus keeping the terminal pressure practically constant under all conditions. Alternators for trans- mission of power are not " composite " wound. 240 Compensated Revolving- Field Alternators. The General Electric Company in October, 1899, placed on the market a new type of multiphase alternator, which is claimed to overcome many of the faults common to the old style of machine, especially when used on combined lighting and motor loads. While it has been found a compara- tively easy matter to compound and over-compound for non-inductive loads, it has been heretofore quite difficult to add excitation enough to compound for inductive loads which require considerably more field current than do loads of a non-inductive nature. The following description is taken from the bulletin issued by the makers describing the machine, which is of the revolving field type : — "The means by which this result is accomplished areas follows: The shaft of the alternator which carries the revolving field carries also the armature of the exciter, which has the same number of poles as the alter- nator, so that the two operate in synchronous relation. In addition to the commutator, which delivers current to the fields of ooth the exciter and the alternator, the exciter has three collector rings through which it receives current from one or several series transformers inserted in the lines leading from the alternator. This alternating current, passing through the exciter armature, reacts magnetically upon the exciter field in proportion to the strength and phase relation of the alternating current. Consequently the magnetic field and hence the voltage of the exciter, are due to the combined effect of the shunt field current and the magnetic reaction of the alternating current. This alternating current passes through the exciter armature in such a manner as fro give the necessary rise of exciter voltage as the non- inductive load increases, and without other adjustment, to give a greater rise of exciter voltage with additions of inductive load." Following are cuts of the types mentioned above. magneto dynamo Fig. 18. separately excited dynamo Fig. 19. / \ / \ d b J Fig. 21. 241 CONNECTIONS OF TYPE AS SINGLE-PHASE ALTERNATINB CURRENT GENERATORS WITH COMPOSITE FIELD 2300 VOLTS &s e-Go- /is 0-90-900 r( *s a-iao-soo r< Be \Col lector s'de Leach succeeding spool, pra 24. — General Electric Composite wound alternator. 242 DYNAMOS CONNECTIONS OF TYPE AS SINGLE-PHASg ALTERNATINB CURRENT GENERATORS WITH COMPOSITE FIELD 1150 VOLTS y?se-60 **S 6-90-9-00 Foi jqs s-120-900 rQi Stationary Shur -nutator-Col lector L toward the observer.The arrows correspond to U— !j (H__JfJ those on spool f I anges. the spools being so placed : — that the arrows point in opposite directions on Collector side^ each»succeedlng spool. Fig. 25. — General Electric Composite wound alternator. CURRENTS. 248 CONNECTIONS OF TYPE AS SINGLE-PHASE ALTERNATING CURRENT GENERATORS WITH COMPOSITE FIELD E300 VOLTS AS 14-120-1070 Form A. , r— The observer- Is supposed -to be lookinjg ♦I Sir face of Pole Piece marked A. The series -fi A ■ w'mdfng should be nearest armat.ure,that tLJt UL toward observer. The arrows cor-responc ^rri-J^ ■ -Chose on spool flanges the spools being s< Fig. 26. — General Electric Composite wound alternator., 244 CONNECTIONS OF TYPE AS SINGLE-PHASE ALTERNATING CURRENT GENERATORS WITH COMPOSITE FIELD 1150 VOLTS AS 14-120-1070 Form A Commutator-C. inerof plaong spools. ~U II It if face of pole piece marked A.-The series fiel A winding should be nearest armature.that i Ji tUt iL *ra"d °tose™%- The *z°V cor?spond * Fig. 27. — General Electric Composite wound alternator CURRENTS. 245 •ept for alternating current uilly rinds a struct r:iihv;ty unos are separately excited constant current circuits, Magneto dynamos are now used in the United States only for ringing tele- phone bells,* and for other .signalling purp< Separately excited dynamos are seldom m production ; with the exception that one o power-house where the shunt fields of all t from one generator. Series dynamos are used for arc light) _ where many lamps are distributed over wide area The constant potential arc lamp, both for continuous and alternating currents, has reached such a degree of perfection and low cost as to encourage its use to a very great extent to displace the old style constant current lamp. Series dynamos are also often used as boosters to vary the voltage on a line automatically in proportion with load. Shunt dynamos are used for charging storage batteries, and for large cen- tral stations supplying constant potential current, and this applies especially to the " Edison " stations throughout the country. It is easier to adjust the load between large machines when shunt wound, and in these large stations attendance is always at hand. Compound wound dynamos are used in street railway power-houses, in order to keep the pressure somewhere near constant under the great varia- tion in output ; and are used to a very considerable extent, it may be said almost wholly, in isolated plant work, in order to save attendance and adjustment of the field rheostat DIIAMO GHABACIBRISTIGi. Dr. John Hopkisson is said to have devised the " characteristic " or curve of properties of the dynamo, to show the results to be expected in a certain design of machine, and to indicate actual results after completion, although it is also said that Deprez first used the name. The characteristics most commonly developed are as follows : — 1. Magnetization or saturation curve. 2. External characteristic. 3. Curve of magnetic distribution. 1. Mag'iietization Curve. — This curve is always determined for each newlype of dynamo by reputable builders, and can easily be determined by any one having available a separate exciting current, a voltmeter, and an ammeter. The turns of wire on the field remaining the same, it is sufficient to read the amperes in the field, voltage at the brushes, and revolutions of the arm- ature. Curve, Fig. 28, following shows the result of such a test. In a case where, like the above, the dynamo is already in existence, the field is ex- cited from some outside source, and the ■u curve determined by gradual increase of the current in the field, and the volts at the brushes are read after each such change. The accompanying curve is the re- sultant of trie magnetizing force neces- sary to force the flux through the following parts, in the case of a bipolar dynamo, all of which may be of differ ent character : — a. Armature core. b. Two air-gaps. c. Two polec-pieces. d. Yoke. e. And to overcome leakage of mag- netic lines. Individual curves for each of these parts can be predetermined by use of formulas for calculating the magnetic circuit of dynamos, and from a com- bination of those curves the curve shown above can be constructed, showing the aggregate excitation necessary to produce certain voltages. For sample of such a composite curve the reader is referred to page 149 of the fifth edition of S. P. Thompson's book, Dynamo Electric Machinery. This curve is valuable not only to show the character of one machine, but is useful to compare different machines by, and for that reason some stan- FiG. 28. Magnetization Curve. 246 dard ratio of the scales on which the curves are based should be settled upon. 2. External Characteristic. — This curve is a curve of results, in which the dynamo is excited from its own current, and with the speed con- stant, the terminal voltage is read for different values of load. The curves for series, shunt, and compound wound machines all differ. The observations are best plotted in a curve in which the ordinates repre- sent volt values, and abscissa? amperes of load. Series dynamo. In a series machine all the current flowing magnetizes the field, the volts increase with the current, and if fully developed the curve is somewhat like the magnetization curve, being always below it, however, due to the loss of pressure in overcoming internal resistance and armature reactions. The following diagram (armature reaction being neglected) is a sample of the external characteristic of a series dynamo. To construct this curve from an existing machine, the curve of terminal voltage can be taken from the machine itself by driving its armature at a constant speed, and varying the load in amperes. The curve " drop due to internal resistance," sometimes called the " loss line," can be con- structed by learning the internal resistance of the machine, and computing one or more values by ohms law, and drawing the straight line through these points, as shown. The curve of total voltage is then con- structed by adding together the ordinates of the "terminal voltage" and "drop due to internal resistance." A very good sample of curve from a modern series machine is to be found in the following description of the Brush arc dynamo. Following is a characteristic curve of the new Brush 125-lt. Arc Dynamo V0V-TM3E_ AMPERES LOAD Fig. 29. External Charac- teristic of Series Dynamo. 6832 ^- ^N 0500 _z ^ \ 1 \ t \ 5600 1 4500 1 3500 0U0O 1 2500 1 1 1500 CHARACTERISTIC CURVE SPEED-500 REV. PER MIN. 1000 I soo 0 1 0 7 8 9 10 11 12 13 14 Fig. 30. Characteristic curve of Brush 125-Light Arc Dynamo ivithout Regulator, D Y X A M O C H A R A C T E K T S T I C S . 247 machine without any regulator. The readings were all taken at the spark- less position of commutation. This curve is remarkable from the fact that after we get over the bend, the curve is almost perpendicular, and is prob- ably the nearest approach to a constant current meichine ever attained. By winding more wire on the armature the machine could have been made to deliver a constant current of 9.6 amperes at all loads, without shutting -' nn / -r\ / / COMMERCIAL EFFICIENCY AMPERES 9.6 SPEED £00 R.P.M. "" / 1 " Fig. 31. Electrical Efficiency Curve of Brush 125-Light Arc Dynamo. ' y y y / / COMMERCIAL EFFICIENCY AMPERES 9. S SPEED 500 R.P.M. / / f Fig. 32. Commercial Efficiency Curve of Brush 125-Light Arc Dynamo. any of the current from the field ; but this would have increased trie internal resistance, and also have made the machine much less efficient at light loads. By the present method of regulation the 1-li loss at one-quarter load is reduced from 4,018 to 3,367 watts, the gain being almost one electrical horse-power. Fig. 31 is a curve of the electrical efficiency. It will be noticed that this at full load reaches 94 per cent, which is accounted, for by the liberal allow- ance of iron in the armature, thus reducing the reluctance of the magnetic circuit, and by the large size of the wire used on both field and armature. Fig. 32 is a curve of the commercial efficiency. At full load this is Dver 90 per cent, and approaches very closely the efficiency of incandescent dynamos of equal capacity, but the most noteworthy point is the high effi- ciency shown at one-quarter load. Fig. 33 is a curve of the machine separately excited, with no current in the armature. The ordinates are the volts at the armature terminals, and the abscissa) the amperes in the field. This is in reality a permeability curve of the magnetic circuit. By a comparison of the voltage shown here when 248 DYNAMOS. there are nine amperes in the field, with that of the machine when deliver- ing current, can be seen the enormous armature reaction. The curve also / / / MW / E. M. F. 500 Fig. 33. Permeability Curve of Magnetic Circuit of Brush 125-Light Arc Dynamo. indicates a new departure in arc dynamo design, namely, that the magnetic circuit is not worked at nearly as high a point of saturation as in the old types. Shunt dynamo. The shunt dynamo has, besides an external characteristic, shown below, an internal characteristic. The first is developed from the volts read while the load in amperes is being added, the armature revolu- tions being kept constant. Adding load to a shunt dynamo means simply reducing the resistance of tne external circuit. With all shunt machines there is a point of external resistance, as at n, beyond which, if the resistance is further reduced, the volts will drop away abruptly, and finally reach zero at a short circuit. X $/ y -/f **° 1 d// i v^t \$&. z^^-^^" ! /^"^ AMPE E TURNS IN FIELD Fig. 34. External Characteristic of Shunt-wound Dynamo. Fig. 35. Internal Character- istic of Shunt Dynamo. The internal characteristic, or, more correctly, curve of magnetization, of a shunt dynamo, is plotted on the same scale as those previously described, from the volts at the field terminals and the amperes flowing in the field. DYNAMO CHARACTERISTICS. 249 The resistance line o a only applies to the point a on the curve, and the resistance value a b for that point is determined hy ohms law, or as fol- lows : As the curve of magnetization is determined from the reading of volts plotted vertically and amperes horizontally, and as r = ^- or r = ^-~ 1 o b and — r = tang a ob, therefore the resistance at any point on thecui've will be the tangent of the angle made by joining that point to the origin o. Compound di/namo. As the compound dynamo is a combination of the series and shunt machines, the characteristics of both may be obtained from it. The external characteristic is of con- siderable importance where more than one dynamo is to be connected to the same circuit, or when close regulation is necessary. Fig. 36 is a sample curve from a com- pound-wound dynamo, where the in- crease of magnetization of the fields due to the series coils and load causes the terminal voltage to rise as the load is increased. This is commonly done to make up for drop in feeders to the centre of distribution. It is impossi- ble in ordinary commercial dynamos AMPERES Characteristic of Over- FlG. compounded Compound - wound Dynamo, to make this curve closely approach a straight line, and the author has found it difficult for good makes to approach a straight line of regulation nearer than 1| per cent either side of it for the extreme variation. Curve of magnetic Distribution. — This curve .is constructed from existing dynamos to show the distribution of the field about the pole- pieces ; it can be plotted on the regular rectangular co-ordinate plan, or on the polar co-ordinate. The following cuts illustrate the commonest methods of getting the data for the curve. With the dynamo running at the speed and load desired, the Fig. 37. Fig. 38. pilot brush, a, in the first cut, or the two brushes, a and b, in the second cut, is started at the brush x, and moving a distance of one segment at a time, the difference in volts between the brush x and the location of the pilot brush, a, is read on the voltmeter. Where the one pilot brush is used, the total difference between that and the origin is read ; Avhile Avith two brushes, as a and b, which are commonly fastened to a handle in such a manner as to be the width of a segment apart, just the difference between the two adjacent segments is read, and the total difference is determined by adding the individual differences together. 250 DYNAMOS. In taking the distribution curve on a commutator, with the two-brush, method of S. P. Thompson, the curve of potential maybe plotted in two ways, viz. : the heights of the ordinates may be made equal to the sum of all the readings to the given point, or they may be made equal to the reading at each bar, in which case the curve will indicate the value of the induction at each point of the field where a reading is made. Potential curves of this kind are often plotted on a circle, the circle itself representing the commutator, with the segments plotted as radial ordi- nates, which are made equal in value to the readings of the voltmeter brushes. ARiWATlKES. Armatures for continuous current dynamos differ much in practice from those used for alternating-current machines, although the former produce alternating currents that are rectified or turned in the same direction by a commutator. Direct-current armatures are divided into two general forms,— drum arma- tures, in which the conductors are placed wholly on the surface or ends of a cylindrical core of iron ; and ring armatures, in which the conductors are wound on an iron core of ring form, the conductors being wound on the out- side of the ring and threaded through its interior. Another form used somewhat abroad is the disk armature, in which the conductors are arranged in disk form, the plane of which is perpendicular to the shaft, and without iron core, as the disk revolves in a narrow slot be- tween the pole-pieces. Armature Cored. In some early dynamos cores were made of solid iron ; but the heat from Foucault or eddy currents was found so excessive as to endanger the insula- tion of the conductors, and the loss in the core reduced the efficiency greatly. Iron wire wound on a frame constructed for the purpose was then intro- duced in place of solid cores. This answers the purpose for ring armatures fairly well, but there is considerable waste space, as round wire is always used. To-day armature cores are invariably made of thin sheet iron or annealed soft steel from .015 to .025 inch thick. In order to prevent Foucault currents in such laminated cores, it is necessary to insulate the disks from each other in some manner. Very thin tissue paper between disks, rust on the surfaces, varnish, oil, or paint, are all used for the purpose. Most of the better builders of to-day use a light japan on the disks, with a layer of good insulating paper about every half inch. Open spaces are left in the core about every two inches for ventila- tion. Armature cores are divided again as to outer surface into smooth bod;/ and toothed; the latter calied formerly the Pacinnotti armature, after it's inventor. The smooth body armature core is enough smaller in diameter than the inner circle of the pole faces, to allow laying on the winding ; the full diameter of the toothed armature core is only enough smaller than the field pole space to allow proper air-gap, and slots are provided ir» its periphery in which are laid the conductors. The toothed ring armature is used to-day in the United States to perhaps a greater extent than any other form, although the winding is of the drum form used with multipolar dynamos. The toothed armature is said by Professor Crocker to possess the follow- ing advantages and disadvantages over the smooth body. Advantages : 1. The reluctance of air-gap is minimum. 2. The conductors are protected from injury. 3. The conductors cannot slip along the core by action of the electrody- namic force. 4. Eddy currents in the conductors are avoided. 5. If the teeth are practically saturated by the field magnetism, they oppose the shifting of the lines by armature reaction. ■M^B^^^MI A R M A T U RES . 251 Disadvantages, 1. More expensive. 2. The teeth tend to generate eddy currents in the pole-pieces. 3. Self-induction of the armature is increased. If the slots are made less in width than 2| or 3 times the air-gap, so that the lines spread and become nearly uniform" over the pole faces, hut little effect will be felt from eddy currents induced in the pole faces. When it is not possible to make such narrow slots, pole-pieces must be laminated in the same plane as the disks of the armature core, or the gap must be consid- erably increased. Hysteresis in the armature core can be avoided to a great extent by using the best soft sheet iron or mild steel, Avhich must be annealed to the softest point by heating to a red heat and cooling very slowly. Disks are always punched, and are somewhat hardened in the process ; annealing will not only remove the hardness, but will remove any burrs that may have been raised. Disks should be punched of such careful dimensions as to need no filing or truing up after being assembled. Turning down the surface of a smooth- body armature core burrs the disks together, and is apt to cause dangerous heating in the core when finished. Light filing is all that is permissible for truing up such a surface. Slotted cores should be filed as little as possible, and can sometimes be driven true with a suitable mandril. End plates of iron are seldom satisfactory, and the use of gun metal or other bronze is to be commended. Bolts through the core must be insulated, or currents will be induced in them as in any conductor. Cores were formerly designed of small diameter, especially so in those of the drum type ; but now the dimensions of the core take no particular shape, excepting in some cases it is said to be better to make the cross-section of each side of ring-armature cores approximately square, although cores of a rectangular cross-section ansAver better the purpose for avoiding excessive heating, and for least cost. The size of core is determined first by the number and size of conductors it has to carry to produce the required E.M.F. ; and secondly, by the surface necessary to avoid excessive rise of temperature. Armature conductors are usually made GOO to 800 circular mils per ariipere, and the number of paths through the armature between which the current is divided is determined by the design of the winding and the number of poles. In a bipolar closed-coil winding there are two paths, each carrying one-half the total current, while a four-pole closed-coil winding may have either two or four circuits. The method of determining the number of con- ductors necessary to produce the required E.M.F. is explained in the early part of this chapter. For losses in cores of armatures, see chapter on Mag- netic Qualities of Iron. Armature shafts must be very strong and stiff, to avoid trouble from the magnetic pull should the core be out of centre. They are made of machin- ery steel, and have shoulders to prevent too much side play. Core Insulation. — A great variety of material is used for insulating the core, including asbestos, which is usually put next to the core to prevent damage from heating of that part ; oiled or varnished paper, linen, and silk ; press board ; mica and micanite. For the slots of slotted cores the insula- tion is frequently made into tubes that will slide into the slots, and the con- ductors are then threaded through. Special care must be taken at corners and at turns, for the insulation is often cut at such points. The armature conductors of the Niagara dynamos are insulated by a layer of mica wound on to the bar a inch thick, and then pressed, into place under high and hot steam pressure. Armature Windings. For all small dynamos, and in many of considerable size, the winding is of double cotton-covered wire. Where the carrying capacity is more than the safe carrying capacity of a No. 8 B. & S. gauge, the conductor should be stranded. In large dynamos, rectangular copper bars, cables of twisted cop- per, and in some cases large cable compressed into rectangular shape, are more commonly used. If the copper bars are too wide, or wide enough so that one edge of the bar enters the field perceptibly before the remaining 252 DYNAMOS. parts of the bar, eddy currents are induced in it ; such bars are therefore made quite narrow, and it is common to slope the pole face a trifle, so that the bars may enter the field gradually. Methods or arrangement of windings are of a most complex nature, and only the most general in use will be described here, and these only in theory. Parshall & Hobart have described about all the possible combinations ; S. P. Thompson, Hawkins & Wallis, and others have also written quite fully on the subject. Ifiipolar windings are not windings at all, as the armature is simply a cylinder or disk of metal ; and as none have as yet been put to practical use, no further comment will be made on them. ISing- or Cri-amnie Windings. The form of core does not to-day deter line the form of winding, for, while the drum core is always of necessity wound with the drum winding, the ring core can be wound with either the ring or drum winding, as will be explained. The simplest form of ring winding is the two-circuit single winding, where a continuous conductor is wound about the ring, and taps taken off to the commutator at regular intervals. The first variation on this will be the multi-circuit single winding, used Avhere there are more than one pair of poles. Fig. 40 ' shows the four- circuit single winding. Where it is advisable to reduce the number of brushes in use, the multi- circuit winding can be cross-connected ; that is, tbo.se parts of the winding occupying similar positions in the various fields are connected in parallel to the same commutator bar. Fig. 41 shows one of the simplest forms of cross-connected armatures. Where, from the shape of the frame, the magnetic circuits are somewhat unequal, the winding shown in Fig. 42 will average up the unequal induction values, and prevent sparking to some extent. It also halves the number of commutator segments ; that is, there are two coils connected to each segment instead of One, as in the previously mentioned windings. If n = number of coils, and p = number of poles, any coil is connected across to one ( - ± 1 I in advance of it. \ p Multiple Winding'* for Ring' Arniatiirvs. - An important clai of windings much in use at present, and for many purposes invaluable, is the double, triple, quadruple, etc., wound ring In these classes two or more entirely separate and distinct Avindings are employed, each connected to its own set of segments, the segments of the different windings following each other in consecutive order. Fig. 43 shows the simplest form of two-circuit double winding, used in ARMATURES. 253 a bipolar machine. As no two segments of the same circuit are adjacent, the liability of short-circuit of the commutator is diminished. Two-circuit Winding"!* for Multipolar Field*. — This is an important class of windings, and, as it has but two circuits irrespective of the number of poles, has the advantage over the multiple-circuit windings that it needs but — as many conductors as are necessary in that class, and therefore needs but - as much space for insulation. n But two sets of brushes are necessary for the two-circuit windings, unless the current is heavy enough to require a long commutator, in which case other sets of brushes can be added, up to the number of poles. In the short-connection type of this class, conductors under adjacent field poles are connected together so that the circuits from brush to brush are influenced by all the poles, and are therefore equal. In the long-connection type the conductors under every other pole are con- nected, so that the conductors from brush to brush are influenced by but one-half the number of poles. The number of coils in a two-circuit long-connection multipolar winding is determined by the formula Avhere S = the number of coils, n = the number of poles, and y = the pitch. The number of commutator segments is equal to the number of coils, and must be odd for machines with an even number of pairs of poles, but may be either odd or even for machines having an odd number of pairs of poles. The pitch, y, is the number of coils advanced over for end connections, as, for instance, in an armature with a pitch of 7 the end of coil number 1 is connected to the beginning of coil 1 -j- 7 = 8, and from 8 to 8 -f- 7 = 15, and so on. In multipolar ring long-connection windings y may be any integer, but not so in drum windings. Mr. Kapp gives, in the following table, the best practice as to angular distance between brushes for this class of windings. 254 Number of poles. Angular distance between brushes. Degrees. Degrees. Degrees. Degrees. Degrees. 2 180 4 90 6 60 180 8 45 135 10 36 108 180 12 30 90 150 14 25.7 77 128 180 16 22.5 67.5 112 158 18 60 100 140 180 20 54 90 126 162 Fig. 44 shows a simple form of two-circuit multipolar single winding, and Fig. 45 another sample as used with a greater number of poles. Fig. 45. Both of the above samples are of the long-connection type. In the short- connection type the formula for determining the number of coils is S = ny ± 2, and Fig. 46 is a sample diagram of one of the type. Two-circuit Multiple-wound Multipolar Ring's. — The for- mula for determining the number of coils and other factors for this class of windings is ARMATURES. 255 £=— X y ±m where S= number of coils, n = number of poles, y = pitch, m z= number of windings, as double, triple, etc. " m " will equal a number of independently re-entrant windings equal to the greatest common factor of y and m. The following figure is a diagram of a two-circuit doubly re-entrant, double wound ring armature : in Fig. 47. Fig. 48 is a diagram of a two-circuit, singly re-entrant, double-wound ring. Drum WiiB4li]i£'.«>, In order that the E.M.F.'s generated in the coils of a drum armature may be in the same direction, it is necessary that the two sides of each coil be in fields of opposite polarity, and therefore the sides of the coils are connected 256 DYNAMOS. across the ends of the core ; directly across, for bipolar machines, and part way so for those of the multipolar type. Figure 49 shows the Von Hefner- Alteneck drum winding, used principally in small and smooth core armatures. Fig. 49. A sample of two-layer, two-circuit single winding is shown in Fig. 50. Multiple-circuit Single- wound, Multipolar Drums. — In this class of winding there must be an even number of bars ; and for single wind- ings the pitch at one end must exceed that of the other by 2, and must both be odd. If n is the number of poles, and c the number of face conductors, the average pitch y should be about — . For chord windings y should be as much smaller than — as convenient. ■1 ARMATURES. 257 In iron-clad windings the number of conductors must be a multiple of the number of conductors per slot. Following is a diagram of a six-circuit, single windinq. Fig. 51 « Two-circuit. Sing-le-wouud, Brum Armatures. —In this type of winding, the pitch y is always, forward, and must be an odd number, the connections leading the winding* from a certain bar under one pole to a bar similarly situated under the next pole in advance. Two-circuit drum wind- ings have for a given voltage — as many conductors as multiple-circuit windings. 258 DYNAMOS. When as many sets of brushes are used as there are poles, careful adjust- ment of the brushes is necessary in order to avoid excessive flow of current and bad sparking at any one set of brushes, with symbols the same as in the previous paragraph, c = n y ± 2. The following diagram shows the connections of a two-circuit single winding. Two-circuit, Multiple-wound, Drum Armatures.— With the same symbols as before, and m =r number of windings, the general formula is c =: n y ± 2 m. ARMATURES. 259 This is a large class, and many combinations have been worked, Figs. 53 and 54 showing two of the simpler ones ; the first a two-circuit triple wind- ing, and the second a two-circuit double winding. Alternating* current Armatures. Almost any continuous current armature winding may in a general way be used for alternating currents, but they are not well suited for such work, and special windings better adapted for the purpose are designed. Alternating current armature windings are open-circuit windings, except- ing in the rotary converter, where the rings are tapped directly on to the direct current armature windings. Early forms of armature windings of this type, as first used in the United States, had pan-cake or flat coils bound on the periphery of the core. In the next type the coils were made in a bunched form, and secured in lai'ge slots across the face of the core. Both these types were used for single- phase machines. After the introduction of the multiphase dynamo, arma- ture windings began to be distributed in subdivided coils laid in slots of the core ; and this is the preferred method of to-day, especially so in the case of revolving field machines. The single coil per pole type of winding gives the larger E.M.F., as the coils are thus best distributed for influence by the magnetic field. This type also produces the highest self-induction with its attendant disadvantages. The pan-cake and diatril»if< d-cn'd windings are much freer from self-induc- tion, but do not generate as high E.M.F. as does the single-coil windings. In well-considered multiphase windings the E.M.F. is but little less for distributed coils than for single coils, and has other advantages, especially where the use of step-up transformers permits the use of low voltages, and consequently light insulation for the coils. The distributed-coil winding offers better chance for getting rid of heat from the armature core, and the conductor can in such case be made of less cross-section than would be required for the single-coil windings. The greater number of coils into which a winding is divided, the less will be the terminal voltage at no load. Parshall & Hobart give the following ratio for terminal voltage under no-load conditions : 260 DYNAMOS, Single-coil windings 1. for the same total number of conductors, the spacing of conductors being uniform over the whole circumference. Two-coil winding = .707. Three-coil winding = .667. Four-coil winding = .654. When the armature is loaded, the current in it reacts to change the termi- nal E.M.F., and this may be maintained constant by manipulation of the exciting current. With a given number of armature conductors this reac- tion is greatest with the single coil per pole winding, and the ratios just given are not correct for full-load conditions. Single-phase "Winding's. — The following diagram shows one of the simplest forms of single-phase winding, and is a single coil per pole winding. Another similar winding, but with bars in place of coils, is shown in the following figure. It can be used for machines of large output. ARMATURES, 26i The following figure shows a good type of three bars per pole winding, which is simple in construction. Two-phase "Wineliiisrs. — The following diagram shows a good type of winding for quarter-phase machines. It utilizes the winding space" to good advantage, and is applicable to any number of coils per pole per phase. Fig. 59 is a diagram of a bar winding for a quarter-phase machine, with four conductors per pole per phase. Three-phase Winding's. — Fig. 60 is a diagram of a three-phase 262 DYNAMOS. winding connected in Y, in which one end of each of the three windings is connected to a common terminal, the other ends being connected to three collector rings. Fig. 61 is a sample of a three-phase delta winding, in which all the con- ductors on the armature are connected in series, a lead being taken off to a collector ring at every third of the total length. Fig. 61. In the Y windings the proper ends to connect to the common terminal and to the rings may be selected as follows : Assume that the conductor in the middle of the pole-piece is carrying the maximum current, and mark its direc- tion by an arrow ; then the current in the conductors on either side of and ad- jacent to it will be in the same direction. As the maximum current must be coming from the common terminal, the end toward which the arrow points must be connected to one of the rings, while the other end is connected to the common terminal. It is quite as evident that the currents in the two adjacent conductors must he flowing into the common terminal, and there- fore the ends toward which the arrows point must be connected to the com- mon terminal, while their other ends are connected to the remaining two rings. In a delta winding, starting with the conductors of one phase in the mid- dle of pole-piece, assume the maximum current to be induced at the moment in this conductor ; then but one-half the same value of current will be included at the same moment in the other two phases, and its path ARMATURES. 263 and value will best be shown in the following diagram, in which x may be taken as the middle collector-ring, and the maximum current to be flowing from x toward z. It will be seen that no current is coming in over the line y, bat part of the current at z will have been induced in branches b and c. Most three-phase windings can be connected either in Y or delta ; but it must be borne in mind that with the same windings the delta-connection will stand 1.732 times as much current as the Y- ~-i much voltage. Heating- of Armatures. Fig. 62. Path ami Value of Current in Delta- The temperature an armature will attain during connected Armature, a long run depends on its peripheral speed, the means adopted for ventilation, the heating of the conductors by eddy currents, the heating of the iron core by hysteresis and eddy currents, the ratio of the diameter of tbe insulated conductor to that of its copper core, the current density in the conductor, the radial depth of winding, whether the armature is of cylinder or drum type, and the amount and character of the cooling surface of the wound armature. The higher the peripheral speed of the armature the less is the rise of temperature in it. Mr. Esson gives, as the result of some experiments on armatures with smooth cooling surfaces, the following approximate rule : 55 W 350 W ~ S (1 + 0.00018 V) S' (1 + 0.00059 P)' where C° = difference of temperature between the hottest part of the arm- ature and the surrounding air in degrees, Centigrade, W= watts wasted in armature, S = active cooling surface in square inches, S' = active cooling surface in square centimeters, V= peripheral speed of armature in feet per minute, V' ■=. peripheral speed in meters per minute The more efficient the means adopted for ventilating the armature by cur- rents of air, the smaller is the temperature rise. Some makers leave spaces between the winding at intervals, thus allowing the air free access to the core and between the conductors. A draught of air through the interior of the armature assists cooling, and should be arranged for whenever possible. For heavy currents it is sometimes necessary to subdivide the conductors to prevent eddy currents ; stranded conductors, rolled or pressed hydraulic- ally, of rectangular or wedge-shaped section, have been used. Such subdi- vision should be parallel to the axis of the conductor, and preferably effected br the use of stranded wires rather than laminse. Few armature conductors of American dynamos of to-day are divided or laminated in any degree whatsoever. Solid copper bars of approximately rectangular cross-section are often used, and little trouble is found from Foucault currents. The power wasted by eddy currents in an armature core is proportional to the square of the maximum magnetic induction and to the frequency of change of magnetic induction in the iron. Mr. Kapp considers 1.5 square inches (9.7 square centimeters) of cooling surface per watt Avasted in the armature, a fair allowance. Esson gives the following for armatures revolving at 3000 feet per minute. W '=. watts wasted in heat in winding and core, S = cooling surface, exterior, interior, and ends, in square inches, S, = cooling surface, exterior, interior, and ends, in square centimeters, T =. temperature difference between hottest part of armature and surrounding air in C°. m 35 W 225 W Then T = or — - — Specifications for standard electrical apparatus for U.S. Navy say, " No 264 DYNAMOS. part of the dynamo, field, or armature windings shall heat more than 50° F. above the temperature of the surrounding air after a run of four hours at maximum rated output." According to the British Admiralty specification for dynamos, the tem- perature of the armature one minute after stopping, after a six hours' run, must not exceed 30° F. above that of the atmosphere. In this test the ther- mometer is raised to a temperature of 30° F. above that of the atmosphere before it is placed in contact with the armature, and the dynamo complies (or does not comply) with the specification according as the thermometer does not (or does) indicate a further rise of temperature. The best dynamo makers to-day specify 40° and 45° C. as the maximum rise in temperature of the hottest part of a dynamo, or 55° if the tempera- ture of the commutator surface is to be measured. Armature Reactions. In continuous current dynamos, with no special devices for reversing the currents in the armature sections as they successively pass under the brushes, it is necessary, in order to avoid sparking, to give the brushes a forward lead ; the lead usually varies with the output of the dynamo. With the forward lead given to the brushes the effect of the armature cur- rent is to weaken and distort the magnetic field set up by the field-magnets ; a certain number — depending on the lead of the brushes — of the armature ampere-turns directly oppose those on the field-magnets, and render a some- what lai'ger number of these ineffective, except as regards wasting power ; the remaining armature ampere-turns tend to set up a magnetic field at right angles to the main field, with the result that the resultant field is rotated forward in the direction of motion of the armature, and that tbe field-strength is reduced in the neighborhood of every trailing pole-piece horn, and is increased in that of every leading pole-piece horn. When, therefore, the brushes have a forward lead each armature section as it comes under a brush enters a part of the field, of which tbe strength is reduced by the armature cross-induction ; and, if this reduction is great, the field-strength necessary for reversing the current in the section (in the short time that the section is short-circuited under the brush) may not be obtained, and sparkless collection may thus be rendered impossible. Various devices for reversing the currents in the armature sections, as they pass successively under the brushes, without giving a forward lead to the brushes, have been proposed ; a number of these are described in the paper by Mr. Swinburne; an improvement of Mr. W. B. Sayers consists in interposing auxiliary coils between the joints of adjacent armature sections and the corresponding commutator bars. Each auxiliary coil is wound on the armature with a lead relatively to the two main armature sections and the commutator bar which it connects together. Tbe result of this arrange- ment is that the difference between the E.M.F.s in the two auxiliary coils connecting any given armature section to the two corresponding commuta- tor bars may be made sufficient to reverse the current in the armature sec- tion when short-circuited under a brush, even if the brush has a backward instead of a forward lead. Mr. Sayers's invention not only makes it possible to reduce the air-gap very considerably, but also, by enabling a backward lead to be given to the brushes, to make the armature winding assist that on the field-magnets in producing the required magnetic field for the arma- ture. Both these results assist in reducing the weight and excitation of the field-magnets. For a two-pole dynamo the back ampere-turns are given by the formula, 180 where 6 = angular lead of brushes in degrees, JlT— number of conductors, counted round periphery of armature, in series, /=: armature current in amperes ; and, according to Prof. S. P. Thompson, the number of ampere-turns on the field-magnets required to compensate for the back ampere-turns on the armature is v X (A.T.)a, where v is the coefficient of magnetic leakage. ARMATURES. 265 In the Thompson-Ryan dynamo the effects of armature reaction are neu- tralized by a special winding through slots across the faces of the pole-pieces, parallel Avith the axis of the armature; this winding is in series with the armature, and the same current flowing in both, but in such direction tbat all effects on the field magnets are neutralized, the ampere-turns of the shunt are therefore much less than in other dynamos, there is no sparking under any ordinary conditions of load, the brushes are placed permanently when the machine is set up, and the efficiency is high under a wide range. This dynamo is not compound- wound in the usual meaning of the term, but the effects of compounding can be obtained by varying the position of the brushes, a backward lead, tending to raise the voltage by assisting the field magnets, as the current or load increases. I* rag" on Armature Conductors. — In dynamos, each armature conductor has to be driven in opposition to an effort or drag proportional at every instant to the product of the current carried by the conductor into the strength of the magnetic field. This drag on a conductor varies, there- fore, with the position of the conductor relatively to the field-magnet poles, and is a maximum when the conductor passes through that part of the air- gap at which the magnetic induction is greatest. The arrangements for driving the armature conductors must, of course, be adapted to the greatest value of the drag to which a conductor is exposed, and this is given for smooth core armatures by the formulae below. Let /= current in amperes carried by each conductor, (B = maximum induction in air-gap per square centimeter, F ■=. maximum drag on a conductor in lbs. per foot of length. Then F — -^§^ or .00000685 / ($> 140,000 In slotted armatures the drag comes upon the core teeth instead of the conductors. Current Density in Armature Conductors. —This should be determined so that the I2r loss, plus the hysteresis loss in the armature, does not exceed the less of the two limiting values assigned by the condi- tions of efficiency and freedom from overheating respectively ; in practice current densities of 2,000 to 3,000 amperes per square inch are common, and in drum armatures the current density is sometimes higher. American practice gives 600 to 800 circular mils per ampere. Surface necessary for Safe Temperature. Esson gives the following method of determining the surface necessary for a magnet coil to keep its heat within assigned limits. Let w ■=. watts wasted in heating, s =r cooling surface in square inches of coil, not including end flanges and interior, s, = same as above in square centimeters, t =. temperature of hottest part above surrounding air, then £F°=99 - or *C° = 335 - s s/ Maximum current = Vrfe^- F X s1^ 99 X hot r Hot r = cold r -f- 1% for each additional 4.5° F. 266 DYNAMOS. TmI>1«> of Cooling* Surfaces. Excess temperature above sur- rounding air. Cooling surface per watt in F.° C.° square inches. sq. centimeters. 15 3.G7 23.7 30 — . 3.30 21.3 — 20 2.75 17.8 40 — 2.48 ICO — 25 2.20 14.2 50 — 1.98 12.8 — 30 1.83 11.8 60 — 1.65 10.7 — 35 1.57 10.1 70 — 1.41 9.1 — 40 1.3S 8.9 Notes. — The number of ampere-turns necessary to overcome an air-gap of one-half inch equals the number of lines of force per square centimeter. Approximate rule by G. Forbes. Current l>ensity. (Esson.) The current density per square centimeter section in the magnet winding of ordinary machines is about half the current density in the armature. §af« Continuous Output of X&ynamos and Motors. (Albion Snell.) _ | Drums Watts = IcVn .015. uynamos { Cylinders Watts = Mhi .01. I Drums Brake H.P. = IcPn .000015. nh \ Cylinders Brake H.P. = .00001. I rr length of armature in inches, d = diameter of armature in inches, n — number of revolutions per minute. are great, it will be found advisable to place their plane of rotation athwart- ships, in order to avoid as far as possible wear and tear of bearings due to the gyrostatic action. §YACHROIIZKK$». There are numerous methods of determining when alternators are in step, some acoustic, but mostly using incandescent lamps as an indicator. In the United States it is most common to so connect up the synchronizer that the lamp stays dark at synchronism ; in England it is more usual to have the lamp at full brilliancy at synchromism, and on some accounts the latter is, in the writer's opinion, the better of the two, as, if darkness indi- cates synchronism, the lamp breaking its filament might cause the machines to be thrown together when clear out of step ; on the other hand, it is some- times difficult to determine the full brilliancy. The two following cuts show theory and practice in connecting synchro- jy^) ^y® 4© /& ^^vM&AMM/- h I Fig. G4. Synchronizer Connections. Lamp lights to full c.p. when dyna- mos are in synchronism. Fig. 63. Synchronizer Connections. W hen connected as shown, the lamp will show full c.p. at synchronism. If a and b are reversed, darkness of lamp will shoiv synchronism. Two transformers having their primaries connected, one to the loaded and the other to the idle dynamo, have their secondaries connected in series through a lamp ; if in straight series the lamp is dark at synchronism ; if the secondaries are cross-connected the lamp lights in full brilliance at synchronism. Note on tlie Parallel It«iniiing- of Alternators. — There is little if any trouble in running alternators that are driven by water-wheels, owing to the uniform motion of rotation. With steam-engine driven ma- chines it is somewhat different, owing to more or less pulsation during a stroke of the engines, caused by periodic variations in the cut-off, which causes oscillations in the relative motion of the two or more machines, accompanied by periodic cross currents. Experiments have proved that a sluggish governor for engines driving alternators in parallel is more desi- 268 DYNAMOS. rable than one that acts too quickly ; and it is sometimes an advantage to apply a dashpot to a quick-acting governor, one that will allow of adjust- ment while running. It is quite desirable also that the governors of engines designed to drive alternators in parallel shall be so planned as to allow of adjustment of speed while the engine is running, so that engines as well as dynamos may be synchronized, and load may be transferred from one machine to the others in shutting down. Foreign builders apply a bell con- tact to the same part of all engines that are to be used in this way, and throw machines together when the bells ring at the same time. These bells would also serve to determine any variation, if not too small, in the speed of the machines, and assist in close adjustment. Manufacturers do not entirely agree as to the exact allowance permissible for variation in angular speed of engines, some preferring to design their dynamos for large synchronizing power, and relatively wide variation in angular speed, while others call for very close regulation in angular varia- tion of engine speed, and construct their dynamos" with relatively little syn- chronizing power. Dynamos of low armature reaction have large synchronizing power, but if accidentally thrown out of step are liable to heavy cross-currents. On the contrary, machines with high armature reaction have relatively little syn- chronizing power, and are less liable to trouble if accidentally thrown out of step. The smaller the number of poles the greater may be the angular variation between two machines without causing trouble, thus low frequencies are more favorable to parallel operation than high ; and this is especially so where the dynamos are used to deliver current to synchronous motors or rotary converters-. Specifications for engines should read in such a manner as to require not more than a certain stated angular variation of speed during any stroke of the machine, and this variation is usually stated in degrees departure from a mean speed. The General Electric Company states it as follows : — "We have . . . fixed upon two and one-half degrees of phase departure from a mean as the limit allowable in ordinary cases. It will, in certain cases, be possible to operate satisfactorily in parallel, or to run synchronous apparatus from machines whose angular variation exceeds this amount, and in other cases it will be easy and desirable to obtain a better speed con- trol. The two and one-half degree limit is intended to imply that the max- imum departure from the mean position during any revolution shall not exceed ^- of an angle corresponding to two poles of a machine. The angle of circumference which corresponds to the two and one-half degrees of phase variation can be ascertained by dividing two and one-half by one-half the number of poles ; thus, in a twenty-pole machine, the alloAvable angular variation from the mean would be -^ = .25 of one degree." Some foreign builders of engines state the conditions as follows : Calling !N the number of revolutions per minute, the weight of all the rotary parts of the engine should be such that under normal load the variation in speed dur- . ,. Nmax. — Nmin. .ni , 1 _, , , 1 me one revolution — will not exceed — — • Some state -—r- N average 250 200 Oudin says : " The regulation of an engine can be expressed as a percent- age of variation from that of an absolutely uniform rotative speed. A close solution of the general problem shows that 1J° of phase displacement cor- responds to a speed variation, or " pulsation," Avith an alternator of two n poles, as follows : — In the case of a single cylinder or tandem compound engine — 5.5% A cross compound A working out of the problem also shows . . . that no better results are obtained from a three-crank engine than a two-crank. The Westinghouse Company designs its machines with larger synchro- nizing effect by special construction between poles, and allows somewhat SYNCHRONIZERS. 269 larger angular variation, stating it as follows : The variation of the fly- wheel through the revolution at any load not exceeding 25% overload, shall not exceed one-sixtieth of the pitch angle between two consecutive poles from the position it would have if the motion were absolutely uniform at the same mean velocity. The maximum allowable variation, which is the amount which the armature forges ahead plus the amount which it lags behind the position of absolute uniform motion is therefore one-thirtieth of the pitch angle between two poles. The number of degrees of the circumference equal to one-thirtieth of the pitch angle is the quotient of 12 divided by the number of poles. Alternators in Parallel. To connect an idle alternator in parallel with one or more already in use : Excite the fields of the idle machine until at full speed the indicator shows bus bar pressure, or the pressure that may have been determined on as the best for connecting the particular design of alternator in circuit. Connect in the synchronizer to show when the machines are in step, at which point the idle machine may be connected to the bus bars. The load will now be unequally divided, and must be equalized by increasing the driv- ing-power of the idle dynamo until it takes on its proper part of the load. Very little control over the load can be had from the field rheostats. To disconnect an alternator fron the bus bars : Decrease its driving power slowly until the other machines have taken all the load from it, when its main switch may be opened and the dynamo stopped and laid off. Current leads from brushes to binding-posts, must be ample to produce no appreciable drop in voltage. The following table gives current densities, etc., for brush- holders, cables, conductor-rods, cable-lugs, binding-posts, and switches. Average Current Densities for Cross-section and Contact Surface of Various materials. Material. Current density. Square Mils per Ampere. Amperes per Square Inch. 1 Cross section . y J Copper wire . . . Copper rod .... ! Copper-wire cable . Copper casting . . Brass casting . . . 500 to 800 800 " 1,200 600 " 1,000 1,400 '• 2,000 2,500 " 3,300 1,200 to 2,000 800 " 1,200 1,000 " 1,600 500 " 700 300 " 400 Brush contact . Copper brush . . . Carbon brush . . . 5,700 " 6,700 28,500 " 33,500 150 " 175 30 " 35 Sliding contact [ Copper — copper . . ±5i ass <^braBB . . ( 10,000 " 15,000 ) 20,000 " 25,090 62 " 100 40 " 50 Screwed contact j Copper — copper -Brass ^coPPe1" • lirass is constant Twill be proportional to la, and Twill be greatest, there- fore, when the armature is standing, and la = -^. If r — resistance of the armature then /« = ?—' and T .•*« ™X* r X 13.56 X 107 or T = 0, when rev. n * = E x 108. , . E X 108 2n x T x 13.56 X r X 1015 Speed in rev. per sec. = r— - n x $ n2 4>2 If r is small and $ is relatively large, the second term may be neglected. The stronger the field, i.e., the , the slower will be the speed ; and if $ is constant the speed is proportional to E. §ei'ie§-Wound Motor. Yalues in C. G. S. units. In a series motor R= ra -f- rm where ra = resistance of the armature, at " rm = resistance of the fields : Let $ sat. = complete saturation of field magnets, and I' — diacritical cnrrent, or current at half saturation, /+/' "Writing Ffor - / + — -f- 13.56 X 107 — torque in pounds at 1 ft. radius. /= -^p in C. G. S. units. in C. G. S. units. 272 In a series motor the current is the same tinder the same load at any speed. In other words, the torque is almost directly proportional to the current. The following curves show the speed and torque curves ior a series motor on a constant potential circuit. TORQUE Fig. 65. NIiunt-%T$'omi>rf Motor. Values in C, G. S. units. : / — Is, where Is z E T- and if Y "~ E + E' tion in field magnets, - ra I. $ sat. x current in the shunt field. where E' is the E M F, to givt half satura- E E-{- E> _ T E + E' b 1= _x-^— + - =A [*(»+?).-".'] Brushes on a motor must he set back of the neutral point, or with a " backward lead." This tends to demagnetize the fields, and as weakening the fields of a motor tends to increase the speed, tne increase oi load on a shunt-wound motor tends to prevent the speed failing, and the shunt motor is very nearly self-regulating. JLeonar-cTs System oi* Motor Control. Wherever it becomes necessary to vary the speed and torque of a contin- uous current electric motor to any considerable degree, any of the rheostat methods introduce very considerable losses, and are apt to induce bad sparking at the commutator. H. Ward Leonard, E.E., invented the method shown in Fig. 66, which gives most excellent results, although to some extent complicated, and is highly efficient. The driving motor, or rather motor which it is wished to control, is pro- vided with a separately excited field, which can be varied by its rheostat to produce any rate of speed, from just turning to the full speed of which it may be capable. Current is supplied to its armature from a separate gen- erator, and by varying the separately excited field of this generator, the amount of current supplied to the motor armature can be varfed at will, and the torque therefore changed to suit the circumstances. The generator is driven at constant speed by direct connection to a motor which gets its current from an outside source, or to another generator driven by some other motive power, say a steam engine. This driven gen- ALTERNATING CURRENT MOTORS. 273 erator supplies current for exciting the fields of the secondary generator and main motor. By reversing the field of the generator, the current in its armature is reversed, and therefore so is the direction of rotation of the motor armature. Fig. 67 shows the Leonard system adapted to electric street railway motor control. MOTttR. 6ENERAT0R Leonard's System of Motor Control. Fig. 67. Leonard's System of Electric Propulsion. ALTERMATIWG CURRENT MOTORS. While the single-phase alternating current motor has been quite well de- veloped during the last few years, it has as yet come but little into use, owing largely to its inductive effect on the line, and poor efficiency and un- satisfactory operation. On the contrary, the multiphase motor has been so far developed as to bring it into very strong competition with the direct current motor, owing probably to its extreme simplicity, lacking all brushes, commutators, and other troublesome attachments. Fig. 68. Connections for Standard S. F. A. C. Motor of the Fort Wayne Electric Corporation. Only the most elementary formula? will be given here, and the reader is referred to the numerous books treating on the subject ; among others, S. P. Thompson, Steinmetz, Jackson, Kapp, and others. Following is a statement of the theory of the multiphase motor, condensed from a pamphlet of the Westinghouse Electric and Manufacturing Company. 274 Elementary Theory of the Multiphase Induction Motor. If a horse-shoe magnet he held over a compass the needle will take a posi- tion parallel to the lines of force which flow from one pole to the other. It is perfectly obvious that if the magnet he rotated the needle will follow. If a four-pole electromagnet he substituted for the horse-shoe, and current be made to flow about either one of the sets of poles separately, the needle will take its position parallel with the lines of force that may be flowing, as will be seen by the following figures. Fig. 69. Fig. 70. If the two sets of poles are excited at the same time by currents of equal strength, then the needle will take its position diagonally, half way be- tween the two sets of poles, as will be seen by the following diagram. It is now easily conceivable that if one of these currents is growing stronger while the other is at the same time becoming weaker, the needle will be at- tracted toward the former until it reaches its maximum value, when if the currents are alternating, the strong current having reached its maximum begins to weaken, and the other current having not only re- versed its direction hut begun to grow strong, attracts the needle aAvay from the first current and in the same direction of rotation. If this process be continually repeated, the needle will continue to re- volve, and its direction of rotation will he determined by the phase relation of the two currents, and the direction of rotation can be reversed by reversing the leads of one phase. If the compass needle be replaced by an iron core wound with copper conductors, secondary currents will be induced in these windings, which will react on the field windings, and rotation will be produced in the core just as it Avas in the compass needle. Two cranks at right angles on an engine shaft are analogous with the quarter-phase motor, and three to the three-phase motor, which depends on the same principle for its working. Fig. 71. Theory of Multiphase Induction M otor. Condensed from S. P. Thompson. The following names and symbols are used for designating the parts and properties of the induction motor : — ALTERNATING CURRENT MOTORS. 275 Stator — stationary part, nearly always corresponding to the field. Rotor = rotating part, corresponding to the armature of the d.c. motor. Q =1 angular speed of the rotating magnetic field := 2n rev. -^- m, where m = number of pairs of poles, to = angular speed of rotor = 2n rev.., -|-m, where rer.o = number of rev- olutions per second. T = torque between the stator and rotor. Analytical Theory of F*olyi»lia«e Induction Motors. © Let r — resistance per circuit of stator. r, = resistance per circuit of rotor, being reduced to primary system by square of the ratio of turns. Let d = number of poles, x = inductance of primary , per circuit, x, = inductance of secondary, per circuit, reduced to primary system by square of the ratio of turns. Let S = per cent of slip, I— current per circuit of stator, E =z applied E.M.F. per circuit, Z = impedance of whole motor per circuit, JV= frequency of applied E.M.F. Let the primary and secondary consist of p. circuits on a p. phase system. n — primary turns per circuit, tij = secondary turns per circuit, Let a — — ratio of transformation, n, Then /(neglecting ex. current) - ^ + ^ '_ ^ ^ + ^ Torque T- Max. torque : -teJXKrsSrp + SUXj + xM p r, E'2S (\ — S) Z (P/ + gr)* _|_ £2 (X/ + xf dp £2 ' S7riV;[r-f ^r2 + {X/ + x)2\ Max. power = — — f- — — at the slip S = '. r. 2 [r + r, + Z] r rt + Z Starting current = i — -=, cj. j.- dpE- r. Starting torque = "T^v x \72 Note that the maximum torque is independent of rotor resistance r„ and thus the speed at maximum torque depends on the rotor resistance. Current at maximum torque is also independent of rotor resistance. The maximum torque occurs at a lower speed than the maximum output. A resistance can be chosen that when inserted in the rotor, the maximum 276 torque will be obtained at starting ; tbat is, the speed at which maximum torque occurs can be regulated by the resistance in the rotor. o SYNCHRONISM Fig. 72. Torque curves for Polyphase Induction Motor. Curves 1, 2, and 3 show the effect of successive increases of rotor resist- ance, rotor run on part of curve a — b ; for here a decrease of speed due to load increases the torque. Speed of Induction Motor, — The speed or rotating velocity of the magnetic field of an induction motor depends upon the frequency (cycles per second) of the alternating current in the field, and the number of poles in the field frame, and may be expressed as follows : — rev. = revolutions per minute of the magnetic field, p = number of poles, / =: frequency ; then f rev. = 120 - P The actual revolutions of the rotor will be less than shown by the formula, owing to the slip which is expressed in a percentage of the actual revolu- tions ; therefore the actual revolutions at any portion of the load on a motor will be rev. X slip due to the part of the load actually in use. actual speed = rev. (1 — % of slip.) Tbe following table by Wiener, in the American Electrician, shows the speeds due to different numbers of poles at various frequencies. Speed of Rotary Field for Different H>iml»ers of I*oles and for Various Frequencies. o Speed of Revolving Magnetism, in Revolutions per Minute, when %l Frequency is : 25 30 33§ 40 50 60 66| 80 100 120 125 133J 2 1500 1870 2000 2400 3000 3600 4000 4800 6000 7200 7500 8000 4 750 900 1000 1200 1500 1800 2000 2400 3000 3600 3750 4000 6 500 600 667 800 1000 1200 1333 1600 2000 2400 2500 2667 8 375 450 500 600 750 900 1000 1200 1500 1800 1875 2000 10 300 360 400 480 600 720 800 960 1200 1440 1500 1600 12 250 300 333 400 500 600 667 800 1000 1200 1250 1333 14 214 257 286 343 428 514 571 686 857 1029 1071 1143 16 188 225 250 300 375 450 500 600 750 900 938 1000 18 167 200 222 267 333 400 444 533 667 800 833 889 '20 150 180 200 240 300 360 400 480 600 720 750 800 22 136 164 182 217 273 327 364 436 545 655 682 720 24 125 150 167 200 250 300 333 400 500 600 625 667 ALTERNATING CURRENT MOTORS. 277 Slip. — The slip, or difference in rate of rotation between rotating field and rotor, is due to the resistance opposed to rotor current. Slip varies from 1 per cent in a motor designed for very close regulation to 40 per cent in one badly designed, or designed for some special purpose. Weiner gives the following table as embodying the usual variations : Slip of Induction Motors. Capacity of Motor, H.P. Slip, at full load, per cent. Usual limits. Average. § 20 to 40 30 I 10 " 30 20 i 10 " 20 15 1 8 " 20 14 2 8 " 18 13 3 8 " 16 12 5 7 " 15 11 n 6 " 14 10 10 G " 12 9 15 5 " 11 8 20 4 " 10 7 30 3 " 9 6 50 2 " 8 5 75 1 " 7 4 100 1 " 6 35 150 1 " 5 3 200 1 " 4 25 300 1 " 3 2 Core of Stator and Rotor. — Both the field-frame core, or Stator, and the armature core, or Rotor, are built up of laminated iron punchings in much the same manner as are the armature cores of ordinary dynamos. The windings in both cases are laid in slots across the face of either part, and for this reason both parts are punched in a series of slots or holes for the reception of the windings. The following cuts, taken from the " Ameri- can Electrician," show the usual form of slots used. s. 73 and 74. Forms of Punchings of Induction Motors. The number of slots in thesfafor mustbe a multiple of the number of poles and number of phases, and Weiner gives the following table, in the " Ameri- can Electrician," as showing the proper number to be used in various cases, both for two- and three-phase machines. In practice the number of poles is determined by the speed required and the available frequency ; then the number of slots is so designed as to be equally spaced about the whole inner periphery of the stator. 278 MOTORS. H'uml»er of Slots in W idd-Frame of Induction Motors. Capacity of Motor. Number of Poles. Slots per Pole. Slots per Pole per Phase. Two-Phase. Three-Phase. i H.P. to 1 H.P. 4 to 8 3 4 -h 1 \ H.P. to 1 H.P. 4 to 6 6 "2 3 - 4 to 10 5 6 2* 3 2 2 H.P. to 5 H.P. 4 to 6 7 8 9 3* 4" 3 6 H.P. to 50 H.P. 6 to 12 7 8 9 f 3 4 to 8 10 11 VI 5 H 6 1 10 to 20 8 4 3 50 HP. to 200 H.P. 8 to 12 10 11 12 13 5 9 64 4 6 to 10 14 15 16 h - The number of slots per pole per phase in the rotor must be prime to that of the stator in order to avoid dead points in starting, and to insure smooth running, and commonly range from 7 to 9 times the number of poles, or any integer not divisible by the number of poles, in the squirrel cage or single conductor per slot windings. The proper number of slots may be taken from the following table by Weiner : ALTERNATING CURRENT MOTORS. 279 IVumuer of Rotor Slots for Squirrel-Cag-e Induction Motors up to 9 H.JP. Capacity. Number of Poles, p. Limits of Slots, Number 7 p. to 9 p. Number of Rotor Slots. 28 to 36 42 " 54 56 " 72 29, 30, 31, 33, 34, 35, 37. 43, 44, 45, 46, 47, 49, 50, 51, 52, 53. 57,58,59,60,61,62,63,65,66,67, 68, 69, 70,71. In large machines, where there is more than one conductor in each slot and in which the winding is connected in parallel, the number of slots in the rotor' must be a multiple of both the number of phases and the number of pairs of poles. The following table gives numbers of slots for various field-slots : H~uinl»er of Rotor-Slots for Induction Motors of Capacities over S» H.I*. Number of Field-Slots per Pole. Number of Rotor-Slots. (n3 Field-Slots.) f n«. or § n« Alls. * n«. " S n3 | ns. " §"« f lis. " # lis ftn.. " fns |ns. " |ns Flux Rensity. — This must be settled for each particular case, as it will be governed much by the quality of iron and the particular design of the motor. Hysteresis loss increases as the 1.6 power of the flux density; and eddy current losses are proportional to the square of the density and also to the square of the frequency. The following table shows practical values : f lux-Rensities for Induction Motors. ("Wiener.) Flux Density , in Lines of Force pei Square Inch. Capacity of Motor, H.P. For Frequencies from 25 to 40. For Frequencies from 60 to 100. For Frequencies from 120 to 180. Practical Values. Aver- age. Practical Values. Aver- age. Practical Values. Aver- Age. i \ 12000 to 18000 15000" 25000 18000 " 32000 15000 20000 25000 10000 to 15000 12000 " 18000 15000 " 25000 12500 15000 20000 7000 to 11000 7500 " 12500 8000 " 17000 9000 10000 12500 280 MOTORS. JTlux-Densities for Induction Motors — (Continued). Flux-Density, in Lines of Force per Square Inch. Capacity of Motor, H.P. For Frequencies from 25 to 40. For Frequencies from 60 to 100. For Frequencies from 120 to 180. Practical Values. Aver- age. Practical Values. Aver- age. Practical Values. Aver- age. 1 5 10 20 50 100 150 200t 20000 to 40000 25(100 " 45000 30000 " 50000 4oooo " 60000 50000 " 70000 60000 " 80000 70000 " 90000 80000 " 100000 90000 " 110000 30000 35000 40000 50000 60000 70000 80000 90000 100000 18000 to 32000 20000 " 40000 25000 " 45000 30000 " 50000 35000 " 55000 40000 " 60000 45000 " 65000 50000 " 70000 60000 " 80000 25000 30000 35000 40000 45000 50000 55000 60000 70000 9000 to 11000 10000 " 25000 11000 " 29000 12500 " 32500 15000 " 35000 17500 " 37500 20000 " 40000 25000 " 45000 30000 " 50000 15000 17500 20000 22500 25000 27500 30000 35000 40000 In the earlier induction motors it was considered the most efficient method to connect the driving current to the revolving part or rotor; and as it is highly important that the number of windings on the rotor he prime to that of the stator, Fig. 75 shows a winding with an odd combination of con- ductors, being 51, or three times 17. The stator windings would then be bars, con- nected at either end to a heavy copper ring, this forming a sort of " squir- rel-cage." In the modern ma- chines the winding shown would be in coils on the stator, the three ends being carried to terminal blocks on the outside of the machine instead of to rings as shown, and the " squirrel- cage" would then be placed on the rotor and be made of bars as men- tioned. Starting- and Meg'- Fig. 75. ulating- Devices. — Small induction motors, up to about 5 h. p. capa- city, are started by closing the circuit directly to the motor. In large ma- chines this would not be safe, as the rotor is 'standing, and would act in a lesser degree as the short-circuited secondary of a static transformer, and cause a heavy rush of current. Resistance in Rotor. — This is a favorite method with the General Electric Company. A set of strongly constructed resistances is secured inside the rotor ring, and so arranged Avith a lever that they may be closed or short-circuited after the motor has reached its full speed. These resist- SYNCHRONOUS MOTORS. 281 ances are in the armature circuits. In order to give maximum starting torque total armature resistance should be r, = Vr2 + (X/ -\- x)* Where rx = rotor resistance per circuit reduced to held system. xy = rotor reactance per circuit reduced to field system. r = resistance per field circuit. y = reactance per field circuit. This' method serves the double purpose of keeping down the starting cur- rent and increasing the starting torque. Resistances in Stator. — Resistance boxes may be connected in the circuits supplying induction motors ; three separate resistances in three- phase circuits, and two separate resistances in two-phase circuits. They must be all connected in such a manner as to be operated in unison. Under these conditions the pressure at the field terminals is reduced, as is of course the starting current and the starting torque. In order to start a heavy load, under this arrangement, a heavy starting-current is necessary. Compensators or Auto-Transformers. — This method is greatly favored by the Westinghouse Electric and M anuf acturing Company, is used to some extent by the General Electric Company, and consists of introducing an impedance coil across the line terminals, the motor being fed, in starting, from some point on the winding where the pressure is considerably less than line pressure. This avoids heavy drafts of current from the line, thus not disturbing other appliances attached thereto, but as regards starting- current and torque has the same effect as resistances directly in the line ; that is, greatly reduces both. Rotor Windings Commntated. — In this arrangement all or a part of the rotor windings are designed to be connected in series when starting, and are thrown in parallel after standard speed is attained. Another design has part of the conductors arranged in opposition to the remainder in starting, but all are thrown in parallel in regular order when running at standard speed. These commutated arrangements have not been much used in the United States. iYACHROXOUS MOTORS. Alternators are convertible into motors ; and one alternator will run in synchronism with another similar machine after it is brought to the same speed, or, if of unlike number of poles, to some multiple of the speed of the driven dynamo, provided the number of pairs of poles on the motor is divisible into the multiple. Such motors will run as if geared to the driven dynamo even up to two or three times its normal full torque or capacity. Single-phase synchronous motors have no starting-torque, but synchronous motors for multiphase circuits will come up to synchronism without much load, giving about 25 % starting-torque, starting as induction motors, with the d. c. field open. When connected to lines on which are connected induction motors that tend to cause lagging currents and low-power factor of the line, over excita- tion of the synchronous motor fields acts in the same manner as a condenser introduced in the line, and tends to restore the current to phase with the impressed E.M.F., and therefore to do away with inductive disturbances. . It is necessary to provide some source from which may be obtained con- tinuous current for exciting the fields of the synchronous motor ; and this is oftenest done by the use of a small d. c. dynamo belted from the motor- shaft, the exciting ciirrent not being put into use until the motor armature reaches synchronism. In starting a synchronous motor the field is open-circuited, and current is turned on the armature. In practice, field coils are connected in various ways to obviate the dangers of induced voltage, and a low resistance coil similar to the series winding of the d. c. machine is sometimes so arranged on the field poles as to give the necessary reaction for starting. Another way is to use a low-pressure excitation, and therefore few turns on the field coils ; also the field coils are " split up " by a switch at starting. The field excitation is thrown on after the rotating part approaches synchronism, wbich may be indicated by a lamp or other suitable device at the operating switchboard. Considerable care must be exercised in the use of synchronous motors, and their best condition is where the load is quite steady, otherwise they intro- 282 duce inductive effects on the line that are quite troublesome. The field of such a motor can be adjusted for a particular load, so there will be neither leading nor lagging current, but unity power factor. If the load changes, then tlie power factor also changes, until the field is readjusted ; if the load has been lessened the current will lead, and if it increases the current Avill lag. If induction motors are connected to the same line, with a synchro- nous motor that has a steady load, then the field of the synchronous motor can be over-excited to produce a Leading current, which will counteract the effect of the lagging currents induced by the induction motors. If two or more synchronous motors are connected to the same circuit, and the load on one of them is quite variable, and its field is not changed to meet such changing conditions, a pumping effect is liable to take place in the other motors, unless especial provision has been made in the design of the motors to prevent it. It is only necessary to arrange one of the motors of the number for preventing this trouble, but better to make all alike. A copper shield between pole- pieces, and covering a portion of the pole-tip, will prevent the trouble ; and the Westinghouse Electric and Manufacturing Company use a heavy copper j strap around each pole-piece, with a shoe covering part of the pole-tip in the air-gap. Theory of the synchronous motor. GENERATOR Let R =z resistance of Avhole circuit. L =z self-inductance of whole circuit. ^o resultant. Fig. 77. Take the origin at 0. Let E represent maximum value. e = instantaneous value, ex = Ex sin (pt. + <£), e2= E2 sin (pt. — $), where p — 2w n, and n number of complete periods per second. e = E0 sin (pt. — , E,>EX e; - tarn// - E, - /:., cos E9 + E, (Et + E2) Eo Eq and <$> are known. SYNCHRONOUS MOTORS. 283 Energy. Shifts the origin hy the angle \f/. ex = Ex sin (pt. -f 4, + if,). e2 — E2 sin (pt. — -f- »//). E0 Now I~- and / lags behind E0 by the angle 6 where . LP tan 8 = -g- ■ By introducing the angle ty we are referring the E.M.F.'s of both machines to the zero point of the resultant wave as origin. In general ei dt- if. EI , where w = the energy in watts, and © = lag or lead of /with respect to E. E and /are maximum values. T=r — , or the periodic time. H ojj = energy given to the circuit by the generator, cj2 = energy absorbed from the circuit by the motor. J_ / e, idtz Ei _ [i = I sin (pt — 5)] _E\ En X~ 2 V7?2+^2X' ^w + r-p* E,E0 0 cos (0 + \f> + 8 [cos ( -(- »W cos 5 - sin ($ -f- i//) sin 6] cos 5 - VlP+p*!,* " 1 — 2(/?2+^2i2 and substituting — for + we get Re 3 (^-j-^)_ipsin(0 + ^) -2(/P+j3«i>) ( J i? cos (A) + -£j9 sin ( - Now sin \f/ = V" — — sin ©> cos \f/ : Ex — E2 Substituting and reducing 1 An angle -\- Lp sin 20) — E2 R sin 2 01 = cos 201 = y/&+p*Lz Lp VR2-j-p*lS 284 Substitute in w2 , and E2 2 li2 +2)*L2 \ + 2 <£i =: 90° or j J£a Vy>>2 _j_ p2 i2 sin (2 0 + 2 «£i) — 7^27? \ that is, the " sine term " = unity. w2 is positive provided Ei ^ It E.z Vlp+p*& which shows that it is possible to have E2 greater than Ex if there is the proper ratio of resistance and reactance in the circuit. Now, if we plot from an actual motor the armature current and the field excitation we get a curve shown in Fig. 79. This shows that the armature current varies with the excitation for a given load. The flatter curves are for increase of load. Point a shows under excitation, b shows over excitation, c shows the excitation which makes the power factor unity ; it is well from the point of stability of operation to slightly over excite, and this makes E2>EX , and also counteracts the inductive dfop in the line, thus showing that the action of an Fig. 79. over excited synchronous motor is similar to a condenser. Graphical treatment. Eg = generator E.M.F. Em = motor E.M.F. Eo = resultant E.M.F. Jo = resultant current. O Igz=z projection of I0 on O Eg. O lm = projection of Jo on O Em. O Ig . O Eg — u)g = energy given up by the generator. O lm . OEm=. to™ — energy absorbed by the motor from the cir- cuit. coto is negative, which shows that wm is the motor, because it is taking energy from the circuit ; and similarly wg is the gener- ator, because O Eg . O Ig is positive, and gives up energy to the circuit. [For further discussion see Jackson's Alternating Current and Jlternotivfj (.'ar- rent Machines ; also Electrical World for March 30 and April 6, 1895, by Bedell and Ryan. The latter is the classic paper on the subject.] ANGLE OF LAG POSITIVE Fig. ? MOTOR CHEHTEMATOIBS OH BTJTAMOTOHI. These are of two styles, one for transforming continuous current of one voltage into continuous current of a different voltage, and usually called in America motor-generators; the second class transforms alternating current into continuous current, or vice versa, the voltage not being changed except- ing from A.C. Vmean2 values to d.c. values equal to the top of the A.C. wave ; these latter machines are now called rotary converters, and are largely used in connection Avith the circuits of the Niagara Falls Power Company and other power transmission stations. DYNAMOTOKS. 285 Motor-generators are now largely used in telegraph offices for reducing the pressure of the supply current to voltages suitable for use in telegraphy and for ringing and charging generators in telephone offices. Theory. Let E = voltage at motor terminals. e — voltage at generator end terminals. / = current in motor armature. /• = resistance of motor armature. n = number of conductors in motor armature. Ix z= current in generator armature part. ?-, = resistance of generator armature part. Wj = number of conductors in generator armature part. — — k := coefficient of transformation. E = induced E.M.F. in motor part. Ex = induced E.M.F. in generator part. E — rev. x n X <$>■ Ey = rev. — nx x 4>- E — E — r I E1 = e + rJl. ke— E = rl—Jerx Ix. If it be assumed that losses by hysteresis and eddy currents be negligible, or that E I = Ex Ix whence Ix = k I, then -i -(**?)* Such machines run without sparking at the commutator, as all armature reactions are neutralized. Continuous Current Boosters. This is a type of motor generator much in use for raising or lowering the pressure on long feeders on the low-pressure system of distribution, and is to be found in most of the larger stations of the Edison companies. It is also much used in connection with storage-battery systems in charging cells. The " booster " consists of a series generator driven by a motor direct con- nected to its armature shaft. The terminals of the generator are connected in series with one leg of the feeder ; and it is obvious that the current in the feeder will excite the series field just in proportion to the current flowing, provided the design of the iron magnetic circuit is liberal enough so that the field is way below saturation (on the straight part of the iron curve way beloAV the knee). As the armature is being independently rotated in this field, it will produce an E.M.F. approximately in proportion to such excitation, which E.M.F. will be added to that of the feeder or will oppose that E.M.F., ac- cording as the terminal connections are made. On three-wire systems two generators are direct connected to one motor, and for convenience on one bed-plate. Such a booster can be so adjusted as to make up for line loss as it in- creases with the load. One danger of a booster that is not always taken into account is. that if the shunt of the driving-motor should happen to open, or, in fact, anything should happen to the driving-motor that would result in its losing its power, the generator would immediately become a series motor, taking current from the line to which it is connected, and by its nature would reverse in direction of rotation, and increase in speed enormously, and if not discon- nected from its circuits in time would result in a complete Avreck of the machine. It is always safest to have the generator terminals connected to their line through some automatic cut-out, so arranged that should the shunt break, as suggested, it would actuate the device, and automati- cally detach the booster from the circuit before harm could be done. 286 MOTORS. KOTiRY COlfVERlERg. A rotary converter is the name given to a machine designed for changing alternating currents into continuous currents. If the same machine be used inverted, i.e., for changing continuous currents into alternating, it is sometimes known as an inverted converter. Again, if the same machine be driven by outside mechanical power, both alternating and continuous cur- rents may be taken from it, and it then becomes known as a double current generator. Theoretically the rotary converter is a continuous current dynamo with collector rings added, which are connected by leads to certain parts of the armature windings, sometimes at the commutator segments. In the following figure, which represents in diagram the sing!e-])fiase rotary converter, the collector rings r and rx are connected by leads to dia- metrically opposite segments or coils of the armature at c and cx. It is obvious that as the armature revolves the greatest difference of potential between the rings, or maximum E.M.F., will be at the instant the segments c and 1 IN rH b 1 | Sl« |tM(! >I1 in li Hl* CO CO 3 II II 41 ii II >h Two- phase. II in ii o li II Three- phase. li -IS CO li |eoi| > 1 IN li | |> ■ fcJO o3 PI ^ CO II ||0q © II n l«M > <* ii > 3 o O o - - - - o a SP.E?o lis C3 10'*, For three-phase machines E = 3.69 Tn . It should be pointed out that tests at high voltages considerably in excess of the normal voltages are admissible on new machines, to determine whether they fulfil their specifications, but should not be made subse- quently at a voltage much exceeding the normal, as the actual insulation of the machine may be weakened by such tests. Stf. The test for dielectric strength should be made with the completely assembled apparatus, and not with its individual parts ; and the voltage should be applied as follows : — 1st. Between electric circuits and surrounding conducting material, and. 2d. Between adjacent electric circuits, where such exist, as in trans- formers. The tests should be made with a sine wave of E.M. F., or, where this is not available, at a voltage giving the same striking distance between needle points in air as a sine wave of the specified E. M. F., except where expressly specified otherwise. As needles, new seAving-needles should be used. It is recommended to shunt the apparatus during the test by a spark gap of needle-points set for a voltage exceeding the required voltage by 10 per cent. 38. The following voltages are recommended for apparatus, not including transmission lines or switchboards : Rated Terminal Voltage. Capacity. Testing Voltage. Not exceeding 400 volts Under 10 k. W. . 1000 volts " " " 10 k. w. and over 1500 " 400 and over, but less than 800 volts. Under 10 k. w. . 1500 " " " " " 10 k. w. and over 2000 " 800 " 1200 " Any ..... 3500 " 1200 " 2500 " Any ..... 5000 " < Double the nor- 2500 " .... Any ..... j mal rated ( voltages. Synchronous motor fields and fields of converters started from the alternating current side 5000 volts. Synchronous motors and synchronous converter field-coils should be tested at 5000 volts, since in the starting of such machines a high voltage is induced in their field-coils. wm^^^^^mi^m^mmmma^ REPORT OF COMMITTEE ON STANDARDIZATION. 301 Alternator field circuits should be tested under a breakdown test voltage corresponding to the rated voltage of the exciter referred to an output equal bo the output of the alternator ; i.e., the exciter should be rated for this test as having an output equal to that of the machine it excites. Condensers should be tested at twice their rated voltage, and at their rated frequency. The above values are effective values, or square roots of mean square, reduced to a sine wave of E. M. F. 30. In testing insulation between different electric circuits, as between primary and secondary of transformers, the testing voltage must be chosen corresponding to the high-voltage circuit. •A©. In transformers of from 10,000 volts to 20,000 volts, it should be con- lidered as sufficient to operate the transformer at twice its rated voltage, by connecting first the one, and then the other terminal of the high-voltage winding to the core and to the low-voltage winding. The test of dielectric resistance between the low-voltage winding and the core should be in accordance with the recommendation in Section 39 for similar voltages and capacities. 41. When machines or apparatus are to be operated in series, so as to employ the sum of their separate E.M. F.'s, the voltage should be referred to this sum, except where the frames of the machines are separately insu- lated, both from ground and from each other. 43. The term regulation should have the same meaning as the term " in- herent regulation," at present frequently used. 43. The regulation of an apparatus intended for the generation of con- tant potential, constant current, constant speed, etc., is to be measured by the maximum variation of potential current, speed, etc., occurring within the range from full load to no load, under such constant conditions of opera- tion as give the required full-load values, the conditions of full load being considered in all cases as the normal condition of operation. : 4-4. The regulation of an apparatus intended for the generation of a potential, current, speed, etc., varying in a definite manner between full load and no load, is to be measured by the maximum variation of potential, current, speed, etc., from the satisfied condition, under such constant con- ditions of operation as give the required full-load values. If the manner in which the variation in potential, current, speed, etc., between full load and no load, is not specified, it should be assumed to be a simple linear relation. The regulation of an apparatus may, therefore, differ according to its qualification for use. Thus, the regulation of a compound wound generator specified as a constant-potential generator, will be different from that it possesses when specified as an over-compounded generator. 45». The regulation is given in percentage of the full-load value of poten- tial, current, speed, etc., and the apparatus should be steadily operated dur- ing the test under the same conditions as at full load. 40. The regulation of generators is to be determined at constant speed ; of alternating apparatus at constant impressed frequency. 4?. The regulation of a generator unit, consisting of a generator united with a prime mover, should be determined at constant conditions of. the jrime mover ; i.e., constant steam pressure, head, etc. It would include the nherent speed variations of the prime mover. For this reason the regula- tion of a generator unit is to be distinguished from the regulation of either the prime mover, or of the generator contained in it, and taken separately. 48. In apparatus generating, transforming, or transmitting alternating currents, regulation should be understood to refer to non-inductive load ; that is, to a load in which the current is in phase with the E. M. F., at the output side of the apparatus, except Avhere expressly specified otherwise. 40. In alternating apparatus receiving electric power, regulation should refer to a sine wave of E.M. F., except where expressly specified otherwise. 50. In commutating machines, rectifying machines, and synchronous machines, as direct-current generators and motors, alternating-current and aolyphase generators, the regulation is to be determined under the follow- ng conditions : a. At constant excitation in separately excited fields, 6. With constant resistance in shunt-field circuits, and 302 DYNAMO AND MOTOR STANDARDS AND TESTING. c. With constant resistance shunting series fields ; i. e., the field adjust- ment should remain constant, and should be so chosen as to give the required full-load voltage at full-load current. 51. In constant-potential machines the regulation is the ratio of the maximum diit'erence of terminal voltage from the rated full-load value (occurring within the range from full-load to open circuit), to the full-load terminal voltage. 53. In constant-current machines, the regulation is the ratio of the maxi- mum difference of current from the rated full-load value (occurring within the range from full-load to short circuit), to the full-load current. 53. In constant-power machines, the regulation is the ratio of maximum difference of power from the rated full-load value (occurring within the range of operation specified) to the rated power. 5-1. In over-compounded machines, the regulation is the ratio of the maximum difference in voltage from a straight line connecting the no-load and full-load values of terminal voltage as function of the current, to the full-load terminal voltage. 55. In constant-speed continuous-current motors, the regulation is the ratio of the maximum variation of speed from its full-load value (occurring within the range from full load to no load) to the full-load speed. 5©. In transformers, the regulation is the ratio of the rise of secondary- terminal voltage from full load to no load (at constant primary impressed terminal voltage), to the secondary terminal voltage. St. In induction motors, the regulation is the ratio of the rise of speed from full load to no load (at constant impressed voltage), to the full-load speed. The regulation of an induction motor is, therefore, not identical with the slip of the motor, which is the ratio of the drop in speed from synchronism to synchronous speed. 5S. In converters, dynamotors, motor-generators, and frequency chan- gers, the regulation is the ratio of the maximum difference of terminal volt- age at the output side from the rated full-load voltage (at constant impressed voltage and at constant frequency), to the full-load voltage on the output side. 50. In transmission lines, feeders, etc., the regulation is the ratio of max- imum voltage difference at the receiving-end, between no-load and full non- inductive load, to the full-load voltage at the receiving-end, with constant voltage impressed upon the sending-end. ©O. In steam engines, the regulation is the ratio of the maximum varia- tion of speed in passing from full load to no load (at constant steam pressure at the throttle), to the full-load speed. ©1. In a turbine or other water-motor, the regulation is the ratio of the maximum variation of speed from full load to no load (at constant head of water ; i.e., at constant difference of level between tail-race and head-race), to the full-load speed. Variation and Pulsation. ©2. In prime movers which do not give an absolutely uniform rate of rotation or speed, as in steam engines, the "variation" is the maximum angular displacement in position of the revolving member from the position it would occupy at uniform rotation, expressed in degrees, that is, with one revolution at 300° ; and the pulsation is the ratio of the maximum change of speed in an engine cycle to the average speed. ©3. In alternators or alternating-current circuits in general, the varia- tion is the maximum difference in phase of the generated wave of E. M. F. from a wave of absolutely constant frequency, expressed in degrees, and is due to the variation of the prime mover. The pulsation is the ratio of the maximum change of frequency during an engine cycle to the average fre- quency. n ©4. If n = number of poles, the variation of an alternator is — times the variation of its prime mover if direct connected, and ~p times the variation of the prime mover, if rigidly connected thereto in the velocity ratio p. ©5. The pulsation of an alternating-current circuit is the same as the pulsation of the prime mover of its alternator. m^^m^^^MHHIMMHB REPORT OF COMMITTEE ON STANDARDIZATION. 303 RATI\fi, 66. Both electrical and mechanical power should he expressed in kilo- watts, except when otherwise specified. Alternating-current apparatus should be rated in kilowatts on the basis of non-inductive condition ; i. e., with the current in phase with the terminal voltage. 69. Thus the electric power generated by an alternating-current appara- tus equals its rating only at non-inductive load, that is, when the current is in phase with the terminal voltage. ©S. Apparent power should be expressed in kilovolt-amperes, as distin- guished from real power in kilowatts. ©i>. If a power factor other than 100% is specified, the rating should be expressed in kilovolt amperes and power-factor at full-load. 9©. The full-load current of an electric generator is that current which, with the rated full-load terminal voltage, gives the rated kilowatts, but in alternating-current apparatus only at non-inductive load. 91. Thus, in machines in which the full-load voltage differs from the no- load voltage, the full-load current should refer to the former. If P = rating of an electric generator, and Ez=. full-load terminal voltage, the full-load current is : p 1= — in a continuous-current machine or single-phase alternator. 1= — — - in a three-phase alternator. p . -2E1 9*3. Constant-current machines, such as series arc-light«generators, should be rated in kilowatts based on terminal volts and amperes at full load. 93. The rating of a fuse or circuit breaker should be the current strength at which it will open the circuit, and not the working-current strength. Classification of "Voltag-es and frequencies. 94-. In direct-current, low-tension generators, the following average ter- minal voltages are in general use, and are recommended : 125 volts. 250 volts. 550 volts. 95. In direct-current and alternating-current, low-pressure circuits, the following average terminal voltages are in general use, and are recom- mended : 110 volts. 220 volts. In direct-current power-circuits for railway and other service, 500 volts may be considered as standard. 9©. In alternating-current, high-pressure circuits at the receiving-end, the following pressures are in general use, and are recommended ; 1,000 volts. 2,000 volts. 3,000 volts. 6,000 volts. 10,000 volts. 15,000 volts. 20,000 volts. 99. In alternating-current, high-pressure generators, or generating sys- tems, the following terminal voltages are in general use, and are recom- mended : 1150 volts. 2,300 volts. 3,450 volts. These pressures allow of a maximum drop in transmission of 15% of the pressure at the receiving-end. If the drop required is greater than 15%, the generator should be considered as special. 98. In alternating-current circuits, the following approximate frequencies are recommended as desirable : 25^. or 30~ 40~ 60^ 120^.* These frequencies are already in extensive use, and it is deemed advisable to adhere to them as closely as possible. Overload Capacities. 90. All guaranties on heating, regulation, sparkling, etc., should apply to the rated load, except where expressly specified otherwise, and in alter- * The frequency of 120 ~ may be considered as covering the already existing com- mercial frequencies between 120 <-%- and 140 -— ', and the frequency of 60'"w as covering the already existing commercial frequencies between 60^ and 70--*'. 304 DYNAMO AXD MOTOR STANDARDS AND TESTING. nating-current apparatus to the current in phase with the terminal E.M.F., except where a phase displacement is inherent in the apparatus. 50. All apparatus should be able to carry a reasonable overload with- out self-destruction by heating, sparking, mechanical weakness, etc., and with an increase of temperature elevation not exceeding 15° C. above those specified for full loads. (See Sees. 25 to 31.) 51. Overload guaranties should refer to normal conditions of operation regarding speed, frequency, voltage, etc., and to non-inductive conditions in alternating apparatus, except where a phase displacement is inherent in the apparatus. S°2. The following overload capacities are recommended : — 1st. In direct-current generators and alternating-current generators, 25% for one-half hour. 2d. In direct-current motors and synchronous motors, 25% for one-half hour, 50% for one minute except in railway motors and other apparatus intended for intermittent service. 3d. Induction motors, 25% for one-half hour, 50% for one minute. 4th. Synchronous converters, 50% for one-half hour. 5th. Transformers, 25% for one-half hour ; except in transformers con- nected to apparatus for which a different overload is guaranteed, in which case the same guaranties shall apply for the transformers as for the appa- ratus connected thereto. 6th. Exciters of alternators and other synchronous machines, 10% more overload than is required for the excitation of the synchronous machine at its guaranteed overload, and for the same period of time. APPENDIX I. Efficiency of Phase-Displacing* Apparatus. In apparatus producing phase displacement, as, for example, synchronous compensators, exciters of induction generators, reactive coils, condensers, polarization cells, etc., the efficiency should be understood to be the ratio of the volt-ampere activity to the volt-ampere activity plus power loss. The efficiency may be calculated by determining the losses individually, adding to them the volt-ampere activity, and then dividing the volt-ampere activity by the sum. 1st. In synchronous compensators and exciters of induction generators the determination of losses is the same as in other synchronous machines under Sections 10 and 11. 2d. In reactive coils the losses are molecular friction, eddy losses, and I2r loss. They should be measured by wattmeter. The efficiency of reac- tive coils should be determined with a sine wave of impressed E.M.F., except where expressly specified otherwise. 3d. In condensers, the losses are due to dielectric hysteresis and leakage, and should be determined by wattmeter with a sine wave of E.M.E. 4th. In polarization cells, the losses are those due to electric resistivity and a loss in the electrolyte of the nature of chemical hysteresis, and are usually very considerable. They depend upon the frequency, voltage, and temperature, and should be determined with a sine wave of impressed E.M.F., except where expressly specified otherwise. APPENDIX II. Apparent Efficiency. In apparatus in which a phase displacement is inherent to their operation, apparent efficiency should be understood as the ratio of net-power output to volt-ampere input. Such apparatus comprise induction motors, reactive synchronous convert- REPORT OF COMMITTEE ON STANDARDIZATION. 305 ers, synchronous converters controlling the voltage of an alternating-cur- rent system, self-exciting synchronous motors, potential regulators, and open magnetic circuit transformers, etc. Since the apparent efficiency of apparatus generating electric power de- pends upon the power factor of the load, the apparent efficiency, unless otherwise specified, should be referred to a load power-factor of unity. APPENDIX III. Power Factor and Inductance Factor. The power factor in alternating circuits or apparatus may be defined as the ratio of the electric power in watts to volt-amperes. The inductance factor is to be considered as the ratio of wattless volt- amperes to total volt-amperes. Thus, if p — power factor, q = inductance factor, then p2 -(- q2 =. 1. The power factor is the (energy component of current or E.M.F.) total current or E.M.F. and the inductance factor is the (wattless component of current or E.M.F.) true power (total current or E.M.F.; " volt-amperes ' Since the power-factor of apparatus supplying electric power depends upon the power-factor of the load, the power-factor of the load should be considered as Unity, unless otherwise specified. APPENDIX IV. The following notation is recommended : — JE, e, voltage, E.M.F., potential difference, E, r, resistance, 1, i, current, X, x, reactance, P, power, Z, z, impedance, , magnetic flux, L, I, inductance, (£, magnetic density, C, c, capacity. Vector quantities, when used, should be denoted by capital italics. APPENDIX V. Table of sparking distances in air between opposed sharp needle-points, for various effective sinusoidal voltages, in inches and in centimeters. Kilovolts Distance. Sq. Root of Mean Square. Inches. Cms 5 0.225 0.57 10 0.47 1.19 15 0.725 1.84 20 1.0 2.54 25 1.3 3.3 30 1.625 4.1 35 2.0 5.1 40 2.45 6.2 45 2.95 7.5 50 3.55 9.0 Kilovolts Sq. Root of Mean Square. 60 70 80 90 100 110 120 130 140 150 Distance. Inches. Cms. 11.8 14.9 18.0 21.2 24.4 27.3 30.1 32.9 35.4 4.65 5.85 7.1 8.35 9.6 10.75 11.85 12.95 13.95 15.0 306 TESTS OF DYNAMOS AND MOTORS. TESTS OF DYNAMOS AIV» MOTORS. All reliable manufacturers of electrical machinery and apparatus are now provided with the necessary facilities for testing the efficiency and other properties of their output, and where the purchaser desires to confirm the tests and guaranties of the maker, he should endeavor to have nearly, and in some cases all such tests carried out in his presence at the factory, unless he may be equipped with sufficient facilities to enable him to carry out like tests in his own shops after the apparatus is in place. Some tests, such as full load and overload, temperature, and insulation (except dielectric) tests are best made after the machinery has been installed and is in full running order. Owing to the ease and accuracy with which electrical measurements can be made, it is always more convenient to make use of electrical driving power for dynamos, and electrical load for the dynamo output, and in the case of motors, a direct-current dynamo with electrical load makes the best load for belting the motor to. No really accurate tests of dynamo efficiencies can be made with water- wheels, and only slightly better are those made by steam-engines, owing to unreliability of friction cards for the engine itself and the change of fric- tion with load. "Where it is necessary to use a steam-engine for dynamo testing, all fric- tion and low load cards should be taken with the steam throttled so low as to cut off at more than half stroke, and to run the engine at the same speed as when under load. The tests of the engine as separated from the dynamo are as follows : — a. Friction of engine alone. b. Friction of engine and any belts and countershaft between it and the dynamo under test. Consult works on indicators and steam-engines for instructions for deter- mining power of engines under various conditions. The important practical tests for acceptance by the purchaser, or to deter- mine the full value of all the properties of dynamos and motors, are to learn the value of the following items : — Rise of temperature under full load. Insulation resistance. Dielectric strength of insulation. Regulation. Overload capacity. Efficiency, core loss. Bearing friction, windage and brush friction. I2R loss in field and field rheostat, I2R loss in armature and brushes. Note. — If a separate exciter goes with the dynamo, its losses will be determined separately as for a dynamo. Methods of determining each of the above-named items will be described, and then the combinations of them necessary for any test will be outlined. Temperature. — The rise of temperature in a dynamo, motor, or transformer, is one of the most important factors in determining the life of such piece of apparatus; and tests for its determination should be carried out according to the highest standards that can be specified, and yet be within reasonable range of economy. The A. I. E. E. standards state the allowable rise of temperature above surrounding air for most conditions, but special conditions must be met by special standards. For instance, no ordinary insulation ought to be subjected to a degree of heat exceeding 212° F., or 100° C. And yet in the dynamo-room of our naval vessels the temperature is said to at times reach '130° F., or even higher, which leaves a small margin for safety. It is obvious that specifications for dynamos in such locations should call for a much lower temperature rise in order to be safe under full load. For all practical temperature tests it is sufficient to run a machine under its normal full-load conditions until it has developed its highest temperature, although at times a curve of rise of temperature may be desired at various loads. TEMPERATURE. 307 All small dynamos, motors, and transformers, up to, say, 50 KW., will reach maximum temperature in rive hours run under full load, if the tem- perature rise is normal ; but larger machines sometimes require from 6 to 18 hours, although this depends quite as much on the design and construction of the apparatus as on size, as, for instance, the 5,000 h.p. Niagara Falls Gen- erators reach full temperature in live hours. Temperature tests can be shortened by overloading the apparatus for a time, thus reaching full heat in a shorter period. On dynamos and motors the temperatures of all iron or frame parts, com- mutators, and pole-pieces, have to be taken by thermometer laid on the surface and covered by waste. Note that when temperatures are taken with the machine running, care must be taken not to use enough waste to influence the machine's radiation. Where there are spaces, as air spaces, in armature cores or in the held laminations, that will permit the insertion of a thermometer, it should be placed there. Temperature of field coils should be taken by thermometer laid on the surface and covered with waste, and by taking the" resistance of the coils first at the room temperature and again while hot immediately after the heat run. Temperature rise of arma- ture windings can be taken by surface measurement and by the resistance method also ; although being nearly always of low resistance, very careful tests by fine galvanometer and very steady current are required in order to get anything like accurate results. The formula for determining the rise of temperature from the rise of resistance is as follows : — Temperature l»y rise in resistance; for copper. — The in- crease in resistance due to increase in temperature is 0. 4% for each degree Cent, above zero, the resistance at zero being taken as the base. If then tx = temperature of copper when cold resistance is measured, Rx = resistance at temperature tx, t2 = temperature of copper when hot resistance is taken, B2 = resistance at temperature U, Then first reducing to zero degrees, we have The increase in resistance from 0 to t2 degrees is R2 — R0l and hence we have for final temperature, U - R2~Bo ^ 004 (2)< Substituting (1) _ J?2(1 + .00^1)-JZ1 It is usually most convenient to correct all cold resistances to a tempera- ture of 20° C, in which case we first reduce to zero and then raise to 20°. The general formula for obtaining the resistance at t degrees is JRt = (1 + .004 t) B0. Hence i?20 = 1.08 JR0 and in terms of the cold resistance at temperature t. _ (1-08 Z?n) 2 (1 + .004 1) w# Formula (3) then becomes, when the cold resistance is at 20°, 1.08 i?„ 1 TU t* = mxX0-ynamios, Shunt or Compound, and Alternators. The dynamo must be run for a sufficient length of time at a heavy load to raise its temperature to its highest limit ; the field rheostat is then adjusted, starting with voltage a little low, and bringing up to proper value to obtain the standard voltage at the machine terminals, and since a constant temper- ature condition has been reached, must not again be adjusted during the test. Adjust the brushes, in the case of a D. C. machine, for full-load con- ditions, and they should not receive other adjustment during the test. This is a severe condition, and not all machines will stand it ; but all good dy- namos, Avith carbon brushes, Avill stand the test very well, provided the brushes are adjusted at just the non-sparking point at no load. Load is now decreased by regular steps, and when the current has settled the following readings are taken : — Speed of dynamo (adjusted at proper amount). Current in output (a non-inductive load should be vised). If alternator, current in each line if more than single-phase. Volts at machine terminals. Amperes, field. Volts, field. Note sparking at the brushes (they should not spark any with carbon brushes). Readings should be taken at at least ten intervals, from full load to open circuit (no load) ; and load should then be put on gradually and by the same steps as it was brought down ; and the same records should be made back to full-load point, and beyond to 25% overload. If the readings are to be plotted in curves, as they always should be, it will make little difference if the intervals or steps are not all alike ; and should the steps be overreached in adjusting the load, the load must not, in any circumstances, be backed up or readjusted back to get regular inter- vals or a stated value, as the conditions of magnetization change, and throw the test all out. In case the current is broken, or the test has to be slowed down in speed or stopped, it must be commenced all over again. Finally, when the curves are plotted, draw, in the case of a compound-wound ma- chine, a straight line joining the no-load voltage and the full-load voltage ; and the ratio of the point of maximum departure of the voltage from this line to the voltage indicated by the line at the point will be the regulation of the machine. • The readings as obtained give what is called a field compounding curve. In the case of a shunt or separately excited machine, the procedure for the test is the same ; but when the curve is plotted, the regulation is figured as equal to the difference between the no-load voltage and full-load voltage, divided by the full-load voltage. The curve is called a characteristic in this case. DYNAMO EFFICIENCY. 311 Regulation Tests, Motors, Shunt, Compound, and Induction. After driving the motor under heavy load for a length of time sufficient to develop its full heat, full-rated load should be applied, the field rheostat, if any is used, and brushes adjusted for the standard conditions ; then the load should be gradually removed by regular steps, and the following read- ings be made at each such step : — Amperes, input. Volts at machine terminals (kept constant). Watts, if induction motor. Speed of armature. Note sparking at brushes. Amperes, field (in D. C. machines). At least ten steps of load should be taken from full-rated load to no load. The ratio of the maximum drop in speed between no-load and full-load, which will be at full-load, to the speed at full-load, is the regulation of the motor. Efficiency Tests. Urnamos. The term efficiency has tAvo meanings as applied to dynamos ; viz., electrical and commercial. The electrical efficiency of a dynamo is the ratio of elec- trical energy delivered to the line at the dynamo terminals to the total electri- cal energy produced in the machine. The commercial efficiency of a dynamo is the ratio of the energy delivered at the terminals of the machine to the total energy supplied at the pulley. Otherwise the electrical efficiency takes into account only electrical losses, while the commercial efficiency includes all losses, electrical, magnetic, and frictional. Core-loss Test, and Test for friction and Windage. These losses are treated together for the reason that all are obtained at the same time, and the first can only be determined after separating out the others. A core-loss test is ordinarily run only on new types of dynamos and motors, but is handy to know of any machine, and if time and the facilities are available, should be run on acceptance tests by the consulting engineer. It consists in running the armature at open circuit in an excited field, driv- ing it by belt from a motor the input to which, after making proper deduc- tions, is the measure of the power necessary to turn the iron core in a field of the same strength as that in which it will work when in actual use. Connect as in the following diagram, in which A is the dynamo or motor under test, and B is the motor driving the arma- ture of A by means of the belt. The field of A must, of necessity, be separately excited, as its own armature circuit must be open so that there may be no current generated in its conduc- tors. Fig. 3. Connections for a test of core loss. The motor field is sep- aratelv excited and kept constant, so that its losses and the core loss ot tne motor' itself being constant for all conditions of the test, may be cancelled in the calculations. The motor B should be thoroughly heated ; and bear- ings should be run Ion? enough to have reached a constant friction condi- tion before start imi tins test, so that as little change as possible will taite place in the different "constant" values. It is necessary to know accu- rately the resistance of the armature, B, in order to determine its I-R loss at different loads, and to use copper brushes to practically eliminate the It is well t(f make a test run with the belt on in order to learn at what speed it is necessary to run the motor in order to drive the armature A at its proper and standard speed. UNDER. TEST 312 TESTS OF D/NAMOS AND MOTORS. Friction, core loss, and windage of motor. — The speed having been determined, the belt is removed, and the motor field kept at its final adjustment, and enough voltage is supplied to the motor armature to drive it free at the standard speed. The watts input to the armature is then the measure of the loss (I2li) in the motor armature plus the friction of its bear- ings, plus its windage, plus core loss, or the total loss in the motor at no load. This is called the " running light " reading. friction and windage of dynamo. —After learning the losses in the driving motor, the belt is put on and the dynamo is driven at its standard speed without excitation, and in order to be sure of this a volt- meter may be connected across the armature terminals ; if the slightest indication of pressure is found, the dynamo field can be reversely excited, to be demagnetized, by touching its terminals momentarily to a source of E.M.F. Take a number of readings of the input to the motor in order to obtain a good mean, and the friction and windage of dynamo is then the input to the motor, less the " running light" reading previously obtained, the I2R of motor armature having been taken out in each case. Let Wx = watts input to motor, nx — I2R in motor armature when driving dynamo, /=: " running light" reading of motor, }\ = friction and windage of dynamo armature, n9 = I2R of motor armature when " running light," then /i = Wx - (% +/ +/i + n2) . ISrnsh friction. — The friction of brushes is ordinarily a small portion of the losses ; but when it is desirable that it should be separated from other losses, it can be done at the same time and in the same manner as the test for bearing friction. The brushes can be lifted free from the commutator or collector rings when the readings of input to the driving motor for bearing friction are taken ; dropping the brushes again onto the commutator and taking other readings, the difference between these last readings and those taken with brushes off will be the value of brush friction. Note, that allow- ance must be made as before for increase of I2R loss in the motor armature. Test for core loss. — Having determined the friction and other losses that are to be deducted from the total loss, a current as heavy as will ever be used is put on the dynamo field, the motor is supplied with current enough to drive the dynamo at its standard speed, and the reading of watts and current input to the motor armature is taken. The dynamo field current is now gradually decreased in approximately regular steps, readings of the input to the motor being taken at each such step until zero exciting current is reached, when the exciting current is reversed and the current increased in like steps until the highest current reading is again reached. This may now be again decreased by intervals back to zero, reversed and increased back to the starting-point, which will thus complete a cycle of magnetization ; ordinarily this refinement is not, however, necessary. This test must always be carried through without stop ; and although it is desirable to make the step changes in field excitation alike, if the excitation be changed in excess of the regular step it must not be changed back for the purpose of making the interval regular, as it will change the conditions of the residual field. When the readings are plotted on a curve, regularity in intervals of magnetization is not entirely necessary. The following ruling makes a convenient method of tabulation : — Dynamo. Motor. Speed amperes in field Speed amperes in field amperes in armature i volts in armature e Constant Constant. Constant. ^■MHH^^M DYNAMO EFFICIENCY. Computations, 313 watts in armature belt on W„ = ie Kunning light reading / I*R in arm. belt on im in arm. belt off Core loss Plot on curve with exciting-current values on the horizontal scale, and the core loss on the vertical, and the usual core-loss curve is obtained. Separation of core loss into Hysteresis and JEddy current loss. 3 due to hysteresis and friction vary directly with the speed ; losses due to eddy currents vary "as the square of the speed. Current and voltage must now be applied to the dynamo armature to drive it as a motor at proper speed, with the current in the separately excited field kept constant at proper value. Drive the motor (dynamo) at say two different speeds, one of which may be A' times the other ; let L — total loss in watts, f\~=- loss in friction, JI=z loss by hysteresis, D = loss by eddy currents, or L =f1 + H+ D at the first speed, Lx = Kf, + KH-\- K"-D at second g K X (1) = AA = Kf\ + lcH+ KB, (2) — (3) = L, — KL — A2D — KD, L1 — KL = KD{K—\), If K = 2, then D = A(A-l) _ LX — 2L CD (2) (3) (4) (6) (G) 2(2-1) 2 Kapp and Housman separately devised the above method of separating the losses, but stated them somewhat differently. With the field separately excited at a constant value, different values of current are supplied to the armature at different voltages, to drive it as a, motor. The results are plotted in a curve which is a straight line, rising as the volts are increased. The following diagram shows how the losses are plotted in curves. The test as a separately excited motor is run at a number of different values of voltage and current in the armature, and the results are plotted in a curve as shown in the following diagram. The line a, b, is plotted from the results of the current and volt readings. The line a, c, is then drawn parallel to the base, and represents the sum of all the other losses, as shown by previous tests, and they may be further separated and laid off on the chart. Foucault currents are represented in value by the triangle a, c, b. If another run be made with a different value of excitation, a curve, %, 61} or one below the original a, b, will be gotten, according to whether the total losses have been increased or decreased. If the higher values of current tend to demagnetize, by reason of the eddy currents in the armature, the curve a, b, will curve upward somewhat at the upper end. It is thus seen how to measure core-loss, and friction and windage of a dynamo; knowing this and the resistance of the various parts, the efficiency is quickly calculated, thus Let W= core-loss -f- friction (obtained as shown), V = voltage of armature, I=i current of dynamo armature, Ix =r current of dynamo field, R = resistance of armature and brushes, jBx — resistance of field. 314 TESTS OF DYNAMOS AND MOTORS. FOUCAUlT currents HYSTERESIS BOSH FRICTION Then considering the above as the only losses (i.e., neglecting rheo- stats, etc.), Vc EfficienCy = Vc+ItR + VRi+W This is the simplest method of getting the efficiency, but does not take in "load losses" «if any should exist. Anotnea' test for efficiency. — It' the dy- namo under test is not of too large capacity, and a load for its full output is available, either in the form of a lamp bank, water rheostat, or other adjustable resistance, then one form of test is to belt it to a motor. By separately exciting the motor fields, and run- ning the motor free with belt off, its friction can be determined, and with the resistance of the ar- mature known, the input to the motor in watts, less the friction and the I2R loss in its armature at the given load, is a di- rect measure of the power applied at the pul- ley of the dynamo. The output in watts, meas- ured at the dynamo terminals, then measures the efficiency of the machine. Let, W= watts input to motor, I = losses in motor, friction, I-R, and core-loss, W1 = watts output at dynamo terminals. % of efficiency = 100 X ,,r _|_ . = commercial efficiency. Knowing the current flowing in the armature and in the fields, and also knowing the resistance of the same, the I2R losses in each may be calcu- lated, which, added to the output at the dynamo terminals, shows the total electrical energy generated in the ma- chine. If in ■= the 7 2R loss in the armature, /= the I2R loss in the fields, The electrical efficiency will be % electrical efficiency BEKBINQ. FRICTION AND WINDAGE VOLTS IN ARMATURE". d Fig. 4. Diagram showing separation of losses in dynamos. :100 ) "', W^ + m+f GENERATOR WATER RHEOSTAT. J FOB LOAD The following diagram shows the connections for this form of test. It must be obvious that a steam-en- gine, or other motive power that can be accurately measured, may be used in place of the electric motor ; but measurements of mechanical power Fig. 5. Connections for efficiency are so much more liable to error that test of a generator. Driven by an they should be avoided where possible. electric motor. The only objection to this method is that the friction of the driving-motor varies with the load, and the loss in the belt is not considered. DYNAMO EFFICIENCY. 315 Kapp's Test with two Similar I)j-nanij due to the fact that both machines are not running same load, since that one supplies the losses of both. Eleming-'s Modification of Hopkin- son Test. — In this case the two dynamos under test are connected together by belt or shafts, and are driven electrically by an external source of current, say a storage battery or another dynamo, which is connected in series with the circuit of the two machines. Figure 12 shows the con- nections for this test, which will be found car- ried out in full in Fleming's " Electrical Labo- ratory Notes and Forms." MOTOR TESTS. Fig. 12. PRONY BRAKE Fig. 13. Probably the most common method of testing the efficiency and capa- city of motors is with the prony brake, although in factories where spare dynamos are to be had, with load available for them, there can be no question that belting the motor to the dynamo with an electrical load is by far the most accurate, and the easiest to carry out. Prony brake test. — In this test a pulley of suitable dimensions is applied to the motor-shaft, and some form of friction brake is applied to the pulley to absorb the power. The following diagram shows one of the simplest forms of prony brake ; but ropes, straps, and other appliances are also often used in place of the wooden brake shoes as shown. Note. — See Flather, " Dyndmmeters and the Measurement of Power.''' As the friction of the brake creates a great amount of heat, some method of keeping the pulley cool is necessary if the test is to continue any length of time. A pulley with deep inside flanges is often used ; water is poured into the pulley after it has reached its full speed, and will stay there by reason of the centrifugal force until it is evaporated by the heat, or the speed is lowered enough to let it drop out. Rope brakes with spring bal- ances are quite handy forms. The work done on the brake per m inute is the product of the following items: I = the distance from the centre of the brake pulley to the point of bearing on the scales, in feet, n — number of revolutions of the pulley per second, Power : H.P.: : weight in lbs. of brake bearing on scales. -2tt I ni» r I mo ~550 : foot-pounds per second, and The input to the motor is measured in watts, and can be reduced to horse- power by dividing the watts by 74G ; or the power absorbed by the brake can be reduced to watts as follows : — If the length, Z, be given in centimeters, and the weight, w, be taken in grams, the power absorbed by the brake is measured directly in ergs, and as one watt = 107 ergs, the Watts output at the brake = — ^ — = W. The watts input = If the output is measured in I W=2.72 W, Input in h.p. = —^ Output H.P. = °m 107 W . ' W, - feet and w = lbs., then • I w. = h.p. 550 h.p. MOTOR EFFICIENCY. 323 If it is desired to know the friction and other losses in the motor, after the brake test has been made, the brake can be removed, and the watts neces- sary to drive the motor at the same speed as when loaded, can be ascertained. Electrical load test (including loss in belting, and extra loss in bear- ing s due to pull of belt) . — This test 'consists in belting a generator to the motor and measuring the electrical output of the generator, which added to the friction and other losses in the generator, makes up the load on the motor. The efficiency is then measured as before, by tbe ratio of output to input. The great advantage of this form of test is, that it can be carried on for any length of time Avithout trouble from heat, and the extra loss in bearings due to pull of belt is included, which is therefore an actual com- mercial condition. In this form of test the losses in the generator are termed counter torque, and the method of determining them is given following this. Counter torque. — In tests of some motors, especially induction mo- tors, the load is supplied by belting the motor under test to a direct current generator having a capacity of output sufficient to supply all load, including overload. , In determining the load applied to the motor and the counter torque, it is necessary to know, besides the /. E. or watts output of the generator, the following : — I2R of generator armature, Core loss of generator armature, Bearing and brush friction and windage of generator, Extra bearing friction due to belt tension. It is necessary to know the above items for all speeds at which the com- bination may have been run during the testing. This is especially useful in determining the breakdown point on induction and synchronous motors, both of which can be loaded to such a point that they " fall out of step." While the motor is under test especial note should be made of the speeds at which the motor armature and generator armature rotate, and of the watts necessary to drive the motor at the various speeds without load. The counter torque will then be the sum of the following three items : — W = 1 2R of generator armature, Wc = core loss of generator armature, F = bearing and brush friction and windage of the generator armature. The field of the D. C. machine must be separately excited and kept at the same value during the load tests and the tests for" stray power " and does not enter into any of these calculations. Belt-on test. — After disconnecting current from the motor under test, and with the belt or other connection still in place, supply sufficient volt- age to the D. C. machine armature to drive it as a motor at the speeds run during the motor test, holding the field excitation to the same value as before, but adjusting the voltage supplied to the armature for changing the speed. Take readings of Speed, i.e., number of revolutions of D. C. armature, Volts at D. C. armature, Amperes at D. C. armature. Construct a curve of the power required to drive the combination at the various speeds shown during the motor test. Belt-off test. — Throw the belt or other connection off, and take read- ings similar to those mentioned above, which will show the power necessary to drive the D. C. machine without belt. Then for any speed of the combination the " stray power'" will be found as follows : — W, =■ watts from belt-off curve, required fo drive the D. C. machine as a motor. W// = watts from belt-on curve, required to drive the combination. Wc = core loss in D. C. armature. F=z friction of D. C. machine, belt off. F, = friction of motor under test, running light and without belt. / = increase in bearing friction of D. C. machine, due to belt tension. fj = increase in bearing friction of motor, due to belt tension. 324 TESTS OF DYNAMOS AND MOTORS. From the belt-off curve, W, = wo + F. (1) From the belt-on curve, Wit = Wc + F + Fi +/ +/,. (2) Subtract (1) from (2) W„- W,-F,+f+f. Wit-Wt — Ft =/+/,. (3) The values of / and /, cannot be determined accurately ; but if the ma- chines are of about the same size as to bearings and weights of moving parts, it is very close to call them of equal value, when, C Wit — W, — F,) /or/, = ^ lJ *l) (4) The friction F, of the motor under test has been previously found by noting the watts necessary to drive it at the various speeds. If it is an in- duction motor, the impressed voltage is reduced very low in determining the friction in order that the core loss may be approximately zero. As all the values of the quantities on the right-hand side of the equation (4) are now known, /is determined, and may be added to W, to give the total " stray power." A curve is then plotted from the values of " stray poioer " at different speeds. Counter torque = W,-\- f +, Total load = I E + IUi + ( W, -f /), Where I E =z watts load on the D. C. machine when it is being driven by the motor, If S = W, -f /= ^ stray poioer" then Total load = I.E.+ I*R + S. The value of / is so small when compared with the total load, that any ordinary error in its determination will cut no figure. Test of Street-Railway Motors. The " pumping -bach " test, as described before, with some little modifica- tion serves for testing street-railway motors. The following diagram shows the arrangement and electrical connections. The motors are driven mechanically by another motor, the input to which is a measure of the losses, frictional, core losses, gears, bearings, etc., in the two motors ; the two motors are connected in series, through a booster, B, care being taken to make the connections in such a manner as to have the direction of rotation the same ; Fig. 14. Diagram of connections and arrange- and their voltages op- ment of street-railway motors, posing. Headings are taken and the efficiencies are calculated as in the " pumping- back " test. In eliminating the friction of bearings, etc., and of the driving-motor, it is run first without belts, the input being recorded as taken, at the speed necessary. The belt is then put on and a reading taken at proper speed, with both the motors under load. The load being adjusted by varying the field of booster B, the total losses of the system are then IE from booster plus the difference between belt-on reading with full load through the motors, and belt-off reading as noted (allowance being made for change of I2R of driving-motor). If the two motors are similar, half this value is the loss in one motor, from which the efficiency can be calculated as previously shown. Induction motors. — In addition to the tests to which the D. C. motor MOTOR EFFICIENCY. 325 is ordinarily submitted, there are several others usually applied to the in- duction motor, as follows : — Excitation ; Stationary impedance; Maximum output ; and some variations on the usual heat and efficiency tests. Excitation : This is also the test for core loss-f- friction, allowance being made for 1*11 of field ; with no belt on the pulley the motor is run at full impressed voltage. Read the amperes of current in each leg, and total watts input. The amperes give the excitatiou or " running-light" current, and the watts give core loss + friction -f- I'2R of excitation current. Stationary impedance: Block the rotor so it cannot move, and read volts and amperes in each leg, and total watts input. This is usually done at half voltage or less, and the current at full voltage is then computed by proportion. This then gives the current at instant of starting, and a meas- ure of impedance from which, knowing the resistance and core loss, other data can be calculated, such as maximum output, efficiency, etc. Maximum output : This might be called a brealc-doion test; as it merely consists in loading the motor to a point where the maximum torque point is passed and thus the motor comes to rest. Keep the impressed voltage constant and apply load, reading volts, am- peres in each leg, the total Avatts input, and revolutions ; also record the ,load applied at the time of taking the input. Then take counter torque as explained before, from which the efficiency, the apparent efficiency, the power factor, and maximum output are immediately calculated. Iff eat test. — Run motor at full load for a sufficient length of time to develop full temperature, then take temperatures by thermometer at the following points : — 1. Room, not nearer to the motor than three feet and on each side of motor. 2. Surface of field laminations. 3. Ducts (field). 4. Field or stator conductors, through hole in shield. 5. Surface of rotor. 6. Rotor spider and laminations. 7. Bearings, in oil. During heat run, read amperes and volts in each line. Efficiency test. — Apply load to the motor, starting with nothing but friction ; make readings at twelve or more intervals, from no load to break- down point. Keep the speed of A. C. generator constant, also the iinpressed voltage at the motor. Read, Speed of motor. Speed of A. C. dynamo. Amperes input to motor, in each leg. Volts impressed at motor terminals. "Watts input to motor, by wattmeter. Current and volts output from D. C. machine belted to motor, Counter torque as explained above, and excitation reading watts. From the above the efficiency, apparent efficiency, power factor ( zr -^- — . — ^—. ) , and maximum output can be calculated. \ real efficiency / In reading watts in three-phase motors, it is best to use two wattmeters, connected as shown in following sketch : — 1, 2, 3, are the three-phase lines leading to the motor. A and B are two wattmeters. b is the current coil of A, and b1 of B. a is voltage coil of A, and a1 of B. The sum of the deflections of A and B give total watts input. At light loads one wattmeter usually reads negative, and the difference is the total watts. Results. — At the end of the preceding tests the following results should be computed, and curves plotted from them. _ Speed of motor x 100. ') synchronism = Synchronous speed. 326 TESTS OF DYNAMOS AND MOTORS. % real efficiency = % apparent efficiency = Power factor - _ Output of motor X 100 Input by wattmeter Output of motor x 100 volt x amperes Watts _ apparent efficiency " Volt X amperes Torque-pounds pull at 1 ft. radius = real efficiency 5,250 H.P. revolutions per minute" The above results should be plotted on a sheet in curves similar to the fol- lowing, taken from Steinmetz's article on " Induction Motors." Fig. 16. Curves of results of tests of induction motor. Synchronous motor. — Synchronous motors are separately excited, and the D. C. exciter should have its qualities tested as a dynamo. Syn- chronous motors are tested for Break-down point ; Starting current at differ- ent points of location of the rotor ; Least exciting current for various loads. All these in addition to the regular efficiency and other tests. Core losses, friction, T^R losses, etc., can be found by any of the usual methods pre- viously described. Break-down point. Synchronous motors have but little starting-torque ; and it is necessary to start them without load, throwing it on gradually after the motor has settled steadily and without " hunting" on its synchro- nous speed. The break-down point is found by applying load to the point where the motor falls out of step, which will be indicated by a violent rush of current in the ammeter simultaneous with the slowing down. This test is usually carried out at about half voltage, the ratio of the load on the motor at the moment of dropping out of step will be to the full load of break-down as the square of the voltages, the load being adjusted at minimum input in each case. For example, say a certain motor, built to run at 2,000 volts, breaks down at 150 K.W., with an impressed voltage of 1,000. Then the true full break -down load will be 2,0002 X 150 = ) K.W. ^■■■■i^^^BlHH SYNCHRONOUS MOTOR. 327 Starting current. Owing to consequent disturbance to the line, it is desi- rable that the starting current of a synchronous motor be cut down to the lowest point ; but it is difficult to reduce this starting current lower than 200% of full-load current. A synchronous motor also starts easier at certain positions of its rotor as related to poles. With the rotor at rest, and the location of the centre of its pole-pieces chalked on the opposite member, the circuit is closed, the impressed voltage is kept constant, and the current flowing in each leg of the circuit is read, and the time to reach synchro- nism. Care should be taken to note the amount of the first rush of current, and then the settling current at speed. Least exciting current. The power factor of a synchronous motor will be 100 only when," with a given load on the motor, the exciting current is ad- justed so that there is neither a leading nor lagging current in the armature. Sometimes it is desirable to produce a leading current in order to balance the effect of induction motors on the line, or inductance of the line itself. This is done by over-exciting the fields. With a given load on the motor, the 100 power-f actor is found by com- paring the amperes in the motor armature with the exciting current in the field. Starting with the excitation rather low, the armature current will be high and lagging ; as the excitation is increased, the armature current will drop, until it reaches a point where, as the excitation is still increased, the armature current begins to rise, and keeps on rising as the exciting current is increased, and on this side of the low point the armature current is leading. With no reason for making a leading current, the best point to run the motor at is, of course, that at which the armature current is the lowest ; and at that point the power-factor is 100. Synchronous Impedance.- The E.M.F. of an alternating dynamo is the resultant of two factors, i.e., the energy E.M.F. and inductive E.M.F. The energy E.M.F. may be determined from the saturation curve by run- ning the machine without load, and learning the field strength necessary to produce full voltage. The inductive E.M.F. is at right angles to the energy E.M.F., and is de- termined by driving the machine at speed, short-circuiting the armature through an ammeter, and exciting the field just enough to produce full-load current in the armature. The amount of field current necessary to produce full load is a measure of the inductive E.M.F., which can be determined from the saturation curve as before, and the resultant E.M.F. will be Resultant E.M.F. = Venergy E.M.F.2 + inductive E.M.F.2. Saturation test. — This test shows the quality of the magnetic cir- cuit of a dynamo, and especially the amount of current necessary to saturate the field cores and yokes to a proper intensity. In this test it is important that the brushes and commutator be in good condition, and that all contacts and joints be mechanically and electrically tight. The dynamo armature must be driven at a constant speed, and the leads from the voltmeter placed to get readings from the brushes of the dynamo must have the best of contacts. The fields of the dynamo must be separately excited, and must have in the circuit with them an ammeter and rheostat capable of adjusting the field current for rather small changes of charge. The armature must be without load, and a Aroltmeter must be connected across its terminals. Should there be residual magnetism enough in the iron to produce any pressure without supplying any exciting current, such pressure should be recorded ; or perhaps a better way is to start at zero voltage by entirely demagnetizing the fields by momentary reversal of the exciting current. To start the test, read the pressure, due to residual magnetism if not de- magnetized, or if demagnetized, start at zero. Give the fields a small ex- citing current, and read the voltage at the armature terminals ; at the same time read the current in the fields, and the revolutions of the armature. Increase the excitation in small steps until tbe figures show that the knee of the iron curve has been passed by several points ; then reverse the operation, decreasing the excitation by like amounts of current, until zero potential is reached. This is usually as far as it is necessary to go in practice ; but occasionally 328 TESTS OF DYNAMOS AND MOTORS. it is well to complete the entire magnetic cycle by reversing the exciting cur- rent, and repeating the steps and readings as above described. The readings should be plotted in a curve with the amperes of exciting current as abscissae, and volts pressure as ordinates. The E.M.F. will be found to increase rapidly at first ; and this increase will be nearly proportional to the exciting current until the " knee " in the curve is reached, when the E.M.F. increase will not be proportional to the excitation until after the "knee" is passed, when the increase in E.M.F. will again become nearly proportional to the excitation, but the increase will be at such a low rate as to show that the magnetic circuit is practically saturated ; and it is not economical to work the iron of a magnetic circuit too far above the knee, nor is it expedient to work it at a point much below the " knee," except for boosters. The exciting current must not be broken during this test, except possibly at zero ; nor must its value be reduced or receded from in case a step should be made longer than intended. Inequalities of interval in steps of excit- ing current will make little difference when all are plotted on a curve. For the same value of exciting current the down readings of E.M.F. will always be higher than those on the up curve. Resistance of field, coils. — The resistance of the shunt fields of a dynamo or motor can be taken in any of the usual ways : by Wheatstone bridge ; by the current flowing and drop of potential across the field termi- nals ; and it is usual, in addition, to take the drop across the rheostat at the same time. The resistance of each field coil should be taken to insure that all are alike. Resistance of series fields, and shunts to the same, must be taken by a dif- ferent method, as the resistance is so low that the condition of contacts may vary the results more than the entire resistance required. The test for re- sistance of armatures following this is quite applicable. Of course any test for low resistances is applicable ; but the one described is as simple as any, and quite accurate enough for the purpose. Resistance of armature. — In order to determine the I2R loss in a generator or motor armature, its resistance must be measured with consider- able care ; and the ordinary Wheatstone bridge method is of no use, for the reason that the variable resistance of the contacts is often more tban that of the armature itself. The drop method, so useful with higher re- sistance devices, is not accurate ^-r K ;, enongh for the work ; and the storage ^L- f-^ — ■ ^a resistance most accurate method is probably battery. :=t {armature. . smfr- the direct comparison with a stan- dard resistance by means of a good galvanometer and a storage battery. Clean the brushes, commutator surface, or surface of the col- lector-rings, and in the case of a D. C. machine, see that opposite Fig. 17. Diagram of arrangement for brushes bear on opposite seg- measuring resistance of armatures, ments. Connect the galvanometer and its leads, the storage battery and resis- tances, as in the following diagram. The standard resistance, R, will ordina- rily be about .01 ohm, but may be made of any size to suit the circumstances. The storage battery must be large enough to furnish practically constant current during the time of testing. The galvanometer must be able to stand the potentials from the battery ; and it is usually better to connect in series with it a high resistance, so that its deflections may not be too high. The deflection of the galvanometer should be as large as possible, and pro- portional to the current flowing. The leads a, ax , and b and 615 are so ar- ranged with the transfer switch that one pair after the other can be thrown in circuit with the galvanometer ; and it is always well to take a deflection first with R, then again after taking a deflection from the armature. The leads a and ^ must be pressed on the commutator directly at the brush contacts, and may often be kept in place by one of a set of brushes Test. — Close the switch, k, and adjust the resistance, r, until the am- meter shows the amount of current desired, and watch it long enough to be ARMATURE FAULTS. 329 STORAGE .BATTERY Fig. 18. Test for break in ar- mature lead. pure it is constant. Close the transfer switch on b and bL, and read the gal- vanometer deflection, calling it d. Throw the transfer switch to the con- tacts a, and a,, read the galvanometer deflection, and call it dv Transfer the contacts back to b, and 6t and take another reading ; and if it differs from dlt take the mean of the two. Let x= resistance of the armature, then * = n% d Note. — See Flemming's " Electrical Laboratory Notes and Forms." Tests for faults in Armatures. The arrangement of galvanometer for testing the resistance of an arma- ture is the very best for searching for faults in the same, although it is not often necessary to measure resistance. (See Figs. 15 to 18 inc., page 7.) Test for open circuit. — Clean the brushes and commutator, then apply current from some outside source, say a few cells of storage battery or low pressure dynamo, through an am- meter as in the following diagrams. Note the current indicated in the ammeter ; ro- tate the armature slowly by hand, and if the break is in a lead, the flow of current will stop when one brush bears on the segment in fault. Note that the brushes must not cover more than a single segment. If on rotating the armature completely around the deflection of the ammeter does not indicate a broken lead, then touch the ter- minals of the galvanometer to two adjacent bars, working from bar to bar. The deflec- tion between any two commutator bars should be substantially the same in a perfect armature ; if the deflection suddenly rises between two bars it is indicative of a high resistance in the coil or a break (open circuit). The following diagram shows the connec- tions. A telephone receiver may be used in place of the galvanometer, and the presence of current will be indicated by a " tick " in the instrument as circuit is made or broken. Test for short circuit. — Where two adjacent commutator bars are in contact, or a coil between two segments becomes short- circuited, the bar to bar test with galvanom- eter will detect the fault by showing no deflection. If a telephone is used, it will be silent when its terminal leads are connected with the two segments in contact. See dia- gram below for connections. If there be a short circuit between two coils the galvanometer terminals should include or straddle three commutator bars. The normal deflection will then be twice that indicated between two segments until the coils in fault are reached, when the deflection will drop. When this happens, test each coil for trouble ; and if indi- vidually they are all right, the trouble' is between the two. The following diagram shows the con- nections. Test for grounded arma- ture. — Place one terminal of the galvanometer on the shaft or frame of the machine, and the other terminal on the commutator. (The Fig. 19. Bar to bar test for open circuit in coil. Fig. 20. Bar to bar test for short cir- cuit in one coil or between commuta- tator segments. 330 TESTS OF DYNAMOS AND MOTORS. SHORT CIRCUIT' BETWEEN SEETKMB storage battery, ammeter, and leads must be thoroughly insulated from ground.) If, under these circumstances, there is any deflection of the gal- vanometer, it indicates the presence of a ground, or contact between the armature conductors and the frame of the machine. Move the terminal about the commutator until the least deflection is shown, and at or near that point will be found the contact ; in the particular coil connected be- tween two segments showing equal deflection, unless the contact happens to be close to one segment, in which case there will be zero deflection. Contacts in field coils can be located by the same method. The following diagram shows the connections. To determine if armature of mullipola, Fig. 21 . Alternate bar test for short circuit between sections. Fig. 22. Test for ground in armature coils. In the above the brushes should 1 - dynamo is electrically centred, put down brushes 1 and 2, and take volt- age of machine ; put down brush 3, and lift 1, take voltage again ; put down brush 4 and lift 2, again tak- ing voltage ; repeat the operation with all the brushes, and the volt- age with any pair should be the same as that of any other pair if the armature is electrically central. The same thing can also be deter- mined by taking the pressure curves all around the commutator as shown in the notes on characteristics on dynamos. exactly at the neutral point. Test for E.M.r. of Xfeynamo without Running- it. Prof. F. B. Crocker gives the following method (page 247 Trans. A. I. E. E., 1897), for determining the E.M.F. of a dynamo without driving it by outside power, provided a current of the proper voltage is at hand sufficient to give it full torque as a motor. Clamp a lever to the pulley, and weigh the torque, as a motor, at radius r, with a spring balance or a platform scale. r = radius of torque lever. s = speed of revolutions per minute, as a dynamo. p = pounds pull at radius r. I— current. E — E.M.F. EI 2tt r sp 746 ~~ 33,000 E- r s p Field strength is the same as if running as a dynamo ; and by tapping the shaft when test is made, friction losses are partially eliminated, and the method is sufficiently correct for all efficiencies. ^^^^M^^H THE STATIC TRANSFORMER. The static transformer is a device used for changing the voltage and cur- rent of an alternating circuit in pressure and amount. It consists, essen- tially, of a pair of mutually inductive circuits, called the primary and secondary coils, and a magnetic circuit interlinked with both the primary and secondary coils. This magnetic circuit is called the core of the trans- former. The primary and secondary coils are so placed that the mutual induction between them is very great. Upon applying an alternating voltage to the primary coil an alternating flux is set up in the iron core, and this alternat- ing flux induces an E.M.F. in the secondary coil in direct proportion to the ratio of the number of turns of the primary and secondary. Technically, the primary is the coil upon which the E.M.F. from the line or source of supply is impressed, and the secondary is the coil within which an induced E.M.F. is generated. The magnetic circuit or core in transformers is composed of laminated sheet iron or steel. The following cuts represent sections of several dif- ferent types. pv. fi 1 1 pJs f SjP ? ( If Si VM FlG. 1. Cores of some American Transformers. p = primary winding ; s = secondary winding. In those showing a double magnetic circuit the iron is built up through and around the coils, and they are usually called the " Shell " type of trans- former. 331 332 THE STATIC TRANSFORMER. Those having a single magnetic circuit, and having the coils built around the long portions or legs of the core, the short portions or yoke connecting these legs at each end, are called " core " type of transformer. The duties of a perfect transformer are : (1) To absorb a certain amount of electrical energy at a given voltage and frequency, and to give out the "same amount of energy at the same frequency and any desired voltage. (2) To keep the primary and secondary coils completely isolated from one another electrically. (3) To maintain the same ratio between impressed and delivered voltage at all loads. The commercial transformer, however, is not a perfect converter of energy, although it probably approaches nearer perfection than any form of appa- ratus used to transform energy. The difference between the energy taken into the transformer and that given out is the sum of its losses. These losses are made up of the copper loss and the core loss. The core loss is that energy which is absorbed by the transformer when the secondary circuit is open, and is the sum of the hysteresis and eddy cur- rent loss in the core, and a slight copper loss in the primary coil, which is generally neglected in the measurements. The hysteresis luss is caused by the reversals of the magnetism in the iron core, and differs with different qualities of iron With a given quality of iron, this loss varies as the 1.6 power of the voltage with constant fre- quency. Steinmetz gives a law or equation for hysteresis as follows : Wh = V (ft 1-6- We. = Hysteresis loss per cubic centimeter per cycle, in ergs (= 10-7 joules). ■q = constant dependent on the quality of iron. If Ar= the frequency, Vz=. the volume of the iron in the core in cubic centimeters, P = the power in watts consumed in the whole core, then P = f]N V (ft1"6 107, 'l-N y (£1-6 10-7- In Table A, on page 13, this hysteresis constant t\ is given for several different transformers. In the construction, the core loss depends on the following factors : (1) Magnetic density, (2) Weight of iron core, (3) Frequency, (4) Quality of iron, (5) Thickness of iron, (6) Insulation between the sheets or laminations. The density and frequency being predetermined the weight or amount of iron is a matter of design. The quality of the iron is very variable, and up to the present time no method has been found to manufacture iron for trans- formers which gives as great a uniformity of results as to the magnetic losses as could be desired. On the thickness of the laminations and the insulation between them de- pend the eddy current losses in the iron. Theoretically1 the best thickness of iron for minimum combined eddy and hysteresis loss at commercial fre- quencies is from .010" to .015", and common practice is to use iron about .014" thick. The copper losses in a transformer are the sum of the I2E losses of both the primary and secondary coils, and the eddy current loss in the conductors. In any well-designed transformer, however, the eddy current loss in the conductors is negligible, so that the sum of the I2R losses of primary and secondary can be taken as the actual copper loss in the transformer. i Bedell, Klein, Thomson, Elec. W., Dec. 31. 1898. ■m^hhmb^h^hm^^HHI 333 o O 8 II X ci 2.90 " 3.04 " 2.10 " 6.57 " 2.12 " 2.22 " 2.24 " 2.06 " 2.28 " ill o X § 3.20 " 3.52 " 2.39 " 6.26 " 2.24 " 2.38 " 2.54 " 2.42 " 2.75 " $ sb g e s ii o 4280 5400 7500 7720 75G0 4080 49G0 8460 6650 Ill O t- CO 8 N © 2050 2600 3G40 3750 3630 1960 2380 4670 3210 III in co co "el o O o S II q 015 011 017 020 067 021 007 031 011 021 018 H * rt t- o o o o ii b- q iHooeooi-iTijia-noooO'* OOOOOOOOOOO 2 £ 5 C^ o o o o a o o n ■*00t>«Oi0^(NCSOO-*Ort*t- — total flux. (§/' ■=. lines of force per square inch. A = section of magnetic circuit in square inches. N = frequency in cycles per second. T= total turns of wire in series. 2tt 4-44 = ^ = VTXjr Then E - 108 (1) This equation is based on the assumption of a sine wave of electromotive force, and is the most important of the formulae used in the design of an alternating current transformer. By substituting and transposing we can derive an equation for any un- known quantity. Thus if the volts, frequency, and turns are known, then — Ex 108 (2) But $ = ($/' A Therefore A z 4.44 X A" X T x ( (3) (4) which equation gives at once the cross section of iron necessary for the magnetic circuit after we have decided on the total primary turns, and the density at which it is desired to work the iron. Again, if the volts, frequency, cross section of core, and density are known, we have, transposing equation (4), Tzzz E X 108 4.44XA^X( 'XA 1 L r i JR 1 or r E : 1 THE STATIC TRANSFORMER. 335 Fig. 2 is a curve giving the total fluxes as ordinates and capacities in k.w. as abscissae. This curve represents approximately common practice for a line of lighting transformers, to be operated at 60 cycles. For any other frequency or for power work, a curve of total fluxes can be drawn after three or more transformers have been calculated with quite widely differing capacities. HEag-netic densities in the cores of transformers vary considerably with the different frequencies and different designs of various makers. The practical limits of these densities are as follows : For 25 cycle transformer from G0,000 to 90,000 C.G.S. lines per square inch. For 60 cycle transformers from 40,000 to 60,000 lines per square inch. For 125 cycles from 30,000 to 50,000 lines per square inch. Densities for other frequencies are taken in proportion. Current densities in transformer windings vary between 1000 and 2000 circular mills per ampere. Some makers design for greater current density in the secondary than in the primary. The circular mils per am- pere in transformers of the best design are often 1000 or 1500 in the primary coil, and 1200 or 2000 for the secondary coil. The proper adjustment of the current density should be such as to give equal heat distribution throughout the coils, and the relative densities in the two coils should be based on their relative radiating surfaces. FEATURES OF DEilGIT. In the design of a successful transformer, the features to be given partic- ular attention are : (1) Insulation between primary and secondary, (2) Heating, (3) Efficiencies, (4) Regulation, (5) Cost, (6) Power factor and excitingcurrent. Insulation. The insulation of a transformer is really a measure of its durability, and it must be obvious that if it is not well designed and properly constructed to prevent the breakdown of its insulation, it is not a good investment ; and the same reasoning holds good if the insulation deteriorates rapidly. Sim- plicity of form and constructive details is a good point, and as transformers are liable to be exposed to all sorts of weather and other conditions, they should always be designed to withstand all of them. Insulation between coils must be of the best possible kind, as electrical connection here is a menace to life and property, and destruction of the transformer also means costly repairs, loss of income while current is off, and what is of more importance, great annoyance to customers. A liberal margin of overload is necessary, and if specifications call for a rise of temperature not exceeding 40° C, at full load, any ordinary overload will do no harm, provided the insulation is safe. The rules of the Committee on Standardization of the A. I. E. E. state the proper voltages to be used in testing transformers for insulation, and the values so stated will be found in the part of this chapter devoted to tests of transformers. The writer has never been thoroughly satisfied with the methods in common use for deter- mining the rise of temperature in transformers or dynamos or similar appli- ances. The thermometer test is too superficial, and the resistance test is the average only, while what is wanted is the hottest temperature at any point, for "that is the danger point. It is probable that the ordinary small commercial sizes of transformers do not need such refinements, but the larger sizes would be much better tested with a special copper test coil placed at the danger point during construction, with leads brought outside for testing. This might not be necessary in more than one or two of the same type and size, but would never be out of place in every one of the larger sizes now coming so commonly into use in the modern power trans- mission plant. Insulation materials for transformers are of numerous kinds, and no two makers use identical combinations, although most use the same or similar materials ; following is a list of those in common use ; FEATURES OF DESIGN. and the reader is referred to the list of specific resistances (see index) for the breakdown point of most of them. Oiled linen, Oiled silk, Mica, Micanite, flakes of mica pasted together in different forms, Fiber, and all the other forms of artificial board. "Wires are nearly always double cotton covered. As for oils for the oil-insulated transformers, the "Westinghouse Company uses a clear thin oil much like signal-oil, and called lied Seal, while the Gen- eral Electric Company uses a special transformer oil, which is heavy, but is simply a good machine-oil freed from Avater. An order to the Standard Oil Company for transformer oil will bring an oil that will serve every ordinary purpose, and many times it will be found that unless some particular oil is specified they will seldom send the same twice. The laboratory of the National Board of Fire Underwriters has used a number of different kinds in its high-testing transformers (40,000 volts), and has never found any difference in results although ordered as stated above. Heating- and "Ventilation. One of the necessary requirements of any piece of machinery is that it must be able to operate for certain periods of time at its full load, and in some cases over-load, without undue heating. Fig. 3. G. E. Co. Type H Transformer — 20000 watts, oil-cooled. In a transformer, the capacity for work increases directly as the volume of material, densities and proportions remaining constant. The volume, however, increases as the cube of the dimensions, and the radiating surface as the square of the dimensions ; therefore, it is evident that the capacity for work increases faster than the radiating surface. Since the losses are also in proportion to the volume, the designer soon reaches a point where it is necessary to provide additional means for ventilation or radiation of heat, in order that the transformer may run under load without undue tempera- ture rise. Self-cooled transformers are those which require no artificial means for THE STATIC TRANSFORMER. 337 dissipating the heat energy lost in the apparatus during operation. These can be divided into two classes, the Ventilated or Natural Draft, and the oil-cooled. Fig. 4.— 500-k.w. Self-Cooling Transformer. W. E. & M. Co. Type, Oil-cooled. The Ventilated or Natural Draft transformer is one in which air is the direct means of absorbing the heat, it being designed so that cur- rents of air readily pass through the transformer. Such transformers are not well adapted for out-door installations, as they require a separate housing; otherwise there is a liability of water or moisture getting inside of tbe case. Oil-cooled transformers are those in which the coils and core are immersed in oil, the oil acting as a medium to conduct the heat from the coils to the surrounding tank. In addition to acting as a heat-conducting medium, the oil also serves to preserve the insulation from oxidation, increases the breakdown resistance of the insulation, and re-insulates the insulation in case of a puncture. The use of oil in a transformer results in a more rapid conduction between the transformer proper and its case or tank, and the lowering of the tem- perature increases the life of the transformer. Again, instances are known of the discharge of " atmospheric electricity," or a discharge of lightning at a distance that has punctured the insulation of a transformer, and when tilled with oil, the oil flows in and repairs the rupture, which may be too small to cause immediate damage. If a sufficient space is left inside the case, the oil will get up a circulation by its own convection currents, the cooler oil rising inside as it becomes more and more heated, the hot oil on the top falling as it is cooled by contact with the inside surface of the tank. This cooling may be further increased by making the containing case with deep vertical corrugations, thus largely increasing its radiating surface. The curves on page 1 8 serve to show the effect on the temperature of the use of oil. Curve 1 represents the temperature rise (by resistance method) of a transformer without oil ; curve 2, the temperature rise of the same transformer with oil ; curve 3, the temperature rise of the oil ; curve 4, the temperature rise of another transformer run without oil ; and curve 5, the highest temperature rise accessible to thermometer, whose actual tempera- ture (by resistance) is shown in curve 4. When the transformers are of such a size that sufficient radiating surface cannot be had in the tank to dissipate the heat, it becomes necessary to provide artificial means for cooling the same. Some of the means adopted are, water circulation, forced oil or air circulation. For both the water and oil circulation the coils and core are immersed in oil. The water-cooled transformer has its heatad oil cooled by means of water cirevdating pipes placed in the oil. The transformer thus has the ad- vantage of oil insulation, and the circulation of the cold water through the pipes requires much less power than the pumping of the oil, and in addition does not require external cooling apparatus. This method is subject to a slight danger, due to possible leak of water pipes. 338 FEATURES OF DESIGN. Transformers have been constructed in sizes up to about 2000 k.w., using water circulation for s r. - i- _L S^ - s ■" E n >* • - 3 SI / CURVES SHOWING DUE TO USE 0 Fig. 5. S TRANSFORMERS Am Air-Blast Transformer — or one in which ventilation and radi- ation of heat is, by means of a blast or current of air, forced through the transformer coils and core is shown in Fig. 8. In this transformer, the Figs. 6 and 7. Natural Draft Transformer — Showing Case Removed. coils are built up high and thin, and assembled with spaces between them, the air being forced through these spaces. The iron core is also built up of numerous openings through which the air is forced for cooling pur- poses. This style of transformer has been constructed in sizes tip to about 1000 k.w. THE STATIC TRANSFORMER. 339 The following tables show results of tests on a number of commercial transformers by Mr. A. H. Ford. Fig. 8. Air-Blast Transformer. Si Heating* Tests. Transformers in their cases. (Ford.) Rise Watts ra- diated per sq. in. of Case. w2. Watts ra- diated Rise Watts ra- diated per Watts ra- diated No. in Tempera- per sq. in. of Core No. in Tempera- sq. in. of Case. per sq. in. of Core ture °G. and Coils. ture °C. and Coils. w2. 1 31.4 .143 .175 9 310 .172 .300 24.3 .091 .107 39.4 .134 .234 57.4 .168 .198 2 20.1 .052 .110 12 31.6 .086 .145 15.2 .047 .098 20.5 .067 .113 47.8 .102 .214 51.8 .125 .211 30.8 .0S5 .190 21.5 .122 .206 3 20.8 .105 .121 13 60.0 .113 .131 17.5 .080 .093 49.4 .079 .104 50.2 .168 .195 38.4 .134 .155 5 21.8 .118 .166 14 43.4 .168 .266 19.1 .090 .127 32.1 .079 .130 40.8 .172 .242 101.8 .250 .396 40.6 .144 !203 76.9 .150 .234 6 62.4 .388 .542 15 25.4 .099 .150 52.3 .246 .346 21.2 .074 .112 86.8 .412 .580 67.5 .168 .255 72.2 .455 .640 51.6 .149 .225 7 20.0 .082 16 73.4 .225 .396 17.8 .058 66.1 .175 .242 56.3 .144 100.0 .340 .466 36.0 .100 70.0 .242 .334 840 THE STATIC TRANSFORMER. C Heating- Tests. Transformers out of their ct (Ford.) Watts Watts Rise in radiated per Rise in radiated per No. Temperature CC. sq. in. of Exposed No. Temperature sq. in. of Exposed Surface. Surface. W. W. 1 27.9 .175 11 27.0 .274 21.2 .107 18.9 .208 51.0 .222 52.2 50.4 .372 .320 2 14.6 13.6 .110 .098 41.4 .240 12 19.7 .145 42.4 .220 12.3 55.9 .113 .229 3 20.3 12.4 33.2 .122 .093 .167 53.8 .195 30.8 .136 14 29.1 24.0 .266 .125 4 16.2 .160 96.7 .382 13.4 .110 77.0 .286 59.4 .240 51.4 .200 15 25.1 .150 6 50.0 .547 14.3 .112 24.4 .346 61.3 .270 72.0 .595 59.4 .250 58.9 .655 7 14.0 .082 16 44.3 .396 6.4 .058 31.4 .243 75.0 .185 64.3 .438 19.0 .121 42.9 .304 Efficiencies. The efficiency of a transformer is the ratio of the output watts to the input watts. Thus Efficiency : Output watts _ Input watts Output output -\- core loss -4- copper loss The core loss, which is made up of the hysteresis loss and eddy current loss, remains constant in a constant potential transformer at all loads, while the copper loss, or I2R loss, varies as the square of the current in the pri- mary and secondary. Methods for determining all the losses are fully described in the chapter on transformer testing. In a service where a transformer is generally worked at full load, while connected to the circuit, as in power work, the average or " all-day" effi- ciency will be about the same as its full-load efficiency. By " all-day" effi- ciency is meant the percentage which the energy used by the customer is of the total energy sent into the transformer during twenty-four hours. In lighting work the transformers are usually connected to the mains or are excited the full twenty-four hours per day, while the customer draws current from them during from three to five hours in the twenty-four. As- suming on an average five hours full load, the losses will be 5 hours 12R and FEATURES OF DESIGX. 341 24 hours core loss. The calculation of the " all-day " efficiency can, there- fore, be made by the following formula : . ■ _ . Full load X 5 All-day efficiency = = -. — — , — j^-jz _ , _, „ , = J J Core loss x 24 -4- 1 2E X 5 + Full load X 5 From this it is evident that while for power work or continuous full load, the relative amount of the core and copper losses will not affect the " all- day" efficiency seriously, yet in the design of transformers which are worked at full load only a short time, but are always kept excited, a large core loss means a very low " all-day " efficiency. The two tables on pages 112 and 113 show various efficiencies of a number of transformers, giving maximum efficiencies and "all-day" efficiencies. They also show the core loss of various commercial transformers as found by Mr. Ford. i him inn i -C TRANSFORMER IRON.AGEING TESTS. BY H. F. PARSHALL HYSTERESIS IN THE IRON AS RECEIVE HYSTERESIS TRANSFORMER AFTER SHORT PERIOD OF LIGHT WORK HYSTERESIS TRANSFORMER AFTER THREE YEARS OF HEAVY WORK / / / / / / / / / / / / / y / / s i 1 LINES PFR SQUARE CENTIMETER. Fig. 9. J tk./ V / / / / / ,' TY / MANUFACTURE EY A.H. FORD,AT UNIVERSI OF WISCONSIN. JAN. FEB. MAR. 1897. B- TEST ON WAGNER TRANSFORMER, FEB. MAR. APR. 1897. /. / / T^ L- E 342 THE STATIC TRANSFORMER. ll c3 O u o o Pm © II 634 520 730 565 615 400 465 656 657 590 672 384 656 182 630 430 595 585 620 700 640 1 n o«oot-ooanoisffl®oooiooo(oooo — . ■ : -i -'■.■'■, i .■ - ■ . < ". :-i~.- . ..: ,.. ' \ .-- t-t.OOO«C'JOl-fflt-0»t't-lO«lOfflW<0 q p * o 8 II Hrn-o-fOLOOccoan^iaoHOoooNoo oot'm :. ■ •;• i - : i r, : , : -:- - OOOr-c-JHr-'O'-iOi-Hi-iT-tOOO^HiMOi-iCN § II :oo9«o-f«cJ-ti-tiocONLOOi-OOtoocooo :-. : i -f I- >o l- i- i- 01 I" 1- 1 i ■• --OWC03 OOOOOi-tOOTHOOOOOOtMi-irHOOTH SI © CO II fflooo'-o-#oq«qoq«NcoiqLoqiMq(»Hq oilOeo6MOiOiodi005tOI>iCfflodHr<)Or!N CDGOCOGOOOCOOOOOt-OOt-COC-t^OOCOCiClOOCiCi II ^«H»L'5oq-tooqqioqi>«oO'Hi»iaoi» oo t^ d -* d d d d d d d d d -* d i- d oi d to d t-O0000000t-O000t~O0CC0000O000l>OiOS00CSCi 1! q^qqq-sijioNHN^fflqNioq^tDioiqN © -*-#-* -+' l~ -* -*' O -*' CO -V Ol' ri -+ ci 'i 6 * ffl l> oi o © c; © co © © © © © © © © © co © © © © © s II loqqqooiONNNffliotONqwoNqqHiq oi •* id -t< >* i-H d -+ —I d -t- d co ■* l.o oi d t^ d t> c^ Si 6^ c3 cNoqt>;C^e^oo^icu5cqG5uqcocqoo-#ocqqoq oi oi © d d -*' cc" i~ d -# d d d <£-v-+dciccdd> rHCON-*COriC01SCO'qcqt>ioioio*qqqqioq d ^ t-^ d id d oi d t-^ i-l t^ d t^ d oi d Tji © © © d CMOT E^ r-T -# ©~ •. aqoinj^ rHcqeor^iocot-oooso^HCNco^iccot-oocJsorH GENERAL ELECTRIC TYPE H TRANSFORMERS. 343 B bg 3-1 PS o fe ® » |5 ^ S • d "53 a O O lO O lO lO O O tfi lA N to S3 t O CM CM CO Th IQ 1 1 1 1 1 c/. «PU H H H H fc o 5 ^•d ,5 iyc3 © bo £ Q,u ri 5, ffl ,rl (fl CO CO C5 C5 rH lO t~ « iH T3 .2I~' >> ^ g « -i o O Ol Ci C5 CI O Ol C3 C3 Q 3 I? JO HfcS -d ID os» s c o,2 .2^ S. CO CD CO t- ^H to iq iq o n h iq to n oo CO ■+ id io CO '5 11° ° " ° " " Ci © OS © © O 0) *"- o © c o © is i> 6 is * T^ CO ^ Ci ci ci fi ci ri i-l O lO CD -# CO CO CO 2 ©CNQj eS a 1 OOg s |oi d oo « i; m h rt CO t- tH lO 5 cS 11111 O O O © O o o © © © CO ■* io to h O O O O O 8 8 8 8 § ^ o 344 THE STATIC TRANSFORMER. Magnetic fatigue or aging* of iron subjected to magnetic reversals is now well recognized, and precautions are taken to prevent it by all the better class of transformer manufacturers. Unless great care is taken in this respect the core loss is liable to increase very considerably after time has elapsed, this loss increasing from 25 % to often more than 100 % of the ori- ginal core loss. The following curves show the difference between carefully selected and prepared iron, and ordinary commercial iron. The upper curve shows a very great increase in iron loss after 80 days' run, while the two lower curves show but little increase after the same length of time. Curves lu and 11 also show results of aging tests by Mr. W. F. Parshall and Mr. A. H. Ford. AS E /■ • ~l -10 ' E / 0 " r or'e|lqss!d AGEING / 1 p 4 u>v|s r BTA1 / ON A tlV / fa'ctured by / TRAN sf,ormer|buil / | i: i [j | ^ ,kVe 30 ** TYPF ^ WNSFORM ^R t | | | erJc :v't ON E . C C S fP ht | Nl R I E 2% C SI N ov 1 96 | CURVES SHOWING Regulation. The most important factor in the life of incandescent lamps is a steady voltage, and a system of distribution in which the regulation of pressure is not maintained to within 2 % is liable to considerable reduction in the life and candle power of its lamps. For this reason it is highly important that the regulation, i.e., the change of voltage due wholly to change of load on the secondary of a transformer, be maintained within as close limits as possible. In the design of a transformer, good regulation and low-core loss are in direct opposition to one another when both are desired in the highest de- gree. For instance, assuming the densities will not be changed in the iron or in the copper, if we cut the section of the core down one-half, we decrease the core loss one-half. The turns of wire, however, are doubled, and the reactance of the coils quadrupled, because the resistance changes with the square of the turns in series. A well-designed transformer, however, should give good results, both as regards core loss and regulation, the relative values depending upon the class of work it is to do, and the size of the transformer. The following table shows the results of tests for regulation of a number of commercial transformers obtained in the open market by Mr. Ford. REGULATION OF TRANSFORMERS. 845 Q 2 Q D £ 0 S rt » ■p< * fc t ** -d es O) N *"" a; & =J e 13 IP"! CO § cq M H i ^ h « ft * >» S 33 5 =5 ft o fl fcfi c3 1 Hi s II ^5 I> M CO rH r-J t* OS .in •-# ■Min|t-HO«HC)M in " r-5 rH " .H . ' .HHrtn" s II OHOCSOHO .^t>ON-HH»Q0NN00t-H 00 ' 'iHrJrici . CN "rHrHC^C^ci^ -d o o CO d II OOWOMtCNOOKOHtOOOUO -* rr o oo os -r — 1 1 1 - i - o - .-: -r O 0O II rroooo-fj-Hinooocs-fjoeoorHO " ' °°. * " T3 SO II Oioi C3 oi>*t>i>05qioaoo5ioioooH/o5 * * «q co to o6^eit^o6o2eo^t^o4i>c4oot>ooio ' " c4 o4 r-i II (N «o os os co ■* co in os cs ic-irH § ii qHMio®Mqqqt>oqoq»ff]WHqq^qi> ci co c co t-h H 3 ttOOOOO ooooooooooooooo i- ~ .-. - •; : r :---■:.■ o © OO H/ONi* i-OCO d 6 HNn^mffl^ooaOrtNco^iatot-coaOH 346 THE STATIC TRANSFORMER. Comparative Expense of Operating- JLarg-e and Small Iransforme rs. It is obvious that the design of the distributing system has quite as much to do with the maintenance of a steady voltage as does the regulation of the transformers, and the proper selection of the size of transformers to be used requires skilled judgment. When transformers were first used it was the custom to supply one for each house, and sometimes two or three where the load was heavy. Expe- rience and tests soon made it evident that the installation of one large transformer in place of several small ones was very much more economical in first cost, running expenses (cost of power to supply loss), and regulation. Where transformers are supplied one for each house, it is necessary to provide a capacity for 80 % of the lamps wired, and allowing an overload of 25 % at times, where one large transformer is installed for a group of houses, capacity for only 50 % of the total wired lamps need be provided. For resi- dence lighting, where the load factor is always very low, it is often best to run a line of secondaries over the region to be served, and connect a few large transformers to them in multiple. A study of the following curves will show in a measure the results to be expected by careful selection and placing of the transformers. The first curve, Fig. 13, shows the relative cost per lamp or unit of transformers of different capacity, showing how much cheaper large ones are than small s. fe2 o I > 1 H 3 V -~. c 1 0 2 0 3 .1 0 IGH 4 TS 0 5 0 6 0 Fig. 12. Relative Cost of Transformers of Different Capacities. The second set of curves, Fig. 14, shows the power saved at different loads, and using different sizes of transformers. 0 100 200 300 400 Fig. 13. Relative Efficiency of Large and Small Transformers. COMMERCIAL TRANSFORMERS. 347 Power JFactor is the ratio of the actual watts in a line to the volt amperes or apparent Avatts in that line. It is also defined as the cosine of the angle of phase displacement of the current from the voltage in the circuit. The power factor of most commercial transformers is low at no load, varying from 50 % to 70 %, while at high loads the power factor is very nearly 100 per cent. For this reason it is better to distribute the trans- formers on the line so that they will carry load enough most of the time to keep the power factor reasonably high. COIHMERCIA£ TRANSFORMERS. The following tables show the trade numbers, capacities, and the ordinary characteristics of some of the transformers in more common use at this time, including Stanley Electric Co. : Westinghouse Electric and Manufac- turing Co. ; " Wood," the Fort Wayne Electric Corporation ; Wagner Electric and Manufacturing Co. ; General Electric Co., table for which will be found on page 1 13 In order to show a comparison of the qualities of transformers as made some time ago and at present, a table of tests by Dr. Fleming, F.R.S., is also included. IVAVLEY ELECTRIC MAlfrFACTURIIIfG CO. LIGUXiaTG TRANSFORMERS. Frequency == 66 P.P.S. Efficiencies. Regulation uniformly 2\ % at full load. Type. Full Load Output in K.W. Full Load. | Load. \ Load. \ Load. | Load. 2G \ 93.0% 93.1% 92.2% 88.8% 80.7% 3G I 93.0 93.2 93.0 89.5 82.5 4G 1 95.5 95.7 95.0 92.0 85.0 6G 1| 95.8 96.0 95.5 92.8 87.6 8G 2 95.9 95.9 95.5 93.5 88.5 10 G n 96.0 96.2 95.8 93.5 90.4 15 G 3| 96.6 96.7 96.3 94.3 91.3 20 G 5 96.7 96.9 96.6 95.0 91.5 30 G *2 96.8 97.0 96.7 95.5 92.2 40 G 10 96.8 96.9 96.8 95.7 92.6 60 G 15 97.2 97.2 97.2 96.9 94.8 80 G 20 97.8 97.7 97.5 96.9 95.1 100 G 25 97.6 97.8 97.8 97.2 95.5 348 THE STATIC TRANSFORMER. ..fifi P ,p.e. H — FUL 7b L LOAD 37.91 1 3/4 1A ( ' 97.86 I 97. 49 95 s 1A 1/a ' , 95.89 92.4-7 $ V, REGULATION 1'/2 i / TYPE 400-W. POWE^R TRANSFORMER / ST ANLEY ELECTRIC M'F'G CO. | i OUT PUT N KIL 3WAT rs i FFFIC IENCN AT 33 P. P.S. FUL •7/B L LC )AD 93.1 93.15 * 3/4 V? 93.14 97.91 > v4 1/8 ; 96.69 93.96 RE GULA TION 1 Vz k "6^. P TV OWE PE 400-Vy. R TRANSFORM FR ST ANLE »■ ELE CTRIC M'F' SCO. OUT PUT N KIL DWAT TS 25 50. 75. 10 STANDARD C. S. TRANSFORMERS. 349 ST4SDABI> C. S. IRAASf OltTIEIt^ OE WE§TIXC}< HOUSE ELECTRIC .U'U MAJVVEACTURIVG CO. Iron Eosses. True. Apparent. Watts. ir= i33£ ^=60 JST= 133± J^=60 1 250 6.80% 9.40% 8.90% 13.00% 2 500 5.20 6.80 6.60 9.70 4 1000 3.00 4.10 3.70 5.60 6 1500 2.50 3.30 3.20 4.70 8 2000 2.20 2.90 2.80 4.10 12 3000 1.70 2.20 2.20 3.10 16 4000 1.70 2.20 2.20 3.10 20 5000 1.60 2.10 2.10 2.85 25 6250 1.57 2.05 2.02 2.84 30 7500 1.54 2.00 1.90 2.70 40 10000 1.30 1.70 1.71 2.31 50 12500 1.06 1.40 1.40 1.85 60 15000 1.02 1.32 1.35 1.80 75 18750 0.92 1.20 1.17 1.61 100 25000 0.86 1.12 1.12 1.53 STAWAIID C. S. THAHTSEORIflEItS OP WEiTISG- HOUSE EIEC1RIG -AJ¥B» MAMFACTrRIJITG CO. Efficiencies. Full Load. | Load. \ Load. \ Load. JV=133| JY=60 ^Tz=133i ^"=60 iV=133| N=m A=133J jV=60 25 15 23 32 38 46 53 60 1 90.3% 87.7% 88.8% 85.3% 84.7% 79.8% 71.6% 62.0% 2 91.7 90.1 90.7 88.7 88.0 84.9 78.4 72.0 4 94.0 93.0 93.8 92.3 92.5 90.3 97.3 83.0 6 94.5 93.6 94.3 93.3 93.4 91.8 89.2 86.0 8 95.1 94.4 95.0 94.1 94.3 92.8 90.5 88.8 12 95.8 95.2 95.8 95.1 95.4 94.3 92.6 90.5 16 96.34 95.8 96.3 95.5 95.7 94.6 92.8 90.7 20 96.5 96.0 96.34 95.8 95.85 96.8 93.1 91.1 25 97.0 96.54 96.83 96.23 96.15 95.23 93.36 91.52 30 96.96 96.50 96.72 96.21 96.17 95.25 93.47 91.63 40 97.04 96.64 97.02 96.49 96.56 95.76 94.35 92.75 50 97.24 96.90 97.31 96.86 97.03 96.35 95.34 93.98 60 97.38 97.08 97.44 97.04 97.16 96.56 95.52 94.32 75 97.48 97.20 97.58 97.20 97.36 96.80 95.92 94.80 100 97.74 97.48 97.81 97.45 97.58 97.06 96.21 95.17 850 THE STATIC TRANSFORMER. eq ' 8 o CO OS CO o OS 01 ON o »o Si eo lJ- t- rH ?1 us LO os (N 1 8 M ^ OS os CO os 8 - to ,_, > s rt a5 ^gP4 o o C o o * ft 3S £ -55 g o ■j CD 11 d © bo o ood r/J OJ ft o an .3 c3 >> PI ft£ bo 3 £ cS An 80 -d ft ■ pi o be] £ WAGNER TRANSFORMERS. 351 55 » a * & s 1 OS ft OS s ol m s eS q m s s © c4 CO OS OS < • t- o t- q © q A 8 lO s S3 oS io o3 co -H O co 8 J £ r"' oo Tf a OS o bo 3 o s o oo o 8 8 8 I 352 THE STATIC TRANSFORMER. e3 ft co eo oo 3 _ 00 t- CO CO CO to 8 oo ? © $ I- od os CO 00 oo oo C0 OS CO CO - os CO CO 1 £ §8 8 © CO OS d o i-q in N to 3 8 s CO © p © CO CO .2 rj5 od d c4 CO <# in id oo OS co CO CO CO CO CO CO a rtW 'o 50 H d e3 S 00 « LO 8 00 CO 3 CO £ in "3 3 6 -+i to o d d OS CO CO CO CO CO os CO os 6 H|S „ 03 CO in o lO ^ § »o CO 00 § « CO ® 3 -IH 1 LO d d d d t^ os CO CO os CO CO CO CO os os d c3 s * © in 3 CO "# in 3 00 g to in d l^ OS CO CO CO CO OS 1-1 d c3 O eo CO 00 CO $£ 3 -* s CO s 00 CO ,_} CO -l< in d d t^ HW CO OS OS OS CO OS OS CO OS OS Per Cent of eg illa- tion. to CO © co ig o in o o in o o tf u & "3 a" OS c4 O m 3 OS 00 oq s CO CO p-So« lO in 1 8 Pho£^ Tji co to « 00 cq 2 *j *2 o ^ o 0 § o o o o o s o to o o &, c3 to 2 m 1 5 1 CO fr- s o e§ a o3 ac o a 8 § 3 o 1 | o m o 8 3 eo WAGNER TRANSFORMERS. 353 8 * S 5 e » & .5 h H * ! g 0 >> o3 ft 3 rh Ol CO £ o q q *# C5 05 ^ d o CO in to o 3 CO s o 5 s s CO 00 $ CO c§ cd © -d -* "* to CO CI 3 q q © Tt< id 05 ©5 .® 1 w •d 03 s i? 3 iq gj g2 CO C5 id id OS OS 05 C5 C5 o H^ 5 s O « ^ Per Cent Coppe Loss S lO lO 8 8 q 8 o oq ci oi c4 Ol T~' 7-1 1-1 i? S 2 © CO o oi 8 % 3 3 CO OS «s 1 § o 1 § o 8 8 8 s o 8 £> * 1-1 oq CO "* CO '5 ft 6 3 o a o o 8 8 8 I 354 THE STATIC TRANSFORMER. >-> <# 50 os CO t- o S CI A iq o iq co CO t> OS © 3 ■* id 00 OS © as as © © © <1 T3 ce 0 os -# co "t ci 3 g£ -V 3 »d ed ed t> [> OS © © © © © © © © a ® o c3 O h! id oi eo ed l> © OS © © © as as © © © c3 O Ttl o CO © ^ OS CO © s © iq S CO T* id © © OS OS as © © © © © © 1-1 c3 8 05 OS CO 00 CO CO S8 s s 9 © <* 3 si? $ >r; id ed © t^ OS as as © © © © as as © ^ o i- o ia 10 s © 3 3 s © °# J eo eo co CN ci ci > 2416 2400 2435 2447 2389 2400 2400 2400 2400 2400 2400 2400 2400 2400 2392 2400 2400 Magnetiz- ing current in am- peres. t- 1CCOC-1CO CO (O rt (i) O rHCOOICOlOrtOoSi-lt-clboOOT-l Maximum output in watts from secondary. 1875 3750 7500 11250 15000 3750 7500 11250 15000 11250 3000 6000 6500 6000 750 4500 4000 B p a 03 ainti (1885) type . (1885 rewound) (1892 type) . (1892 rewound) burne Hedgehog . inghouse . . . .ey-Brush . . . ison-Houston . . D 356 THE STATIC TRANSFORMER. SPECIAL TYPJES OJP TRASSFOMMER. The ordinary static transformer is generally understood to be a constant potential transformer, which is adapted to operate when connected in parallel across a constant potential circuit. When transformers are designed for special uses, it is customary to designate them by name, indicative of the special work they are intended to perform. A few of these transformers are here described. Special High JPotential Transformer. In making high potential tests of apparatus, it is very desirable to have a transformer which is adapted to tbis work. The General Electric Company is now supplying a transformer designed for the purpose of making high potential tests up to 10000 volts. This trans- former is tested up to a pressure of 35000 volts, and is so constructed as to avoid any danger of breaking down as far as possible. Below is a cut, to- gether with a diagram of its connections. SECONDARY 5T~ .. VOLT MAINS' g (OR 52 VOLTS WITH ~ connection parallel) WATER RHEOSTAT The core is rectangular in form, the primary or low-tension side being wound on one leg of the core, while the secondary or high-tension side is divided into four separate coils, and mounted on a sleeve of heavy insulating material, and placed over the opposite leg, the Avhole being immersed in oil. In making high potential tests of apparatus, it is very desirable to have a transformer which is adapted to the work. A micrometer spark gap is mounted on top of the box or case, and con- nected in shunt across the high potential terminals. The spark gap is set for the desired voltage by the use of a calibration curve, or by a preliminary calibration by means of a voltmeter connected to the low-tension side, the ratio of transformation being known. The apparatus to be tested is then connected to the high potential terminals, and the potential raised to the desired amount. SPECIAL TYPES OF TRANSFORMERS. 357 This transformer is most invaluable in testing all kinds of apparatus for high-tension work. Fig. 17. High Potential Testing Transformer. Transformers for Constant; Secondary Current. Several methods have been tried with more or less success to obtain con- stant current at the secondaries of transformers. The simplest and earliest system for obtaining a constant current in the secondary is by means of transformers whose primaries are connected in series, and a constant current maintained in the primary. This is shown in diagram in Fig. IS. Series transformers for this purpose have never been very successful, due to the trouble caused by the rise of potential in the secondary when opened for any cause. Various devices (Fig. 18), such as short-circuiting points separated by a paraffined paper, or a reactive or choking coil connected across the secondary terminals, have been intro- duced to prevent any complete opening of the secondary by reason of any defect in the lamp or other device connected in the circuit. CONSTANT CURRENT LINE SERIES TRANSFORMERS SHORT- CIRCUIT .POINTS ■ARC LAMPS-* Fm. 18. Reactive coils used as shunt devices have been used under different names ; as compensators, choking coils, and economy coils. 358 THE STATIC TRANSFORMER. A device of this kind has been introduced by the Westinghouse Electric and Mfg. Company, and others, for use in street-lighting by series incan- descent lamps. It is shown diagrammatically in Fig. 19. The lamp is placed in shunt to the coil ; when the filament breaks, the total current passes through the coil, maintaining a slightly higher pressure between its terminals than when the lamp is burning. It is thus evident that the regu- XT] CX_d_ -\Zr lation of the circuit is limited, due to the excessive reactance of the coils when several lamps are taken out of circuit. Economy Coils or Compensators. A modification of the above is built by several companies for use on ordi- nary low potential circuits, where it is desired to run two or three arc lamps. It is a single coil transformer, and is shown in Fig. 20, and diagram- matically in Fig. 21, same page. If any lamp is cut out or open-circuited the current in the main line decreases slightly. As more lamps are cut out SECONDARY J CIRCUIT 1U v. J D. P. SWITCH ] D. P. FUSE BOX _J»i K 1 1 A n i 14 AMPS s *±$X- <- Fig. 20. Westinghouse Econ- omy Coil. For A.C. arc lamps- Fig. 21. Arrangement of Apparatus for use of Economy Coil or Compensator. the remaining lamps receive less current, and it is necessary to replace the bad lamps in order to obtain normal current through the circuit. Transformers for Constant Current from Constant Po- tential. The transformers represented in Fig. 22 show a design that will give out an approximately constant current when connected to constant potential circuits. The transformer has its core so designed that there is a leakage path for the flux between the primary and secondary. This is shown in the SPECIAL TYPES OF TRANSFORMERS. 359 diagram at a and b. At open secondary circuit there is little or no ten- dency for the flux to leak across the gap. When current flows through the secondary, thus creating a counter magneto-motive force, there is then a -■SECONDARY Fig. 22. Constant Current or Series Transformer. leakage across this path, and if properly proportioned, this leakage will, act to regulate the current in the secondary, so that it will be approximately constant. General Electric Constant Current Transformers. The transformer just described has the disadvantage that its regulation is fixed for any transformer, and may vary in transformers of the same design, without any ready means of adjustment. The transformer also regulates for constant current over but a limited range in the secondary loads. The General Electric Company constant current transformer shown in Figs. 23 and 24, is constructed with movable secondary coils, and fixed pri- mary coils. FiG. 23. General Elec. Co. Constant Current Trans- formers for 50 lights. FiG. 24. Connections for Alter- nating Series Enclosed Arc Lighting System, with 50, 75, or 100 Light Transformer. The weight of the movable coil is partially counterbalanced, so that at normal full-load current the movable coil or coils lie in contact (See Fig. 23) with the stationary coil, notwithstanding the magnetic repulsion between them. When, however, one or more lamps are out of the circuit, the in- creasing current increases the repulsion between the coils, and separates them, reducing the current to normal. (See Fig. 24-) At minimum load, the distance between the coils is maximum. The regulation is thus entirely automatic, and is found to maintain practically constant current, or a de- parture from constant current if desired. The transformer can be adjusted for practically constant current for positive regulation ; i.e., increasing current from full load to light loads, or for a negative regulation, i.e., de- creasing current, from full load to light loads. This adjustment is obtained 360 THE STATIC TRANSFORMER. by changing the position of a cam from which the counter-weights are sus- pended. The curves shown in Fig. 27 show the range obtained in a 100-light transformer. Fig. 25. Full-Load Position of Secondary Coils. Fig.. 26. Half-Load Position of Secondary Coils. The transformers are enclosed in cast iron or sheet iron tanks filled with transil oil. The oil, in addition to being an insulating and cooling medium, serves to dampen any sudden movement of the secondary coils. These transformers are connected to the regular constant potential mains, and the larger sizes are arranged for multiple circuits in the secon- dary. After having been started on a run, the transformers need no atten- tion, as they are entirely automatic in their action. Fig. 27. Diagram of Connections. The full-load efficiency of this type is practically the same as that of a constant potential transformer of the same capacity. The power factor of the system at full load is about 85 per cent, due to the reactance of alternat- ing arc lamps. At fractional loads, the power factors necessarily are much lower, and it is therefore not desirable to operate such a system at light load. RE 3ULATION 1 1 1 TEST 100 L. 1. G. ?.00 DISTANT TIVE BEG iUr ION, CI RRE T T ANS "ORIV ER. to' 6. 6 noN vea &GVJL *-n° *- S N' gM" LOAD FULL LOAD Fig. 28. REGULATING RESISTANCE COIL. 361 REGVLATiarG REACTANCE COIL FOBS A. C. ARC CMCU1TS. Another and very simple device for regulating the current in a series cir- cuit for A.C. arc lamps has been put on the market by the Manhattan Gen- eral Construction Company. It consists of a single coil of insulated wire arranged to enclose more or less of one leg of a "W "-shaped magnet, as shown in the following cut. The coil is suspended from one end of a lever Fig. 29. Regulating Reactance Coil by Manhattan General Construction Co. and counterbalanced by a weight on the other, and so arranged that at all points of its travel it just balances the varying magnetic pull of the coil. The arc circuit is connected in series with "this coil with a switch to open the circuit. Without current flowing, the normal position of the coil is at the top or off the leg of the magnet. When the switch is closed, current flows in the circuit (and coil), and draws the coil down on the leg to a point where the reactance of the coil holds the current strength at a predeter- Fig. 30. Diagram of Connections of the Regulating Reactance Coil of the Manhattan General Construction Co. 362 THE STATIC TRANSFORMER. mined point ; as, say, 6.6 amperes. It is said that this device will maintain a current constant within one-tenth of an ampere. The losses are the iron losses and I2R losses in the coil, which, with con- stant current, are the same under all conditions of load. As it is not always, or even often, that it is necessary to provide for regu- lation of an arc circuit to the extent of its full load, the makers have adopted the policy of supplying instruments to care for but that part of the load that is expected to vary, in some cases 10 % of the circuit and in others 75 %, thus avoiding the need for larger apparatus, or for insulation for the total voltage of the circuits. They claim another advantage in being able to connect the device in one leg of the series circuit, and allowing the other end of the circuit to be connected to the mains at any such point as may be the nearest at hand. Fig. 30 shows the apparatus diagrammatically. feeder Regulators. An alternating current feeder regulator is essentially a transformer hav- . ing its primary connected across the mains, and its secondary in series with the mains. The secondary is arranged so that the voltage at its terminals can be varied over any particular range. Fig. 32. Internal Connections of a Stillwell Regulator. MMH^^HMi REGULATING RESISTANCE COIL. 363 The several different styles of feeder regulators have been devised, differ- ing in principle of operation, but all of them have the primary coil con- nected across the mains, and the secondary coils in series with the mains. The " Stillwell " regulator, which was designed by Mr. L. 13. Stillwell, has the usual primary and secondary coils, and effects the regulation of the cir- cuit by inserting more or less of the secondary coil in series with the line. This secondary coil has several taps brought out to a commutating switch, as shown in Fig. 31. The apparatus is arranged so that the primary can be reversed, and therefore be used to reduce as well as to raise the voltage of the line. It is evident from an observation of the diagram that if two of the segments connected to parts of the coils were to be short-circuited, it would be almost certain to cause a burn-out. To prevent this, the movable arm or switch-blade is split, and the two parts connected by a reactance, Fig. 33. this reactance preventing any abnormal local flow of current during the time that the two parts of the switch-blade are connected to adjacent seg- ments. The width of each half of the switch-arm must of necessity be less than that of the space or division between the contacts or segment's. As the whole current of the feeder flows through the secondary of the booster, the style of regulator which effects regulation by commutating the secondary cannot well be designed for very heavy currents because of the destructive arcs which will be formed at the switch-blades. To overcome this difficulty, Mr. Kapp has designed the modification which is shown in Fig. 32 p. 362. In this regulator the primary is so designed that sections of it can be commutated, thus avoiding an excessive current at the switch. This regulator, however, has a limited range, as the secondary always has an E.M.F. induced in it while the primary is excited ; and care must be taken to see that there are sufficient turns between the line and the first contact in order to avoid excessive magnetizing current on short circuit. 'CONTFtflU.iNG HAND Fig. 34. Connections for M. R. Feeder Regulator of G. E. Co. Fig. 35. Diagram of Con- nections of Feeder Po- tential Regulator. The General Electric Company have brought out a feeder regulator, in which there are no moving contacts in either the primary or secondary, and Avhich can be adapted for very heavy currents. This appliance is plainly shown in Figs. 33 and 34. The two coils, primary and secondary, are set at right angles in an annular body of laminated iron, and the central lami- 364 THE STATIC TRANSFORMER. nated core is arranged so as to be rotated by means of a worm wheel and shaft as shown. The change in the secondary voltage, while boosting or lowering the line voltage, is continuous, as is also the change from boosting or lowering, or rice versa. In this regulator, the change of the secondary voltage is effect- ed by the change in llux through the secondary coil, as the position of the movable core is changed by the turning of the hand wheel and shaft. There are, therefore, no interruptions to the flow of current through either the primary or secondary coils, and the regulator is admirably adapted for in- candescent lighting service, where interruptions in the flow of current, how- ever instantaneous, are objectionable. S. St. C. UEVICES fOR RE&riAII^C} A. C. CIRCUIT!. Where polyphase A. C. generators are used for lighting and power it is necessary to provide some method by which the individual phases can be separately and independently regulated. The method used by this company for accomplishing this result is by changing the effective turns on the armature. At one end of the winding of each phase are several regulating coils from which are brought out to suitable regulator heads taps which are mounted upon a terminal board fastened to the machine ; or the regulator heads, if so desired, may be mounted upon the switch-board. The following diagrams illustrate he method of bringing out the regulating taps from the armature coils of a two-phase generator. Fig. 36. Two-phase Generator. The regulator heads are similar to those used in connection with the " Stillwell" regulator, and make use of a modification of the split finger contact arm and choke-coil to prevent short circuit of the regulator coils. DEVICES FOR REGULATING A. C. CIRCUITS. 865 PHASE A-B vVWVWvWAVAVAW-WWV PHASE E-F ^T f— ? L t I [ K { J I \ [\ oo © 1000 Cubic Feet. p u ■ g£ft 53 * -^ Ills St * c 1 0 c3 8% OQOniW ffiOHrt Sfl CCOi-^W oqoi-^S SM $1.00 $0,005 $0.42 $0.10 $0.0056 $0.47 $0.12 1.20 .006 .50 .12 .0066 .55 .14 1.40 .007 .58 .14 .0076 .63 .16 1.50 .0075 .63 .15 .0081 .68 .17 1.60 .008 .67 .16 .0086 .72 .18 1.80 .009 .75 .18 .0096 .80 .20 2.00 .01 .83 .20 .0106 .88 .22 2.20 .011 .92 .22 .0116 .97 .24 2.40 .012 1.00 .24 .0126 1.05 .26 2.50 .0125 1.04 .25 .0131 1.09 .27 2.60 .013 1.08 .26 .0136 1.13 .28 2.80 .014 1.17 .28 .0146 1.22 .30 3.00 .015 1.25 .30 .0156 1.27 .32 3.20 .016 1.34 .32 .0166 1.30 .34 3-40 .017 1.42 .34 .0176 1.39 .36 3.50 .0175 1.46 .35 .0181 1.47 .37 3.60 .018 1.50 .36 .0186 1.55 .38 3.80 .019 1.58 .38 .0196 1.63 .40 4.00 .02 1.67 .40 .0206 1.72 .42 4.50 .0225 1.88 .45 .0231 1.93 .47 5.00 .025 2.08 .50 .0256 ,14 ■M 414 ELECTRIC LIGHTING. Cost of Producing- Electric Eight. No very general investigation has yet been made on this subject in the United States, and few outside the Edison Companies have good facilities for determining the cost. Buckley gives the following : " The profits on electric lighting depend primarily on the average number of hours the lamps burn. Under usual conditions (supplying incandescent current through meter including lamp renewals) the cost per lamp per hour averages as follows : Average Cost of Arc and Incandescent Eamps per Hour. Cost 16 Candle- Cost 2000 Can- Cost 1200 Can- Length Time Burning. Power Lamp, dle-power Arc, dle-power Arc, per Hour. per Hour. per Hour, \ Hour each day .... $.02 $0.16 $0.14 1 Hour each day . . .0112 •08J .07} 2 Hours each day . .0062 .05 .04} 3 Hours each day . .0046 .04 •03| 4 Houi-s each day . .0037 .03J .03 5 Hours each day . .0032 .03 .02* 6 Hours each day . .0028 .02f .02} 7 Hours each day . .0026 m\ .024 8 Hours each day . .0025 .02} .021 .02 9 Hours each day . .0024 .013 .01f 10 Hours each day . .0022 .02 UTotes : — An incandescent lamp gives off from \ to ^ the heat of an equivalent gas-jet. An arc lamp gives off from £$ to ^o as much heat as gas-jets producing an equal light. A 5-foot (16 c.p.) gas-jet vitiates as much air as four men. IIGHTL\(; ICHEDl'LES. General Rule for Construction Schedules. Moonlight Schedules. — Start lamps one half hour after sunset until fourth night of new moon ; start lamps one hour before moonset. Extinguish lamps one hour before sunrise, or one hour after moonrise. No light the night before, the night of, and the night after full moon. During summer months there will be found nights near that of full moon when, under the rule, the time of lighting would be very short. It may not be positively necessary to light up during such times. If better service be desired, but not full every night and all-night service; lamps can be started at sunset and run to 12 or 1 o'clock on full-time sched- ule, and after 12 or 1 on the moonlight basis. The above rules by Alex. C. Humphreys, M.E., have been modified by Frund as follows : Light every night from dusk to 12 o'clock ; after 12 o'clock follow Humphrey's rule for moonlight schedule, excepting there will be no light after 12 o'clock during the three nights immediately pre- ceding full moon. All-Might. Ever j- Wight Schedule. — Start lamps one half hour after sunset, and extinguish them one half hour before sunrise every day in the year. Full schedule commonly called 4000 hours for the year. All the above rules serve to make schedules for any locality, and such schedules must be based on sun time for the locality, and not on standard time. Permanent average schedules are used in New York City, but for other cities they are usually made up fresh every year. Following will be found New York City time tables, also another set by Humphreys that is a good average for sun time in any locality. LIGHTING TABLE. 415 •Sat ■nana etuii, •qsmS utQ.xa jSoooqoSSSoioiomiooiqoiqioioiiJiqiq'Oioiqiniq «r^r/(»Q0Q600l>l>l>l>t^t>t>t^t-t-t>t-t>L-»t-l>l>l>t-t-t- essggss? JSSS^feS^^iSEci'-^^i^ii^ccici •Sai 91UJX •qsin§ ■uiq!xa I "TO§n •gai -u.mgi B.S2-2S2SSS52S22222SSS^««!oo«xodo6«6»od«6 ^§SS^i2SSS3?JSS228qqqq81qqqqo £ cn i o i '^^^^^^S^S^^^l^^^^'-'^'-' si— 'i^22^;32^§§§§oS51?oS g S"#eOCO ^^^^^^-^^^^^^^^cO^CO^COOO^O^COCOCNCNCNOa •qsm§ ^cddcdcddcdyscdo^oyscoo *q • X Ld id id id id id id id 10 id is «5 «o >o « io io « io io »o w »o o » rri«5t!>5Se£:2S§5)§^q?^5; id id id id id ia ia JOW8iK I -x^* — * _ 416 ELECTRIC LIGHTING. O •Sat h.m. 13.32 13.33 13.34 13.36 13.37 13.38 13.39 13.40 13.41 13.42 13.43 13.44 13.45 13.45 13.45 13.45 13.46 13.46 13.46 13.46 13.46 13.46 13.45 13.45 13.45 13.45 13.44 13.44 13.44 13.43 13.42 0_1 •qsmg -niqixg; r:»t-ccoO'-iNco'#Lotot>ooccoQOOHHOi(NMwnn«m'*T)i'* £ co :o co co -r -f -p -p -r -r -r -r -P ~v -r -p in iq in lO lO in in lO lo in LO in in in in •;q§JT[ ^"^■f-t00(»C)0>Oi-i!N - q q q q q q q q q q q q q q q q q q q ioccocccohhth -' in i in m- m' in in in in in in in in in in m" in in m' m' m' in in in in in in in in in" in in « o 525 "§ur -nmg SMMW'ttttCLOOLOCOOOOrirlHTHrlflCjC-lCKjqC'lWW . ^ oi oi c-5 ci oi oi ci oi oi oi ci co co co co' co" co' co' co' co' co' co" co co co co co co co co . LO ■qstnS HT-INMlO(OXmOffim*«Ot-00»OH«mr|(iaQOO)OrtNNM^lO • "g •^§71 rN©LO'*OHOfflCOI>OLO'*M(NHOOOOOXNtO'fflmL0 1ClflTtl^ • ;:i:icici:i:i:iHHHrHHHnHHHr-qqqoqqcocooq . jinmmmininominmminminmin'ininin'minmmininmininmm . e i CO o H O o ■Sui 8UIl£ ^•ft^onaooHMcoffiHijKooHTfMonuot-ow'tt'CC'i'ft-oci Cr -V -t ~H m LO m = — O O O i-i — i-< i-H C4 05 OI CO CO ^rtrtrtHrtHHn'HrirtHrtHHrtM Cl" OI (N CI 01 Cl' CI CN CN iH oi oi c4 oi 5 co g •qsin§ -ui^xg; r.'NWOwc-in-fiLowt-ooaoHiMcoiocoNMoO'Hcico-fincNffiO - oi oi cq co co co o; co ct co co cq -r -r -r; -p -r -h -r_ -t -h in lo in in in in in in in q ^laiQ^inVn^ia^iQ^io^iQ^in^inoiniQ^iA^iA^iQiQiQiaiQco s h wan ^'MHOXt-CMClOCONLl-lfNHQNO-jiniMwr x VC LO CO M O OJ CO SH«p-qq q q q q ,q iq m_ in in ^■ftltltt ". ::. ". :". ". :". c/5 ~: ^2 j®«»66»«6»iaidioioioiQiaidioioioioidiooio«ioioioioio I ft 1 * H H •Sui -u.mg HffiffioddoddododddddccodcddocHHHTHH • OC' •qsinS -ui^xa; r:MC50-HMM-fiO(0|>MO!OHCin-|H-*LOtOt>a300rtlSMrJnn» • ;i-.i-.qooq;;c;:ocHrHrtHHrt«HrirHr;c-ici«iicioiC)N . H '%^11 r-'nHOi»0"*nHOiNL-t-O»O»WW«0«OOO«>OOOOO«5i»OM5C0C0O«D«)tt> . a 5 •SlTI -u.ing • Qtuix r.'OCl'f®CC)HCOOCCOCl-CNffiN-*ll»QC10CllONOni>C5H<*l»aCl ^c©!»c©cdcx3o6aJa3cocscic»cscsc»cscimc»Gsc> to CI ci •qsmS J cj i_- » o o w 01 n * 10 » i> » « o w ci m ^ a a n cc o o -H f i « -J 10 1- •*q§n: -gL'5'tiMrt005l»a>lO-liMHO!»NOiO'*«rtOCO«P3iM'Hg>OOglO - 1 "t t "*i 1 1 c/? c-. t:. "c cc. :". :". ::. ~\ C1. T.' -; °\ 7>. 'i c! *■: t "i 'i1-: R^0. ^ ^L^t^lc^t^l^t^t^tc^L^t^t>^»c^t^Ic^L^t~t^lc^tc^tc^lc^Jc^t^Ic^tc^tc^^ ^ •Sut -ixjiiai ^-t^[>^c^odco"ccco*c co o o ^ « co ^ ■+ 10 to - q q q q q q q qc ^ q - « ^ « ,— — i -h ^_ i-h r-< — < — i oi oi oi ci oi oi oi oi ^ •* ■* "*' -* -*' ■* Tj<" T)5 ■*" Tfi -*' ■*' -*' rj< -#" -#" -*H ■* "* ■* ■* •* •*" ^* Tt< ■* •* -^H •>#' ^ ■*" •(UlSiT ■"2J225Swnf'f'tl-<'Hcaeoai(»(>offlL0')|«cirto»!»«)i> - ~. — . °. °. ^ c. °. —. — . °. q — . q q '-~ >-^ '^ >~ Lq '-"? '-. i~. i-~ u" ,-~. '" l^ *t ■* ■* ■* ^"oooo"cdcooo'c/5cc'cococcc>Dcocdc/>t^L~t^t^t>t^i>^L^i>^t^ JO iqSjK rHO)CN(NCNCNCNCNl0qCOCO LIGHTING TABLE. 417 Summary of New York City JLig-hting- Xahle. Average. Average Day. January . February March . . April . . May . . June . . July . . August . September October . November December 413.10 355.27 341.29 290.17 264.39 238.51 256.12 286.26 316.48 368.50 392.59 424.52 13.19 12.15 11.01 9.40 8.32 7.57 8.16 9.14 10.33 11.54 13.05 13.42 18th 15th 16th 16th 15th 12th 17th 16th 15th 16th 14th 10th Shortest Longest . Average June 21 Dec. 21 Mar. 21 & Sept. 21 7.54 13.46 Note. — Lights started 30 minutes after sunset. Lights stopped 30 min- utes before sunrise. For commercial lighting : add 1 hour for part night lights, add 2 hours for all night lights to above schedule. Tahie Snowing- Hours of Lighting- throughout a Year of §«00 Uoui'N. Daily Lighting. 03 to o3 a, < 6 p is >-» bo P 53 & 5 53 CD ■ 89 117 145 173 201 257 313 92 64 36 67 98 129 160 191 253 315 69 38 36 66 96 126 156 216 276 32 2 6 37 68 99 130 192 254 3 21 52 83 114 145 207 269 24 54 84 114 144 174 234 294 51 21 57 118 149 180 211 273 335 75 44 13 117 147 177 207 237 297 357 103 73 43 140 171 202 233 264 326 3SS 154 123 63 742 1091 1456 1821 2186 2916 3646 728 459 " 9 " . "10 " . "11 " . " " " midnight " " " 2 a.m. „ "4 " . From 4 a.m. to sunrise " 5 " " 15G 187 218 249 311 373 125 04 63 20 50 80 110 170 230 25 56 87 118 180 242 254 1 418 ELECTRIC LIGHTING. IS li.m. 11.40 11.40 11.40 11.40 11.30 11.20 11.20 11.20 11.20 11.20 11.20 11.20 11.10 ' 11.00 11.00 11.00 11.00 11.00 10.50 10.50 10.50 10.50 10.50 10.40 10.30 10.30 10.30 10.30 10.30 10.30 10.20 5, 0 ,0 0 ^ ? li.m. 5.30 5.30 5.30 5.30 5.30 5.20 5.20 5.20 5.20 5.20 5.20 5.20 5.10 5.10 5.10 5.10 5.10 5.10 5.00 5.00 5.00 5.00 5.00 5.00 4.50 4.50 4.50 4.50 4.50 4.50 4.40 to S3 < 2 if' p 0000000000000000000000000000000 G iq lOiOiOOC = - Z SOOO rH r«rHr-HHHHrtHCj OOS50HlMW#lCH01>00»OHtlM^10(DI>COOO HHHHr(HHrlHHN(NN(MNN(NNNNn 5' b zooooooooooooooooooooooooooooooo Sqqio lo iq 5 i ■ ■ ' -r- -+ oo co co co co co co co oi oi c-5 ci -r* i-< r-j <-; ^ t-j h ^OTosodcoo6o6ododo6o6o6odo6o6o6o6odo6o6cocxjo6o6coo6o6odod z 3 c -a £ = z 3^ a 5c ='0000000000000000000000000000000 SooiottL: ■ ■-. > - 1 .0 -+ -H -t •* -f< -f_ -+ -+ -h ■* •* -H 00 00 co 00 00 00 co ^'^Ttico^cocococo^co^cococococo'co'co'coco'co'co'co'coco'coco'cocococo 3 S888888S888SSSSSSSyS2'SSgSSSSSSgg M 3- HNM^WCONCOQOHfinjwONWOOwcJ^OONgOOOjH c 000000000000000000000000000000 ■ - :-. CO CO CN CI C\ ClrtHHrirt ,£] ©v© © O O O © O O O O CS OS OS OS OS OS OS OS OS C O OS OS OS OS OS OS OS OS 0 — ' 5 0 rZ s H <»lSOH«W-*LOOI>COOOH«M^LOffl^OOaO 0ooooooooooooooooo goqqoo«riCi:i:icic-:cMo:M:c lo lo i- lo lo lo © © © © t-h-i # "#' ■* -#" rj< -#' -(< ■#" r)H "* lO lO lO £ fr> or©©©©©©©©©©©©©©©©©©©©©©©©©©©©©© C CO CO CO M CO Ol C! Ol 01 Ol 0) « r- ^ rn rn rt O O O O O C LO LO LO LO LO Of o Hi z iHccooi©i-ic^co-*KOj©t-co©©^o'ico-#«i?ot-ooc50 «S H n » 6 £ s ©©©©©©©©©©©©©©©©©© ©©©©©©©©©©©©© S m io lo lo lo lo lo lo o rn h ri rn h ^ ^ : i : i •; :: :: :: CO -t f f f f LO LO o § h»00C0CCC0»»00OQOOOOOOCSOOOOC)CSQC»OCBOO05O 01 Jl SO©©©©©©©©©©©©©©©©©©©©©©©©©©©©©© W M ^ff^Th^fffffffffffTh^ffffffffffff-* 0 s 2S22SS2222SS§8S88Sgggggg?^?5§^^^ o 3 ,^C-l>l>r>t>t-t-t>t-t-t-t-t-t-t-t>CO<»0?0«Ci«0«i«0?D5DO<10CCOCO ,0 3 1-1 rt >."£ ris»*ioot-»nO'-i;in-tiL03t-cooo^:icoiNl-l>t>l>t-t>l>t-t-l>Nt-t-l>t>t»t-l>«>Nt-t>t>t>t-t^t-l>N g °5 HIMCOflOfflt-OOOOWClCO-f LOONKCC^OlCO^lOCKODOO-l AS H LIGHTING TABLE. 421 h.m. 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.50 13.50 13.50 13.50 13.50 13.50 13.50 13.50 13.50 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 13.40 o ol 2 o "o J: "ci H 3 -o CO r^OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO fl r- f r- < tH t—i t— 1 r-< i-H r— ( t— ' r-1 C} CI CI C} CI CI CI CI CI CI CI 0 1 CI CI oj CI Cl 01 CI CI CI o fi i 3 C992SSSS920020SOOOOOOQOOOOOOQOOO £COCOCOCOCOCOCOCOCOCOCOCOCOCOCOOOCOCOCO-tH-1H--f*--f-t<-f,r-t<^t,-3H.^-^iT3H co p o^3 TH<^Otl^lO»I^OOC50^C1CO-*lOCOt-OOCipT-IC1Cp-f ~' ^oooooooooooooooooooooooooooooo gwco^ffliBioiqiqoioiqrHHrtHHrtHrtHt] c; c; ci ci co co cq cq ^ « ci ci ci ci ci ci ci ci ci ci ci co co ci co co co ci ci co co co" co co co' co" co co co' co K5 M ^"oooooooooooooooooooooooooooooo 5 CO CO CO •*"*■***"*•*•* t »" ' ' - ~ q O © H ^'ifiicoo'ioKiiriiriioicinioioic'^^^o^iriQScD^cocs'ocdcocdcd ho 3 ^oooooooooooooooooooooooooooooo Sqquo uq o lo uq o lo lo o ** ^ -# -# **-*-# -# -* -* -h ~r -v -t< cq cq co cq co 5 ^ a w - : fig HiMCO'tLOONOOCiOHClcO-tO'-Ot-COr. e — < ~ 1 co ;: ■' i_rON00C5O s •ooooooooooooooooooooooooooooooo C ci ci ci ci o i : : t - - -,- Lq iq uq uq io © q r-j ^h h ih r-j th th cq co cq co co co o 8 CO c © H « H ffl O H O O 3| rjooooooooooooooooooooooooooooooo H.OOOOOOOHHrtHHHHHrH | | !N CO CO CO CO CO CO ^"uoiriioioicioiduo^uoouoiouooo'icuo'iriioirio ^lOioooioioioioioioioioioioioioioioioioioioioiOL'jioL'jioaiaio c ~ fig i-icqcO'#io«oi>GOo;0'rMCJco-tiio^cit^cooO'-jcico-+iio«ot-oociO'rH 422 ELECTRIC LIGHTING. Hours of Lighting* per Annum Uy Different Schedule.*. Regular all-night schedule 4000 hours New York City schedule « . . . 3950 hours Philadelphia schedule 4288 hours Providence schedule ........... 4012 hours Philadelphia moonlight schedule ..... t 2190 Lours Frund schedule ............. 3000 hours Hours of Burning- Commercial 'Lights. Time of Sunrise and Sunsets. o o nfl o^ All night "3 «3 X X PX P Poo Pes 03o PS <£o PS 41 O P0^ S 03 x£ lights. ^ o 2 P bco "uarH bo P'Sfl h.ni ll.IIl h.m. h.m. h.m. h.m. h.m. h.m. h.m. 1 Jan. 15 4.55 4.30 3.30 4.30 5.00 5.30 6.30 7.30 7.25 8.00 15.30 Feb. 15 ;>.;n 5.0(1 3.00 4.00 4.30 5.00 6.00 7.00 6.56 7.30 14 30 Mar. 15 <;.<><; 5.30 2.30 3.30 4.00 4.30 5.30 6.30 6.12 6.45 April 15 6.41 (i.15 1.45 2.45 3.15 3.45 4.45 5.45 5.16 5.45 11.30 May 15 7.13 6.45 1.15 2.15 2.45 3.15 4.15 5.15 4.39 5.15 10.30 June 15 ,..H 7.00 1.00 2.00 2.30 3.00 4.00 5.00 4.24 5.00 10.00 July 15 '/.:;■_' 7.00 1.00 2.00 2.30 3.00 4.00 5.00 4.39 5.15 10.45 Aug. 15 ,.00 (1.30 1.30 2.30 3.00 3.30 4.30 5.30 5.08 5.45 11.45 Sept. 15 <;.()!» 5.30 2.30 3.30 4.00 4.30 5.30 6.30 5.40 6.15 12.45 Oct. 15 !>.!!) 4.45 3.15 4.15 4.45 5.15 6.15 7.15 6.13 6.45 14.00 Nov. 15 4.:;:) 4.00 4.00 5.00 5.30 6.00 7.00 8.00 6.52 7.15 15.45 Dec. 15 4.31 6.06 4.00 5.30 4.00 5.00 5.30 6.00 7.00 8.00 7.20 7.45 15.45 Aver'ge )* for y'r ) 1.30 3.30 4.00 4.30 5.30 6.30 5.54 6.26 13.00 Graphic Lighting Schedule for London, England. r— 3 6 1 fri- '"""'"'""i ; i r ijfllfi \H I 1 . : I AHH. \\ ! \\ \ { 1 //\' ■ 1 1 \]\ 1 •£■' |s\ SE. . jA''' ! 11'^ / '' 1 ■ 1 ! ■ % DEC. / v Uii:.!:.:,h;: 1 i i ! : . k Fig. 7. — The shaded area represents the time during which light is required. The horizontal lines show the months of the year. The vertical lines show the hours of the day and night. The inner dotted lines show the time of sunset and sunrise. The outer lines show the time of lighting up and extinguishing. Each square is an hour month, i.e., 30.4 hours. ELECTRIC STREET RAILWAYS. CARS, MOTORS, AID GRIDES. (From Pamphlet by S. H. Short, issued by Walker Company.) Grades and sharp curves should of course be avoided as much as possible, but when unavoidable, the ascent of a 10 per cent or even a 12 per cent grade is possible to a car fitted with a double 15 h. p. or 20 h. p. equipment, and pulling no trailer. The grip of the wheels on the rails may be depended upon, with the aid of sand, to give from 250 to 300 pounds pull for each ton of weight upon them, in even the worst weather. On nearly level roads (having nothing steeper than a 2 per cent grade;, a single 25 h. p. equipment will handle a car, and in a pinch pull a trailer. Ordinarily, however, it is not advisable to use a trail car with a single 20 h. p. equipment, as it makes a slow start and a slow maximum speed. A single 30 h. p. equipment should be able to handle a short car and trailer satisfactorily on roads with noth- ing greater than 2 per cent grades. While the power of a 30 h. p. motor could be depended upon to climb steeper grades, the adhesion of the wheels in bad weather cannot be. Siu.gle 20 or 30 h. p. equipments will handle 20 ft. or 22 ft. cars nicely, when no trailer is used, on as high as 4 per cent grades, and even steeper in good weather, the failure being, as previously explained, not in the power of the motor, but in the adhesion of the wheels to the rails. The 30 h. p. motor has the advantage of the 20 h. p. in giving a quicker start and higher speed on grades. Single motor equipments are, however, not advisable, on account of the liability of a single pair of drivers to slip in bad weather. They will prove especially annoying where snow- storms are of frequent occurrence, or where the track is liable to become icy. All long double-truck cars should have double equipments, as their greater weight requires greater power to bring them up to speed quickly, even on a level. On roads with over 4 per cent grades, whether it is pro- posed to haul trail cars or not, double equipments should be installed. A double 25 or 30 h. p. equipment will handle a trail car on a 6 per cent or 7 per cent grade, the advantage of the 30 h. p. motors again being the higher speed on grades and quicker start. On roads where the traffic is sufficient to warrant the use of trailers with short cars, but the grades exceed 7 per cent, long cars on double trucks, or radial trucks, with double-motor equip- ments should be substituted. These will climb nearly as steep grades as the smaller cars, without trailers. Long cars are not advisable except in the case just named, and on long runs where the stops are few, as the time required for the letting off and taking on of passengers is excessive. Finally, on roads where traffic, such as fairs, base-ball games, etc., has to be handled, giving light loads most of the time, but few exceedingly heavy ones, the most economical arrangement is that of 30 h. p. double equip- ments, hauling two trailers, when the heavy traffic is to be handled. This combination can be depended upon for grades not exceeding 3 per cent in bad and 4 per cent in good weather. CURVES. A 30 ft. radius curve on grade adds about as much to the resistance of a car as 4 per cent additional grade. It will consequently be frequently found impossible to start on such a curve on grade in bad weather without sand. Sand boxes should, then, be a part of every car's equipment. Sharp curves on grade should always be avoided if possible, as they are the cause of great annoyance on wet or icy days. 423 424 ELECTRIC STREET RAILWAYS. §TATIOHf. A station should never contain less than two dynamos. It is desirable also for the steam plant to be composed of two or more units if possible ; but on very small roads, say under five cars, this is of course impracticable. The general plan of a station should be such that the disabling of one dyna- mo or engine could not cause a shut-down on the road. For roads of 15 cars or less, where the fluctuations of load are exceedingly violent, simple high- speed engines are undoubtedly to be preferred. As the road grows larger and the load more steady, simple Corliss engines will give a somewhat better steam consumption. On a road of 40 cars or more, compound condensing engines, of either the Corliss or high-speed type, in units of such size that at least one can be kept fairly loaded at all times, will be economical. Always condense either simple or compound engines when water for that purpose can be had plenty and cheap. Never use compound engines non- condensing. Considering the increased expense and complication, together with the difficulty in regulating under widely and suddenly varying loads, the economy of triple-expansion engines in railway work is doubtful. The size of the engine should be always such as to give the maximum average efficiency with the variations of load in question. It should be noted here that this is not the same size engine which will give the maxi- mum efficiency at the average load. Where it is possible, belt directly from fly-wheels of engines to generator pulleys. Counter shafts give flexibility and make possible the use of larger steam units, but they consume a very appreciable amount of power, and are liable to give trouble otherwise. Concerning the amount of power per car in generators and engines, no general rule can be laid down, as three variables, viz., grade resistances, curve resistances, and traffic, must be considered in this connection. 25 h. p. (rated at J-i cut off), and 30 amperes per car for roads of 5 to 10 cars, and. 20 h. p. with 25 amperes per car for larger roads, would probably cover the demands. This, hoAvever, should be considered only as a rough esti- mate. The question of the amount, character, and location of power should be settled for each road separately by a thoroughly competent engineer, as a small variation from correct principles and design in this respect is liable to considerably increase the running expenses. The whole design should be based on Sir Wm. Thomson's principle, namely, that " The interest on the investment and the cost of such losses as could have been avoided by larger investment should be equal." SPECIFICATIONS vs. STA1¥I>JLI6» TYM3S. The series motor can easily be designed to fill two conditions as to speed and power in the same machine, provided always that the condition for the lesser power calls also for the greater speed, and that these two requirements are not too near alike in speed when the powers called for vary widely or vice versa — too near alike in phase when the speed varies widely. Standard motors for street-railway work are now designed to give a 20-ft. loaded car a speed of from 20 to 22 miles per hour on a level, and to develop NOTE. — In the selection of engines for electrical railway work, the best practice of to-day is to choose the engines in the same manner as for any other commercial manufacturing plant. For large installations, or where storage batteries are used for regulating the load, and so retaining fairly constant power requirements, the size and arrangement of the plant will determine whether the engines should be simple, compound, or triple expansion, and whether they should be run condensing or not, if water is a callable Engines should be designed with all shafts, pins, wearing surface, etc., heavy enough for the maximum loads or brer loads, but their cylinders should be so proportioned that the average loads be secured at the most economical point of cut-off. This gives strength for heavy load and economy for average conditions. Countershafts with friction clutches and pulleys are seldom installed to-day. Either direct-belted or direct-connected engines and dynamos are belter, rc(/uiring less engine-room area, expense for real estate, building, etc., and reduce friction losses aud cost of repairs. J- S. G. MOTORS AND TAX EQUIPMENT. 425 their full rated capacity (of 20 h.p., 25 h.p., etc.), at a speed of 10 miles per hour, when mounted upon wheels of a specified diameter (generally 33 inch). The voltage being kept the same, each speed corresponds to a certain horizontal effort or thrust at the circumference of the wheel, this horizontal effort increasing as the speed decreases. Therefore, for each different tractive resistance, he it due tc the condition of the track, to grade or curve, or to whatever cause, has for a given weight of car and load, a given speed which cannot be altered wi tin >ut altering at the same time the two speeds which the motor was originally designed to give. These speeds are most easily altered by changing the diameter of the wheel to a larger or smaller size than the standard, according as it is desired to increase or decrease the speed, or in S. R motors by changing the ratio of the gearing. In asking for designs for special motors, the weight of the maximum train and the maximum speed on level, together with the weight of the maximum train and the highest speed on the maximum grade, should be given. As before stated, within limits, any conditions as to speed on level and on grade can be approximated by special design. DESIRABLE POINTS I1V MOTORS AITD TAX EO.UIPMEIST. It is desirable that motors should be electrically sound, i.e., that their insulation should be high, mechanically strong, and waterproof. It is of great advantage in this connection if the entire frame of the motor can be insulated from the car truck and consequently from the ground, thus re- lieving the insulation of the armature and fields of half the strain. The mechanical difficulties in the way of accomplishing this, however, go a great way towards counterbalancing the advantage gained. A high average efficiency between three h.p. and full load should be ob- tained if possible, but mechanical points should not be neglected to obtain this. A motor should run practically sparkless up to § of its rated capacity. A low starting current is especially desirable, and for obtaining this nothing can equal a multiple series controlling device, which cuts the starting cur- rent actually in half. This device also enables cars to run at a slow speed with far greater efficiency than any other method. Mechanically, the motor should be simple. The fewer the parts, and es- pecially the wearing parts, the better, It should be well encased in a cover- ing strong enough not only to keep out water, pebbles, bits of wire, etc., encountered on the track, but to shove aside or slide over an obstruction too high to be cleared. At the same time, the case should be hinged so that by the removal of a few bolts access can be had to the whole interior of the motor. The brush holders and commutator should be easily accessi- ble through the traps in the car floor at ail times. As much of the weight of the motor as possible should be carried by the truck on springs ; if practicable, all of it. This arrangement saves much of the wear and tear on the tracks. A switch in addition to the controlling stand should always be provided, by which the motorman himself can cut off the trolley current, in case of accident to the controlling apparatus. Roads having; long, steep grades should have their cars provided with a device for using the motors as a brake in case the wheel brake gives out. There are several methods of accomplishing this, but limited space pro- hibits any description of them. Last, but by no means least, all wearing parts should be capable of being easily and cheaply replaced. NOTE. — Double brakes or track brakes should be used on roads with steep grades. Power brakes a,re seldom, used on ordinary cars. With the increase in the length, and weight of cars they will probably come into more general use, and orders hare been issued by the Railroad Commission of the State of New York that all street cars must be equipped ivith power brakes. 426 ELECTRIC STREET RAILWAYS. WEIGHTS OF BAILN. Pounds per "Weight per Mile. Weight per 1000'. Yard. Long Tons. Long Tons. 640 ^986.7 25 392240 320 39.286 7~2240 2080 7.441 30 472240 47.143 82240 8.929 35 55 1920 55 933.3 10 2240 2026.6 10.417 40 622240 1600 62.587 11 2240 880 11.905 45 702240 960 70.714 132240 635.5 13.393 48 742240 1280 74.428 14 2240 1973.3 14.284 50 782240 1600 78.571 14 2240 1066.7 14.881 52 812240 960 81.714 15 2240 826.6 15.477 55 862240 86.428 16 2240 16.369 56 88 320 88 1604.4 16 2240 . 586.7 16.667 58 2240 2080 91.143 17 2240 1920 17.262 58J 912240 640 91.928 172240 920 17.411 60 942240 960 94.286 172240 1013.3 17.857 62 972240 97.428 18 2240 1680 18.452 63 99 1760 99 182240 2013.3 18.75 63£ 992240 1ftJ20 99.785 18 "2240 773.3 18.899 65 102 U 2240 in„1600 102.143 19 2240 1440 19.245 66 103 2240 103.714 192240 1773.3 19.643 66| 1042240 1ftK640 104.5 19 2240 2106 19.792 67 1 05 — — 2240 1920 105.286 2240 533.3 19.940 68 1062240 106.857 20 2240 2000 20.238 70 110 111280 2240 110 202240 293.3 20.833 71 111.125 21 2240 21.131 WEIGHTS OF RAILS. 427 WFICJHTS OF M AIL§ — Continued. Pounds pei- "Weight per Mile. Weight per 1000 '. Yard. Long Tons. Long Tons. 320 960 72 11322io" 1920 113.143 212240 720.2 21.429 75 1172^40 117.857 22^240 2053.3 22.322 77 121 «™ 320 121 22 2240 480 22.917 78 122 ■ 2210 1600 122.143 232240 1813.3 23.214 80 1252240 1920 125.714 23 2240 906.6 23.810 82 1292240 1280 129.857 24 2240 666.6 24.405 85 2240 960 133.571 25 2240 1760 25.298 90 1412249 141.428 262240 186.6 26.786 91 143 143 27 2240 373.3 27.083 98 154 320 154 29^240 1706.7 29.167 100 1572240 157.143 29~2240~ 29.762 For iron or steel weighing 480 lbs. per cubic foot : Cross-section in square inches = weight in lbs. per yard -J- 10. For iron or steel having J conductivity of copper : Weight in lbs. per yard -^-11.6333 = number of 0000 B. & S. copper wires with combined equivalent carrying capacity. Also, weight in lbs. per yard X 18189.1 = C. M. of equiva- lent copper wire. KJLDIXJS OF CURVE! FOR DIFFEREJX CJI8ABES OF CXTRVATURF. 3 0 © ^ 1 ■S 3 ri b£ £1 b£> bo bo ©■d bD £1 ft « ft « ft « A & A « 11 1 5730 12 521 21 273 31 185 41 139 2 2865 13 477 22 260 32 179 42 136 3 1910 14 441 23 249 33 174 43 133 4 1432 15 409 24 238 34 169 44 130 5 1146 16 382 25 229 35 163 45 127 6 955 17 358 26 220 36 159 46 125 7 818 18 337 27 212 37 155 47 122 8 716 19 318 28 206 38 150 48 119 9 636 20 301 29 197 39 147 49 117 10 573 21 286 30 191 40 143 50 114 Note No. 1. — A 1° curve has a radius of 5730 feet; 2° curve, ^this; curve, I this, etc. 428 ELECTRIC STREET RAILWAYS. GRADES Iltf PER CEUfT AUD RISE IN FEET. Rise in Feet at Given Distances. Per Cent Grade. 500 Eeet. 1000 Feet. 5,280 Feet (1 Mile). i 2.5 5 26.4 1 5 10 52.8 1.5 7.5 15 79.2 2 10 20 105.6 2.5 12.5 25 132 3 15 30 158.4 3.5 17.5 35 184.8 4 20 40 211.2 4.5 22.5 45 237.6 5 25 50 204 5.5 27.5 55 290.4 6 30 60 216.8 6.5 32.5 65 343.2 7 35 70 369.6 7.5 37.5 75 396 8 40 80 422.4 8.5 42.5 85 448.8 9 45 90 475.2 9.5 47.5 95 501.6 10 50 100 528 11 55 110 580.8 12 60 120 633.6 13 65 130 686.4 14 70 140 739.2 15 75 150 792 Note No. 1. — For other distances interpolate the table by direct multi- plication or division. EIEVATIOlf OE OUTER RAIL OK CURVES. o . Speed in Miles per Hour 10 15 20 25 30 35 40 45 1 50 60 ob a A Elevation of Outer Rai in Inches. 1 5730 ft i til 1ft If It* 24 2 2865 * ft ft * H it* 2ft 2* 2ft 411 3 1910 * t* Aft 1* 24 3ft 4* 58 V* 4 1432 J ft A+* 21 3S 4* 4ft «t* y* 5 1146 # | i# 2# 3ft 4ft <* 84 12ft 6 955 tfr i* 2ft 3H 6ft «* 10* 7 818 HI 3 4ft 5f u* yft 11* 8 716 5ft 2i3g 3ft 4+ft 6t* 8i-i- 10* 9 636 8 2| 3x1 54 y* iat* 10 573 t* 2f 4| H 8ft 101 11 521 * H* 3 4nf 6i y* lit* 12 477 t» 3ft 5* »ta 12 J| 14 409 2ft 3x1 5tt «T9TT n+ 16 358 lis 24 4# fitf 9xs 18 20 318 286 W 2| 3ft 4« 5ft 8* 10| 12 Note No. 1. — Wli enE = : elevation ir i inches of o uter r ail above th( s hori- zo ntal p Lane: V = velocity of car in feet per second ; R = radius of curve in feet ; V2 Therefore E =1.7879 —when gauge of track is 4/-8£// 429 SPIKE§. Size. N0-2P^Sg0f ^-Per Spike. Spikes per Lb. 4§ Xi 533 3752 2.66 5 X/5 650 3077 3.25 5 X| 520 3846 2.6 5 XT96 393 50S9 1.96 5| x h 4G6 4292 2.33 5§ X r96 384 5208 1.92 6 XT9s 350 5714 1.75 6 X| 260 7692 1.3 SPIKES PER lOOO' AUTR PEB flttHE SIT¥€nLE IBACK, WITH FOUR SPIRES PER TIE. Spacing of Ties. Per 1000'. Per Mile. 10 ties to 30' rail 13334 7040 11 " " " " 1466| 7744 12 " " " " 1600 8448 13 ■"■"■"■ « 1733^ 8152 14 " " " " 1866|- I 9856 15 " " " " 2000 10560 16 " " " " 2133-J 11264 JOINTS PER MILE OE SIltfCilE TRACK. Per 1000'. Per Mile. Joints Angle Bolts — 30' rails 66§ 133J 266§ 400 533J 800 352 704 — 4 hole bar 6 " 8 " 12 " " 1408 2112 <( 2816 " 4224 TIES PER I©©©/ AITS PER IffflEE. Spacing. Per 1000'. Per Mile. 10 ties to 30' rail 11 " " " " 12 " " " " 13 " " " " 14 " " " " 15 " " " " 16 " " " " 333i 366| 400 433J 463§ 500 533i 1760 1936 2112 2288 2464 2640 2816 BOARD EEET, CUBIC FEET, AID SQUARE FEET OE BEARIH& SURFACE PER TIE. Size. Board Feet. Cubic Feet. Bearing Surface 5" X 5" X 7/ 14.56 1.213 2.91 5" X 6" X 7' 17.5 1.458 3.5 5" X 7" X 7' 20.41 1.7 4.08 5" X 8" X 7/ 23.33 1.944 4.66 6" X 6" X V 21 1.75 3.5 6" X 7" X V 24.5 2.041 4.08 6" X 8" X 7' 28 2.333 4.66 6" X 9" X V 31.5 2.625 5.25 6"X10" X7/ 35 2.916 5.83 6" X 8" X 8' 32 2.666 5.33 6" X 9" X 8' 36 3 6 6"X10" X 8' 40 3.333 6.66 430 ELECTRIC STREET RAILWAYS. REPORT OF E. S. DEPARTMEJIT OF ACiRICEE- 1URE OJT DUKABIMTY OF RAILROAD TIES. White oak 8 years. Chestnut 8 " Black locust 10 " Cherry, black walnut, locust 7 " Elm 0 to 7 " Red and black oaks 4 to 5 " Ash, beech, and maple 4 " Redwood 12 " Cypress and red cedar 10 " Tamarack 7 to 8 " Longleaf pine 6 " Hemlock 4 to 6 " Spruce 5 " PAVIHTG. Paving prices vary so that it is difficult to state even an approximate cost that will not be dangerous to use. Prices are not at all alike for asphalt, even in cities in tbe same localities ; other styles vary according to prox- imity of material, cost of labor, and amount of competition. Square yards of paving between rails, 4' 8|" gauge, less 4// for width of carriage tread : ' Per 10007 run = 485.89 Per mile run z= 25G5.5 Square yards paving for 18" outside both rails : Per 10u0' run = 333§ Per mile run = 1760 Approximate Cost of Pavin §•• (D ivis, > PAVEMENT. Cost of all Material and Labor. Cost of Tearing up Existing Pavement and Repla- cing as Found. 6* xix \ Ti j 1 I ^T J_ up N WIND WITH 20 O.C. COVERED Fig. 6. the circuit and are held in contact by this current also passing through the magnets. The armature B normally rests out of the influence of its magnet. A rod entering from the bottom of this box shoves the armature up into con- tact with the ends of the magnet, and is held in this position until the circuit is broken. The current from the trolley enters first through a lamp, then through the magnet-winding to the frame. When the armature is up the current passes down the arm holding the armature, and then through the signal line to the distant semaphore box. The semaphore box contains a pair of solenoid magnets, which set the semaphore disk and light the lamps. These lamps are arranged behind a red glass disk inserted in the semaphore box. The disk is set by means of a solenoid operating a bell crank and link, which turns the semaphore rod and displays the red disk. The dimensions and methods of general con- struction employed are shown in Fig. 6. The circuit first passes through three lamps, then through the solenoid, and out to the signal releasing-box. The construction of this box is shown in Fig. 5, and consists of a switch and a lamp in circuit with this switch. It is operated by pushing up the rod, and when the rod is released the blade falls back into position, but it will not close the circuit iioav ; for on opening the circuit, the magnet in the cir- cuit-making box dropped its armature, and opened the current at the dis- tant switch, which can now only be closed by the conductor on the car following. The diagram of connections is given in Fig. 4. Covered No. 10 iron wire can be used. Robert Doumblaser developed all the details. 436 ELECTRIC STREET RAILWAYS. EIST OE MATERIAL 1SE(41"IKEI» EOR (tt'B MTEE OE OVERHEAD OAE FOR ELECTRIC STREET RAILWAY. 1 Mile Overhead. Curve Overhead Material. Anchor- Material for Railway Construction. Cross Suspen- sion. Bracket Suspen- sion. Main Line. Branch Line. i o age. H M 33 ■ H P o H 3d 33 H la o ft '3d 33 H 3 o ft 33 H a? O A 33 O ft u No. 0 B. & S. H. D. Trolley Ft. Lb. -,1'SO 10560 3369 52S0 ION,") 10560 3369 250 80 o O No. OB. &S. S.D. F'd'rT'ps Ft. Lb. 400 154 500 192 90 35 180 69 3 7 strand No. 12 span Ft. Lb. :;<;oo 756 3600 756 800 168 koo 168 soo 168 KOI) 168 200 42 400 84 600 122 > 1 7 strand No. 15 guy Ft. Lb. ;;ooo 300 4500 450 1500 150 2000 200 100 10 100 10 100 10 100 10 Plain ears .... Strain ears .... Splicing ears . . . Feeder ears .... 45 1 10 45 45 90 2 20 45 1 10 45 45 90 2 20 5 2 7 10 4 4 4 5 1 6 6 15 2 "If 17 4 4 4 — Insulating caps . . Insulating cones . . 90 90 90 90 M O w Straight line . . Single curve . . Double curve Bracket . . . 45 90 45 90 3 4 3 11 3 12 4 2 2 i Stra Tur Sect Froj Fro Har Eye Cas1 Gas Cros in insulators . . abuckles . . . ion insulators 90 90 2 45 90 45 90 90 4 45 90 45 45 45 48 4 90 90 48 4 4 2 4 4 2 2 2 1 2 2 1 2 2 2 y crossings . . . dwood pins . . 2 -iron brackets . pipe arms . . . s arms (l^'-lS) . Cros & Boll r(" Lag et Lag ar Lag s-arm braces 'X8") s for brackets 'X4") 90 45 144 90 45 144 45 45 90 90 screws for brack- 3 (i"X7'0 . . . screws for cross ms(!"x3") . . screws for braces Poles, 125-ft. apart . 90 90 45 45 2 2 2 2 2 2 Cha Bon Ligfc anel pins . . . ds 800 400 1600 800 KOI) 400 1600 800 tning arresters . 3 3 3 3 Section switch boxes 2 2 2 2 PLATE BOX POLES. 437 Plate Box Poles. BY BUFFALO BRIDGE AND IRON WORKS =LO=L SCREW PIN: M 438 ELECTRIC STREET RAILWAYS. TUBULAR IRON OR STEEL POLES. By Morris, Tasker, & Co. (Inc.). Size. Wrought Iron or Steel. Length. Weight. No. 1, light . No. 1, heavy No. 2, light „ No. 2, heavy No. 3, light . No. 3, heavy No. 4, light . No. 4, heavy 5 in., 4 in., 3 in. 5 in., 4 in., 3 in. 6 in., 5 in., 4 in. 6 in., 5 in., 4 in. 7 in., 6 in., 5 in. 7 in., 6 in., 5 in. 8 in., 7 in., 6 in. 8 in., 7 in., 6 in. 27 ft. 27 ft 28 ft. 28 ft. 30 ft. 30 ft. 30 ft. 30 ft. 350 lbs 500 lbs 475 lbs 700 lbs 600 lbs 1000 lbs 825 lbs 1300 lbs POLES. Dimensions and W^eig-hts W^roug-ht-Iron and Steel Poles. Length. Diameter. Weights. 27 ft. 28 ft. 30 ft. 30 ft. 28 ft. 30 ft. 5 in., 4 in., 3 in. 6 in., 5 in., 4 in. 6 in., 5 in., 4 in. 7 in., 6 in., 5 in. 8 in., 7 in., 6 in. 8 in., 7 in., 6 in. 350 lbs. to 515 lbs. 475 lbs. to 725 lbs. 510 lbs. to 775 lbs. 600 lbs. to 1000 lbs. 775 lbs. to 1260 lbs. 825 lbs. to 1350 lbs. Cubic Contents of Wooden Poles, in Eeet. Length. Diameter. Section. Cubic Feet. 27 ft. 6 in. X 8 in. Circular 7.36 27 ft. 7 in. X 9 in. Circular 9.56 27 ft. 7 in. X 9 in. Octagonal 10.1 28 ft. 7 in. x 9 in. Circular 9.92 28 ft. 6 in. X 9 in. Octagonal 10.46 28 ft. 8 in. x 10 in. Circular 12.52 28 ft. 8 in. x 10 in. Octagonal 13.2 30 ft. 7 in. x 9 in. Circular 10.63 30 ft. 7 in. x 9 in. Octagonal 11.21 30 ft. 8 in. x 10 in. Circular 13.41 30 ft. 8 in. X 10 in. Octagonal 14.15 30 ft. 9 in. X 12 in. Octagonal 19.06 Rake of Poles. Wooden poles should be given a rake of 9 to 18 inches away from the street. Iron or steel poles set in concrete need be given but 6 to 9 inches rake. Corner poles, and those supporting curves, should be given additional rake or be securely guyed. AVERAGE WEIGHTS OF WOOD. 439 AVERAGE WEIGHTS ©IT VADIOUS WOODS, IJ¥ POUUfBS. osa? isl Live oak White oak .... Red oak Chestnut Southern yellow pine Northern yellow pine Long-leaf yellow pine Norway pine . . . Spruce Hemlock Perfectly dry Perfectly dry Perfectly dry Perfectly dry Perfectly dry Perfectly dry Unseasoned Perfectly dry Perfectly dry Perfectly dry The weight of green woods may he from one-fifth to one-half greater than the weight when perfectly dry. DIP Il¥ SPAUT WIRE. (Merrill.) The following tables give the dip of the span wire in inches under the combined weight of span wire and trolley wire, for various spans and strains. Length of trolley wire between supports, 125 feet. Weight of trolley wire, 319 lbs. per 1000 feet. Weight of span wire, 210 lbs. per 1000 feet. Single Trolley "Wire. Spans in Strain on Poles, in Pounds. Eeet. 500 800 1000 1500 2000 2500 3000 30 7.8 4.9 3.9 2.6 1.9 40 10.6 6.5 5.3 3.5 2.7 50 13.6 8.5 6.8 4.5 3.4 2.7 60 16.7 10.4 8.3 5.6 4.2 3.3 2.8 70 19.9 12.4 9.9 6.6 4.9 4 3.3 80 23.2 14.5 11.6 7.7 5.6 4.6 3.9 90 26.7 16.7 13.4 8.9 6.6 5.3 4.5 100 30.3 18.9 15.2 10.1 7.6 6.1 5.1 110 34 21.3 17 11.3 8.5 6.8 5.7 125 37.9 23.7 18.9 12.6 9.5 7.6 6.3 Two Trolley Wires, 1© Feet Apart. Span in Strain on Poles, in Pounds. Feet. 500 800 1000 1500 2000 2500 3000 3500 40 15.4 9.6 7.7 5.1 3.9 3.1 50 20.8 13. 10.4 6.9 5.2 4.2 60 26.3 16.4 13.1 8.8 6.6 5.3 4.4 70 31.9 19.9 15.9 10.6 8. 6.4 5.3 80 37.6 23.5 18.8 12.5 9.4 7.5 6.3 5.4 90 43.5 27.2 21.8 14.5 10.9 8.7 7.3 6.2 100 49.5 30.9 24.8 16.5 12.4 9.9 8.3 7.1 110 55.6 34.7 27.8 18.5 13.9 11.1 9.3 7.9 120 61.9 38.7 30.9 20.6 15.5 12.4 10.3 8.7 Note. — See also chapter on Conductors. For table of stranded wire for spans and guys see page h 18, Properties of Conductors. 440 ELECTRIC STREET RAILWAYS. Span wires should be stranded galvanized iron or steel, sizes J inch diameter &, J, or § inch according to the weight of trolley wire, etc., to be supported. Where wooden poles are used it is not necessary to provide other insulation for the span wire, and the wire can be secured to the loop Fig. 8. Section of Track and Overhead Construction in Broad Streets, showing Double Overhead Wires and Underground Feeder Conduits. Section of Track and Overhead Construction in Narrow Streets, showing Overhead Pipe Brace. Trolley Suspension for Havana Streets, as developed by F. S. Pearson. SIDE BRACKETS. 441 of an eye-bolt that is long enough to pass through the pole at a point from twelve to eighteen inches below the top, and that has a long thread to allow taking up slack. On many roads in the country the span wire is simply wrapped around the pole top, using a number of feet more wire, making it difficult to take up slack, and presenting a slovenly appearance. Where metal poles are used it is necessary to insulate the span wire from the pole This has been done in some cases" by inserting a long wooden plug in the top of tubular poles, capping it with iron, the wooden plug then being pro- vided with the regular eye-bolt. The most modern way is to provide a good anchor bolt or clasp on the pole, then insert between the span wire and this bolt one of the numerous forms of line or circuit-breaking insulators devised for the purpose. If the anchor bolt is not made for taking up slack, the insu- lating device can be so designed as to be used as a turnbuckle. Of course insulation must be provided for both ends of the span wire. Span wire must be pulled very taut when erected so that the sag under load will be a minimum. Height above rail surface should be at least 18 feet after the trolley-wires are in place. This height is regulated by statute in some States, and runs all the Avay from IS to 21 feet. Figures 8, 9, and 10 illustrate one of the most modern installations, that at Havana, Cuba, as designed by Mr. F. S. Pearson for double trolley. \feh *0 COPPER1. 'galv.iron STRETCHED FIG. 10. Views of Trolley Spans with Plus and Minus Feeder connections and Plan of Double Track Y, showing Location of Insulators. SI»E BRACKETS. Along country roads and in such places as the track is along the side of the roadway or street, it is customary to use single poles with side brackets to support the trolley wire. Where side brackets are used it is not safe to place the pole less than four feet away from the nearest rail, and to give flexibility to the stranded sup- porting wire, now always provided for the trolley wire, the bracket should be long enough to reach the distant rail, thus giving a little more than two feet of cable for flexibility. A common length of bracket is 9 feet. Figures 11 and 13 show the simple form of side bracket in most general use, and Figs. 12 and 14 show variations of the same. It is obvious that this method of support may be made as elaborate and ornamental as may be desired. On double-track roads center-pole construction is sometimes used, in which poles are placed along the center line between the two tracks, and brackets are erected on each side of the poles overhanging the tracks. Where wooden poles are vised a good form of construction is to bore the pole at the proper height and run through it the tube for the arms, this long tube being properly stayed on both sides of the pole by irons from the pole- top to the bracket ends, or by braces against the pole. The trolley support- ing wire can extend from end to end of the brackets through the pole, or 442 ELECTRIC STREET RAILWAYS. can be cut at the pole, and eye-bolts be used, as in tbe side-bracket construc- tion shown by Eig. 11. Fig. 11. Single Suspension. For Wood Poles. Figures 15 and 16 illustrate simple forms of center-pole brackets. Fig. 12. Single Suspension. For Wood Poles. Center-pole construction is quite often used on boulevards in cities where tbe brackets and poles can be made quite ornamental. Fig. 13. Single Suspension. For Iron Poles. TROLLEY WIRE SUSPENSION. 443 Fig. 15. Double Suspension. For Wood Poles. FiG. 16. Double Suspension. For Iron Poles. TROLLEY WIBE SUSJPEWSIOW. The support of the trolley wire along straight lines is a. simple matter and needs no explanation ; at curves and ends there have been some simple forms developed in practice that are handy to have at hand. FolloAving are some of the points : Terminal anchorag-e. — Single track. See Fig. 17. Line anchorag'e.-See Figs. 18 and 19. To be placed at the foot of all grades, at the top of hills, and at tangents, three (3) per mile is good practice ; where curves are frequent they will afford all the anchorage necessary. 444 ELECTRIC STREET RAILWAYS. Fig. 18. Single Track. Fig. 19. Double Track. Turnout and Siding* Suspension. — Following is a sketch of a very simple arrangement of suspension and guys for a single-track turn-out. Fig. 20. Curves, Suspension, and Ours. — The suspension of the trolley wire at curves is complicated or simple, according as the track may be single or double, or the curve may be at a crossing or a simple curve. Below are sketches of several types of suspension for different forms of curves, for single and double track, for cross suspension, and for center-pole construc- tion. Fig. 21. Simple Right-angle Curve, Single Track. Fig. 23. Double Track, Right-angle Fig. 24. Double Track, Right- Turn, Cross Suspension. angle Turn, Center Pole. Crossing's, Suspension, and Guys. — Simple crossings of tracks make no complication in the suspension of the trolley wires. When curves are added to connect one track with the other, complications begin, and GUARD WIRES. 445 where double tracks cross double tracks, aud each is connected to the other by curves each way, the network of trolley wires becomes very complicated. Following are sketches of a couple of simple crossings which will clearly enough illustrate the methods of suspension commonly used. L Fig. 25. Single-Track Cross- Fig. 26. Single-Track Crossing, ing, Cross Suspension. Cross Suspension. €HJAIt» WIRE§. Where trolley Avires are used in cities or in any location where there are other overhead conductors liable to fall across the trolley wire, it is custom- ary to place guard wires parallel with but above the trolley wire, as shown in the following sketch. A piece of No. 6 B. & S. galvanized iron or steel CROSS SUSPENSION WITH GUARDS FOR TROLLEY^WIRE Fig. 27. wire is drawn taut above the regular suspension wire ; porcelain insulators are secured to the same at a point about a foot or 18 inches either side of the trolley wire, and through these insulators is threaded and tied a No. 10 gal- vanized iron wire. This guard should be broken at least every half-mile where it is in any great length, as it is not advisable to have it a continuous conductor for any great distance, and it is advisable to avoid its use wher- ever possible. IBMETEJtlMlSATTOlSr OF MOST ECOlVOlfllCAI, »«]¥- SIXT OF CrRBEIl O STREET RAILWAY COWBXCTOItS. (See Chapter on " Conductors," also paper by Mr. H. M. Sayers.*) Wherever there is danger of interference with other properties from elec- trolysis it is desirable to have the drop in rails quite low, the B.T. regula- * See Trans I. E. E. for July, 1900. 446 ELECTRIC STREET RAILWAYS. tions being 7 volts between points on rails. This of course means track return feeders, and in some cases "negative boosters," or boosters on the track feeders. The formula was developed by Professor Perry from Kelvin's law, and following is Mr. Sayers's application of it to tramway work : — formula for Determining- tlie Most Economical Current Density and Drop in Conductors for Tramway Lines. R =. percentage or rate to be charged on complete cost of cables laid ready for use, representing interest and depreciation and maintenance, say 7 per cent. Hours run per year, at 15 hours per day, for 365 days = 5475. w= number of watts continuously wasted in distributing system, that would cost one dollar, at a rate of 1.5 cents per k.w. 100 cents 10 1Er ...... , __ — £^i — ; — tit- = 12. 15 watts for one dollar. 5475 X 1.5 1000 p = cost of copper per ton of 2000 lbs. @ 30 c. per lb. laid complete readv for use = $600. m= tons (2000 lbs.) copper per mile fori square inch cross-section = 10.2 tons. r = resistance per mile of copper of 1 square inchcross-section= .0455 ohms. t= most economical drop per mile in volts. then ^jR.w.p.m.r. _ -\/7 X 12.15 X 600 X 10-2 X -0155 _ 100 100 t — V 236.8 = 15.37 volts per mile. t— — = — — — =z 388 amperes per square inch. .0455 .0455 It is obvious that the distance that the current can be transmitted at the economical density is limited by the permissible drop in the distributing system. The total drop is usually divided somewhat as follows, and is varied to suit conditions. Drop in feeders 50 volts. Drop in trolley 5 " Drop in track return 5 " Drop in return feeders (boosted) .... booster. Thus the distance over which an unboosted feeder will carry current with- out exceeding the drop is determined as follows : 50 volts drop in feeder „ __ „ . ... — — -— .^ ., rr- = 3.25 miles, m this case. t =z 15.3* volts drop per mile Where feeders are " boosted" it is necessary to introduce in the formula, the factors of the cost of the booster and its losses, changing the value of " w " and therefore that of " t" let a =z cost per annum per k.w. for interest and depreciation on cost of booster, say $7.50. b = cost per annum for supplies and maintenance of booster, say $2.50, say the efficiency of the booster is 75 per cent, 100 and w = — „nopT , ^ „ — - — — = 8.37 watts for 1 dollar. /.1827 + 1.5 X 100\ 5475 \ 75 )X 1000 Using the same values as in the first equation, . _ / 7 X 8.37 X 600 X 10.2 X .0455 , f — 4 / Job" — Vi63 — 12.76 volts per mile. t 12 76 and -ttjv^ = ri = 281 amperes per square inch as the most economical cur- rent density for boosted feeders. Determination of the most economical drop, or limiting distance on the track may be made by the above formulae, but calculations may be expe- dited by use of a constant, as follows. Let HORSE-POWER OF ACCELERATION. 447 c = constant for ampere miles. «.= resistance of track per mile, say .03 ohms. d = limit of drop permitted in rails, say 5 volts. Then c = — = — =166 ampere miles. Thus, if each car requires an average of 20 amperes the limit in miles of track for a drop of 5 volts would be for the above values, 166 — (20 x no. of cars, say 5) = 1.66 miles, provided all tbe cars were bunched at the end, or that one or two cars were ascending a heavy grade, requiring the same amount of current. To determine the greatest length of track that can be economically used without feeders, Adhere cars are scattered along a line, the distances intervening between the power-house, or other power or feed- ing center, and each car, are multiplied by the amperes required per car, and the sum of these products must not exceed the value of " C," as follows : 1 car .5 miles from power-house, 20 amperes c = 10 1 " 1.5 " " " " 20 " c= 30 1 " 3. " " « " 20 '• c= 60 1 " 5. " " " " 20 " c = 100 Total c = 200 In this case c = 200, or more than the limit of 166 ; therefore tbe feeder point must be between the third and fourth cars, and the distance will be governed much by the grade between these points, for it is obvious thai; each of the above cars will take a much larger current than stated when ascending grades, and the value of this extra current must be carefully determined before making the calculations. HORSE-POWER ©JF ACCEIERATIO]¥. The following diagram shows the power required to accelerate one ton, when running at any speed, to the next higher speed in miles per hour. HORSE-POWER EXERTED FOR EACH TIME. 1 5 —10 1 5 20 25 3 5 35 4 3 -4 5 50 / / j / / / / / / / / 1 / 1 / / / / / gJ_ 1 / 1 U\ / / / / w £ 1 t / / / y §1 ih -■?/ / / / j\ / « * it!/ ,*7 \i i -A ?/ *>/ / / r_5M7 o/ i/\J/ 2zz: i r~T / r$~i W& - 8 ?/ // j I *w HORSE PO WEF * / 7 ■/ ffs Ly 1/ NECESS !E-tERA-T 1 B AKY :-on ( TO E-TC ~T / / 1 1 / / / > _f / '/ i // / u / C HARLES HENRY DAVIS C. I. T // >/ // '/. / CONSULTING ENC IN.EE R. _T '/ // // // / / // // 'A / NOT 3TH >. NE ;ess RY T 0 AC :ele ^ATE OTH :r /// /// w T.HI S, MULTIP ! DIAGRAM 1 ?ED IN TO LY T BY T E H. HE W 200 P. FOR Q\ :iGHT;-TO 1 LBS-, TH E TON AS iE ACCEL JS: FOR 2 FOU RAT 0 — /// ^ / § / M JLTI ! 2; OR 3 TONS BY : 1 ET 7 - Fig. 28. Copyrighted, 1901, by Charles Henry Davis. All rights reserved* 448 ELECTRIC STREET RAILWAYS. Power Curves. — For convenience in quickly ascertaining the horse- power required to propel a car of known weight under known conditions of speed and grade, the curves shown below have been calculated. The quantities which the various lines represent are clearly marked in the cut, but for the benefit of those who may be unfamiliar with such dia- grams, the following explanatior /.-■ inserted: The left-hand portion of the lower horizontal line represents t_ie speed in miles per hour ; the right-hand portion of same line, the h. p. per car ; the oblique lines in left-hand side of cut, the per cent grade as marked on each line ; the oblique lines on right- hand side of cut, the weight of car as marked ; while the vertical line in centre of cut represents the h. p. per ton. \ \ \ ■ A \ i\ \ c: z -v? 4 <7 \ V s % fc\ I ° u* o> 'S r o > *\ \ \ *" a <*\?o N \ * 0 ^ \ \ \ 5, \ \ 0 CO 7p- ~^ N \ \ g >i \ > \ h s^ jjj NC 1 JJ. ■H -s^ O t J M w o D AY/ IO. E ^ //a / 1 a / f , / / / / - cc / / / 1 / S ^ y / 7^ A / - o o.1: GP / /" A • / V / / / a * 'r / A ^ / o I- ,«f / »* /I A / %J x w/ u / ^c CO < 1 / 7 in O / / w £ S <* y V- / / / / < / / / 3 O s ' /\ / / / IO > / / / / / o / y ../. / / / 1 HORSE-POWER OP CONSTANTS. 449 * e = = $ N ~ fc • sa «B s S«1 <*© I £*£ 0 p h N - H N^ 1^ 82 883S ooo oo o COCOCO -#'#^ 53j cno oo* ot^ ■-T C - oi-* ~ ~. x. - co'cori ooo oScnI ooffl coco ^ oi oi oi oi oi (COM CSO OOOlO CO c MOO COO OCN^ ooo oi oi oi oi oi 5?S ooo oc ri o o7 — -" t5 ciico to - Z, X -. O rH OI CN CO -* 0_ t^RO CO "* 03H l-J 05 CO O O CO o o CO o o woe co_ ^ -^ in o o co o 2 ro c co CI O 0_ rH CI CO ^ CO o o ? -2 — c o q sgco5 si- O 01 I- 0 1 -r :- Oi ro :o -f -h i0 §§3 888 23 3 35 co cm o 23 S3? c 2 c coots 883 §s sss CO O O CO o o O O CO o o SS3 §ss r-i^cq OI CO CO rP^uO : OOO HUM H. P. =z-^=- CST+2000 sin 0). Wr=Load in tons. n= Speed in miles per hour, .AT = Wn X .0026§ (K ± 2000 sin 0). K— Resistance in lbs. per ton. K'=Tq E~= Constants of power required to move one ton on level at speeds in table with K= 10. 5/= Constants of additional powee required to raise ONE TON ON grades and at speeds given. 77" X WK/=.'R. P. required on levels alone for speeds given. IZ'X JF = H. P. additional on grades alone for speeds and % given. W(K'H± RO = total H. P. required. F\ami)l«' : Given a motor car, total Aveight 9 tons, to ascend a 7 per cent grade at a speed of six miles per hour. What is the estimated horse- power required, with K= 30 lbs. ? 450 ELECTRIC STREET RAILWAYS. 30 muuipnea Dy y ; ' overcoming the track resistances alone. #'=2.240, which, multiplied by 9, = 20.16. The sum of the two will give the total theoretical, i.e., 24.48 h. p. required. Allowing 50 per cent as the combined efficiency of motors and gearing, to operate this car would require a draft of 48.96 h. p. upon the line. HORSE - POWER OF TRACTIOIf. (Davis.) m .672 .7253 .778L .832 .036 .0891 .1423 .196 .249i .302^ .356 .4091 .462s .516 .5694 .0223 .676 .7291 .782§ .836 .04 •094 .143 .20 •251 .30f .36 •41* .46f .52 •571 •62| .68 n .84 .048 .1014 .1543 .208 .261A .3143 .368 .4211 •4743 .528 .r,s\y • 6>iH .688 .7411 .7943 .848 .051 .103 .16 .211 .263 !S74 .42f .48 •534 •58f .64 .694 •74§ ■84 •854 •061 .12 .173 !'283 .333 .384 .44 .49| .603 •653 .704 .76 .813 •861 .08 .134 .183 .24 i43 .40 •454 .503 .56 •614 .66f 1774- •82| .88 .091 .143 .20 .254 .303 .36 .411 •46f .52 .574 .623 .68 .534 •783 .84 •S94 •104 .16 .21| .264 .32 '424 .48 .533 .584 .64 .693 •744 .80 •853 •904 • 134 .183 .24 .294 .343 .40 .454 .503 .56 .614 .663 '774 .823 .88 •934 .16 •214 •26| .423 48 .534 •583 .64 c| .80 .854 .96 HOR§E-POWER. SPJEEB, AHI> HORIZONTAL ETJFOKT O POVKD8. Miles Per Hour. Mecli. 2 4 6 8 10 15 20 25 30 40 H. P. Feet Per Minute. 176 352 528 704 880 1320 1760 2200 2640 3520 lbs. lbs. lbs. lbs. lbs. lbs. lbs. lbs. lbs. lbs. 2 375.0 187.0 125.0 93.7 75.0 50.0 37.5 30.0 25.0 18.7 4 750.0 375.0 250.0 187.5 150.0 100.0 75.C 60.C 50.C 37.5 6 1125.0 562.0 375.0 281.2 225.0 150.0 112.5 90.C 75.C 56.2 8 1500.0 750.0 500.0 375.0 300.0 200.0 150.0 120.0 100.0 75.0 10 1875.0 937.0 625.0 468.7 375.0 250.0 187.5 150.C 125.C 93.7 15 2812.0 1406.0 937.0 703.1 562.5 375.0 281.2 225.C 187.5 140.6 20 3750.0 1870.0 1250.0 937.2 750.0 500.0 375.0 300.0 250.C 187.5 25 4687.0 2343.0 1562.0 1172.0 937.5 625.0 468.7 375.0 312.5 234.4 30 5625.0 2812.0 1875.0 14(i(i.ll 1125.0 750.0 562.5 450.0 375.0 3S1.2 40 7500.0 3750.0 2500.0 1875.0 1500.0 1000.0 750.C 600.0 500.0 375.0 50 9372.0 4687.0 3125.0 2344.0 1875.0 1250.0 •937.5 750.0 625.0 468.7 POWER REQUIRED FOR TRUCK CARS. 453 POWER REailRED FOR DOVBIE **■» SIXGEE TRUCK CARS. Wattmeter placed on car. (McCulloch.) £$ Double-truck car. Seats 36; weight, 11.75, tons ;• average for entire day Same as above. Average for heaviest trip . . . Single-truck car, no trailer. Seats 28; weight, 8 tons .... Single-truck car. Trail- ers operated 26% of the time. Average for the entire day Si-igle-truck motor and open trailer. Seats, 63 ; weight, 10.5 tons. Average for heaviest trip MORIZO^TAE EFFORT EXERTED Oltf CURVES. Pounds Per Ton. Feet. Length of Wheel Base, Feet. 25 30 40 50 60 70 80 100 3.5 88.6 73.9 55.4 44.3 36.9 31.7 27.7 22" 4 94.0 7S.4 58.8 47.0 39.2 33.6 29.4 23.5 4.5 99.4 82.9 62.2 49.7 41.4 35.5 31.1 24.9 6 115.6 96.4 72.3 57.8 48.2 41.3 36.1 28.9 6.5 121.0 100.9 75.7 60.5 50.4 43.2 37.9 30.3 7 126.4 105.2 79.0 63.2 52.7 45.2 39.5 31.6 Assumed — 3 miles per hour speed on curve, 4 ft. 8J in. gauge. 454 ELECTRIC STREET RAILWAYS. Formula from Molesworth : Let W = weight on wheels in lbs. A' = coefficient, in this case .27. G = gauge of track = 4/ — 8|" =: feet. B = rigid wheel base in feet. /i=r radius of curves in feet. Tractive force or resistance per ton = W X K X (G + £) HORIZONTAL EFFORT ON C^nADEM. Ponuds per Ton. Speed — Miles per Hour. Grade. Per Ct. 2 4 6 8 10 12 14 16 18 20 0 15.03 15.11 15.24 15.42 15.66 15.95 16.29 16.69 17.14 17.64 1 35.03 35.11 35.24 35.42 35.60 35.95 36.29 36.69 37.14 37.64 n 45.03 45.11 45.24 45.42 45.66 45.95 46.29 46.69 47.14 47.64 2 55 03 55.11 55.24 55.42 55.66 55.95 56.29 56.69 57.14 57.64 2i 65.03 G5.ll 65.24 65.42 65.66 65.95 66.26 66.69 67.14 67.64 3 75.03 75.11 75.24 75.42 75.66 75.95 76.29 76.69 77.14 77.64 3* 85.03 85.11 85.24 85.42 85.66 85.95 86.20 86.69 87.14 87.64 4 95.03 95.11 95.24 95.42 95.66 9555 96.29 96.69 97.14 97.64 5 115.03 115.11 115.24 115.42 115.66 115.95 116.20 116.69 117.14 117.64 6 135.03 135.11 135.24 135.42 135.66 135.95 136.29 136.69 137.14 137.64 7 i.-,r,.o:; 155.11 155.24 155.42 155.66 155.! (5 156.29 156.69 157.14 157.64 8 175.02 175.11 175.24 175.42 175.66 175.95 176.29 176.69 177.14 177.64 9 195.03 195.11 195.24 l!»r> .42 195.66 195.95 196.29 196.69 197.14 197.24 19 215.03 215.11 215.24 215.42 215.66 215.95 216.29 216.69 217.14 217.64 APPROXIMATE CURRENT CONSUMPTION PER CAR. Two 35-H.P., S. R. Ct. Motors. Diameter Horizontal Effort - - Pounds. Inches. 100 200 400 600 800 1000 1200 1400 30 33 25.8 26.6 32.8 34.0 44.6 47.0 54.6 57.6 63.8 72.6 67.4 77.6 82.6 88.4 92.0 98.2 Two 30-H.P., S. R. Ct. Motors. Diameter Wheels. Inches. Horizontal Effort — Pounds. 100 250 500 750 1000 1250 1500 2000 2500 3000 30 33 28.6 29.4 38.8 40.0 51.4 54.0 63.0 65.8 73.2 77.0 84.2 88.8 93.4 98.8 111.8 119.2 130.0 138.4 147.6 158.0 AXLE SPEED. 455 AXIS SPEED PER CAR WITH DOUBIE MOTOR EaUIPME^T - RE V§. PER MOTIE. Averag-e of Several Types 25»-H.P. Motors. Diameter Wheels. Inches. Horizontal Effort — Pounds. 100 200 400 600 800 1000 1200 1400 30 33 308 300 253 248 195 189 170 165 153 149 141 136 131 126 122 119 Average of Several Types of 30 H. P. Motors. Diameter Wheels. Horizontal Effort — Pounds. Inches. 100 250 500 750 1000 1250 1500 2000 2500 3000 30 33 282 272 260 252 202 194 173 166 153 148 139 134 130 125 117 113 107 103 100 95 Formula for close approximation of current required to propel a given car. No. tons in train x [( (% grade -J- 1) 20) + (curve resistance per ton)] = Pounds Horizontal Effort. TMO. OF CARS OJ¥ TES MILEi OE TRACK, VARI- OUS SPEEDS AHT» HEADWAYS. Minutes Average Speed in Miles per Hour. Apart or H'dway. 6 7 8 9 10 12 15 20 25 30 1 100 86 75 67 60 50 40 30 24 20 2 50 44 38 33 30 25 20 15 12 10 3 33 29 25 22 20 17 13 10 8 7 4 25 22 19 14 15 13 10 8 6 5 5 20 17 15 13 12 10 8 6 5 4 6 17 14 13 11 10 8 7 5 4 3 7 14 12 11 10 9 7 6 4 3 3 8 13 11 9 8 8 6 5 4 3 3 10 10 9 8 7 6 5 4 3 2 2 15 7 6 5 4 4 3 3 2 2 1 20 5 4 4 3 3 3 2 2 1 1 30 3 3 3 2 2 2 1 1 1 1 Note. — Fractions above one-half are considered whole numbers, and fractions below one-half are neglected. 456 ELECTRIC STREET RAILWAYS. To obtain the number of cars required to operate any length road, divide the number found in the table under the desired average speed and head- way by ten, and multiply by the length of the road in question. Should it PRESSURE IN POUND PER SQUARE FOOT OF CROSS SECTION. 880 NS s? j\ ! ^ NA S \ 0 ^ § \ \ 4 \ \ a ^4 £o i/, A' $ £\ \ K \ ^ ^ t Vn ^ -^ 3>v ? 3 # » \ \ \ K V £/ ^i V£ H -«- ? V ? v \ \ \ <* * _, s ~% Li V \\ \ &\ > ^ > \ ? -f 3 I & \\ w> \< \ " I I o \ %' \ ^ ' \4 0 D * \ m ° M \ $ $ w \ x" , %' \% "3 j $\ \ \ X \ t\ ^ v N \ \ >^ ° \ ui \ & ^ \ \ ? >^ ^~^ ° \ "- \ ft V \ \ \ ^* ^'- \ \ \ ,? ^S \ _ \ >/ 1 \ ^ \ M >, \ \ \ \ Fig. 30. " Effect of Shape of Moving Body on Air Resistance," Crosby's Experiments. be desired to run at different average speeds on various portions of the road, treat each portion as a separate road, and add the results together. To the number of cars thus obtained should be added 20 per cent for reserve for roads under 20 cars. For roads over 20 cars, 10 per cent reserve will be enough. RATING STREET-RAILWAY MOTORS. 457 Formula : — Let n = number of cars required. m — miles of track. S = average speeds in miles per bour. /= interval or headway in minutes. Tben, m x 60 n = r • SX I HEADWAY, IPEED, AUTR TOTAL IVIMHER OF CAMS. Total number of cars on a given lengtb of street on whicb cars are run- ning botb ways = (lengtb of street X 120) -~ (headway in minutes X speed in miles per bour). MIXES PER HOUR I]tf FEET PER HLWEI Aar» per lEcoxn. (Merrill.) Miles Feet Feet Miles Feet Feet per per per per per per Hour. Minute. Second. Hour. Minute. Second. 1 88 1.46 16 1408 23.47 2 176 2.94 17 1496 24.93 3 264 4.4 18 1584 36.4 4 352 5.87 16 1672 27.86 5 440 7.33 20 1760 29.33 6 548 8.8 21 1848 30.8 7 616 10.26 22 1936 32.26 8 704 11.73 23 2024 33.72 9 792 13.2 24 2112 35.2 10 880 14.67 25 2200 36.67 11 968 16.13 26 2288 38.14 12 1056 17.6 27 2376 39.6 13 1144 19.07 28 2464 41.04 14 1232 20.52 29 2552 42.50 15 1320 22 30 2640 44 RATHfG STREET-RAILWAY MOTORS. (Condensed from W. B. Potter in Street Railway Journal.) Rise of temperature after one hour's run under rated full load not to ex- ceed 75° C. ; room being assumed at 25° C. Average load for a day's run should not exceed 30 per cent of its rated full load, Avhich will give a rise of temperature of about 60° C. The above ratings are based on aline potential of 500 volts, but the aver- age performance can generally be increased in proportion to the increase in line voltage ; that is, a motor will do approximately 10 per cent heavier service for the same temperature rise when operated at 550 volts. With electric brakes, motors must have increased capacity, as heating increases 20 to 25 per cent. The 20 per cent increase is on roads having few grades and stops, while the 25 per cent is on hilly roads with frequent stops. Approximate rated horse-power of motors = (total weight of car in tons) X (max. speed in miles per hour on level). _ 458 ELECTRIC STREET RAILWAYS. For equipments with electric brakes, divide by 4 instead of 5. When maximum speed is not known, it may be assumed as twice the schedule Example 1: 20 ton car (loaded) X 50 m. p. h. „^ , ^ „ , ~ - = 200 h. p., or four 50 h. p. motors. In this case, if the line pressure were raised to 600 volts, electric brakes could be used on the equipment by changing the gear ratio so as to have the same maximum speed. Example 3 : 11 ton car (loaded) X 25 m. p. h. p = 55 h. p., or two 30 h. p. motors, These rules indicate minimum capacity under ordinary conditions. Tractive Effort. Tractive effort is dependent on the rate of acceleration, grade, car fric- tion, and air resistance, which latter is ordinarily included in friction. Acceleration is expressed in miles an hour per sec. 1 mile per hour per sec. = 1.466 feet per sec. Excluding car friction, a tractive effort of 92§ lbs. per ton (2000) will produce an acceleration of 1 mile per hour per sec. on a level track, and the rate of acceleration will vary in direct proportion to the amount of tractive effort. On ordinary street cars, tractive effort during acceleration often rises to 200 or 300 lbs. per ton. On elevated or suburban roads the maximum tractive effort is generally 100 to 150 lbs. per ton. For heavy freight work with slow speeds, the trac- tive effort seldom exceeds 30 to 40 lbs. per ton. Grades are commonly expressed in percentage of feet rise in 100 feet of distance, and tractive effort for a grade is the same percentage of the weight to be drawn as the rise is of the length of 100 feet. For instance, the tractive effort for a weight of one ton (2000 lbs.) up a grade of 3 per cent would be 3 per cent of 2000 lbs., or 60 lbs. For the total tractive effort there must be added to this, the effort for overcoming the car, wind, and rolling friction on a level. Maximum tractive efforts from numerous tests are shown in the following table : Tractive effort in lbs. per ton. 15 ton car, up to 25 m, p. h 25 '" " " " " 50 " " " 50 25 " " " " 25 " " " 20 " " " " " 50 " » " 25 100 "train" " 25 " " " 15 Heavy freight train up to 25 m. p. h 6 to 10. The above figures have to be increased for snow and ice on the track. Tractive Coefficient. This coefficient is usually expressed as the ratio between the weight on the driving-wheels and the tractive effort, and varies largely with the con- dition of the rails. In train work, the weight on drivers should be six times the tractive effort. Example:— Required the weight of a locomotive to draw a 100-ton train up a 2 per cent grade. For train. 100 tons x 15 lbs. for friction = 1500 lbs. " " X 40 " " grade = 4000 " 5500 lbs. RATING STREET-RAILWAY MOTORS. 459 Assume a 20-ton locomotive. 20 tons X 15 lbs. for friction = 300 lbs. 20 " X 40 " " grade = 800 " 6600 lbs. 6600 lbs. equals 16.5 per cent of 20 tons, or a tractive coefficient of 16.5 per cent. Starting the train on a 2 per cent grade with acceleration of J m. p. h. per sec. would mean additional tractive effort equivalent to — ^— = 30.8 lbs. per ton. This would add to the requirements as follows : Train 100 tons, for friction and grade as above . . . 5500 lbs. " " " at 30.8 lbs. for acceleration 3080 " Total for train 8580 lbs. Assume 35-ton locomotive with motors on all axles. 35 tons at 15 lbs. for friction 525 lbs. " " " 40 " " grade 1400 " " " " 30.8 for acceleration 1078 " Total tractive effort . . . 11583 lbs. or a tractive coefficient of 16.5 per cent for the 35-ton locomotive. Tests show the following tractive coefficients : Sanded per cent. per cent. Dry rail 28 30 Thoroughly wet rail 20 25 Greasy moist rail . 15 25 "With ice and snow on the track, the coefficient is lower, and the rolling- friction higher. JLverag-e energy. — Approximate capacity of a power station may be assumed as about 100 watt-hours per ton mile of schedule speed for ordinary conditions of city and suburban service. Example : — 15-ton car, 12 miles per hour schedule, k.w. at station = 100 x 15 X 12 = 18 k.w. If stops are a mile or more apart, only 60 to 70 watt-hours may be neces- sary. Frequent stops and high schedule speeds take 120 or more watt-hours. The following table of efficiencies Avill be found convenient in estimating the power required for operation of motor cars, using three-phase trans- mission and direct current motors. The efficiencies would vary somewhat with the load factor, but can be taken as generally applicable. Considering the I.H.P. of the engine as a basis, for the Average efficiency of engine 90 per cent. " " generator 94 " " " " high potential lines .... 95 " " " " substations 90 " " " " direct current lines .... 92 " " " " motors, including losses of control 72 " " Combined efficiency of the motors and series parallel control during period of cutting out the controller may be taken as 63 " " Efficiency of motors after cutting out the controller, depending on size of motors 80 to 85 per cent. 460 ELECTRIC STREET RAILWAYS. © 03 •sptreg; jo ok o 5 o £ t> £ -r 00 00 CO oo 00 oo eo co -k oo fN ^ cS •oi^y; uoi^onpoy; 03 -# ■* ■<* rH UJ •pi30T[ 8 to . o Xin^; ^ poods . % C PI 0, 6 o 6 O 6 a o O 6 o 6 O ° 6 O 03 © •2W . O O o =5 o £ £ £ ■S £ •S s W w w w -2 03 03 V 03 03 03 P. r^ pj PI Pi g H y H H H H «i^ S - O ■a.gg 1* - OS S3 c« o «1 o o o © 03 1=1 a) a s 03 H H H H 3 © 0 V a 0 ops 9 o IM lO 3 CO 00 co § oo t> CM ^) z> LOc ^o >> t>> !>> >» >. o5 % as cS % cti 1 cS c« te 03 ci > Eh rt 5 §3 Ph « w PI 0=i c3 <5 © M co CO CO CO CO CO 1-1 M CO H .2 So o d CO Ph S Ph g r/i d 6 d d in d d t- "O CO W 6 d ^ A •ofi rH CN eo <* ia CO t- OO OS c s s DIMENSIONS, WEIGHTS, ETC., OF MOTORS. 461 o w s o ~ o? lO o t- lO 2 on % CO to o 00 oo CO co ">* CO 2 o CO CO -* ■* * fr- | to to CO 1 in o lO CM rt< tH <# rH ■* -* « *# "# « ON + co 00 ^ ■* ■* « CM "# •* "* « -* *# "* « <* ■* ■tf -* CO -* rh ■<:» ? "* ^* §33 f. ,. r. ro _^ fn ^ . CO CO m o> yj lO >o UJ co -# -# <* £ 3 -f OS £ ce s ># 3 eg 00 ■ -. Tf ■* ■* , >> >v CD . ^ fl Q >> >-> >» ^ >i oS os S3 e3 S3 c3 o3 crt cS C5 O ,e ,c 1* ^ £ 2 ^ ^ 03 c3 fS r = '? cS r^ J3 c3 « g CD « p £ M « ,0 33 CD B « « - « M P^ 5 02 02 M 02 £ 32 02 M M M O. 02 02 02 tt! 02 02 u "eS d S ^ n 3s^ 02 02 «! P 6 © d 1? | 02 O M l £ o £ o ^ *! 13 £ 'A ki O O O O o m 02 ^ £ 02 ~' 462 ELECTRIC STREET RAILWAYS. rt M ci cS c$ cj cS g 2 2 2 2 S£ 3£ 3^ 1 >** S al M w>*» M+l £ "3 ±! 5 ^3 "oS r^ 'S 43 §"« fS^ 3 "£ 3 £ 3 s 3 ,3 3 £ .S" a .S° a .&» a c •ajiAvojn^'BiujY •sap a •O.IIA1 ppi^ •9jm+«ra.TY C- iH OS O CS > © 2 S a s 2 A £ P cS 1=1 bo b£ bo bfl Tj d 0) X ^ - cS r1 Bj cci c3 O A H H s cc 02 X M cc 0Q W 03 S s d 0 CO ^ CO 53 s ej si crt d ft y d 1 © d* 6 5^ X 1 X © X lO 3 X q x ; d g g s S g B. cS oi HORIE-POWER. H. P. per Lb. Applied at Periphery at 100 Rev. per Min. Diameter Wheel. 26" 28" 30" 33" 36" H. P. .02062 .02221 -0238 .02618 .02656 Pounds at Periphery per H. P. at 100 Rev. per Min Diameter Wheel. 26" 28" 30" 33" 36" Lbs. 48.481 45.018 42.017 38.197 35.014 Lbs. 126050.9 X H. P. Diam. x Rev. H. P. = .00000793 X diam. wheel x rev. x lbs. at periphery. H. P. per lb. at periphery at one mile per hour = .002867. Lbs. at periphery per H. P. at one mile per hour = 374.9. iiVote on Emergency Braking* of Cars. In case of emergency, motormen often reverse the motors, which brings the car up with a severe jerk, and is quite apt to strip gears. This is not necessary, and should never be done unless the canopy switch is first thrown off, then when the motors are reversed and the controller handle thrown around to parallel, the motors will act as generators and will bring the car to an easy stop with no harm to the apparatus. In case circuit breakers are used in place of the plain canopy switches, the reversal of the motors will draw so much current from the line that the circuit breakers, if properly adjusted, will open the circuit and the controller can then be used as suggested above. COPPEB WIRE FUSES FOR RAIIWA1 CIRCUITS. B. &S. Gauges. 17 16 15 14 13 12 11 10 9 390 8 450 7 Fuse Point in Amperes. 100 120 140 166 200 235 280 335 520 466 ELECTRIC STREET RAILWAYS. J3 1 8 feet high in Center Vestibules. Vestibules. Vestibules. Vestibules. Total Weight, 27860 " 23750 I CP CD a o <1 •sqq; if 1 1 | 1 1 1 1 II II 1 II 1 •sqT 4800 4800 4800 4800 4800 4800 3000 3200 3200 3200 5200 5200 m •sqT 'Xpoa 4500 4000 5000 4600 4300 4800 4800 5100 5250 7400 6000 6000 5850 6050 6200 6200 w •UOSJ8J.T8d -lit H @ « I |«« I I I s °> 6S Tt< Th TJ< Tt< ^t< Tf Tt< -^ C 12 I'-IS I I 0 O©00 ©© •# ©-*C0©©00 © © 1 CI oi ci ci 01 01 CO 01 CO CO CI Tt< ■* CO CO i-H CN CO co o A go* d-d °r2 fit^ -SB -^ ■— -v~ tn ea cS © cc -o o o ,d -d bo - 5 X co © PES & •s^as d S o ,2 ? So ;- C u O o - o : P Q APPROXIMATE DIMENSIONS. 469 S 8 1 o © o S o 8 o 8 O § 8 8© > H«Wfo3 ,0 c$ -2 * CD § a 2 53.2 3 >.2 "So a 53 •£ £8 CD .g CD "3 -so 5 >"cl « ta oocn ^CN CD Sh o m i-i 50 ^_ a o o a s t- t- *" ^ CO ^ ti" CO l" ^ 3o2» - - - - b- CO & fc- *> «, CO v v CO CO V ;. x CO QCO CO tH rl 1-1 v ?H co CO CO CO lb b 8 £_ CO cq 1-1 CO (M rH v V o N CO & cn CI ■* cq £ ct sy IO 3 ,— *_ ,— *_ .— *_s — ■— , o CD a . o 13 * ■d" • CD S • ^ • cd 53 3"S o CD . O ,0 'o o tin cl o ^2 cf ,0- cS '3 ' s t:£ 0 >s v% ,2 CD 'ol m §c a a a a 3 O -go ^ fe a N O C O C o Q ft ft 4 4 E 470 ELECTRIC STREET RAILWAYS. o 0 o © O •p^oq; noisanoxg; o 1 o o 1 1 o § 8 8 H •puoq; ^j^nipjo 2 CO 00 n 3 3 •saGguassBj; ^noi^i^ 1 8 1 1 o CO t~ s s o "P^ot; uoismoxg; o o | 8 | O lO o CI CO CO -S 2 1 •p'BOT; jfj'BUtpJO 1 o 1 1 lO i- Ah <» ft , •uoisanoxg; in 8 § § Ah |-3 •Supveag s § 8 •q^Suaq; itJ^oj, CJ & co co d _ - G ■* o o o •qipi^i i«*ojd eo t- © tA A T3 -o o o „_ pq e3 <*h a> o a> m w a > H CM aw gco Ht^N CO COt-lOCOCClQt-C©^ OS 3 g« :-f L-"Vi'^>— To — "— ^'trT-uo co x u~ i> co%t< * * CO -*NNHH • • P3 go co^cf^'i-i To'^H— , ^'coT-iooo coin l~- co1'*' ' • CO rf CM CM rH rH . . CO 3 3W f^CO C5 t- CM £- CO L~ LO 00 00 IC t~ CO "* m iw ^CO CO t- CM r-j CO L— LO CO CO LO L- CO Tt* ' * 2 bi> . Rco r-|t-0»COON»t • • gcb « t>N t H CO iH r. J "J l- C X CO L- L- to tP " " i r"i6 co t- ci -t" " :o — : re ^ co i - i- x x TV i -co"'-* • • CD w .Sfco f^CO cooocot'h:o»dp:i:i co x re l- co o ■* io ob gcO r-Wrtbi ~;:i-::i.c ?- /-;i-rsn rclx-Hfrir-feo "°|H CO CJ T»H -# i-l CO CO CD CO rH oi o l-. o * * CO ■* CO CM CM CM i-l . . « .^co ^co O 05 CO •* r- CO CO p CO o O O CO O LC CO o UO 00 CO CO ieuo ^co 00 05 ■* -* ^H -fi S5 O CO r-i CI IC CO. C5 O 01 UO o CO f^CO CO G5 ■* ■* rH CO CO O CO i-; CI O O ©5 LO CI LO O ' " CO ■* CO CM CM CM rH . . W .^00 ^co 3310 . - .g . .g>g^O>^^C005C5 w ^co cr"- ^? * ' *J0 ' ' L- ^O Cj rnP^L- CO co ao°05 CO g$g | gcoco^co^co^nooicocooc^ ; ; CO fr CO CO' CO -t — :o -+■ p "O C^ -~- »o O. COOCOCOHI " gco co co' co ^r ^h co ^t" p cp — re ih cr. x- co — co •# | y Chicag-o City Railway in 'Street Railway Journal," »ec, 1SOS. Power-Station. Engines, 8 per cent; Boilers, 8 per cent; Gene- rators, 3 per cent ; Buildings. 5 per cent. CaMe Machinery. Cable machinery, 10 per cent ; Cables, 175 per cent. RoadUed. Rails, 5.5 per cent ; Ties, 7 per cent. Paving-. Granite, 5 per cent ; Cedar blocks, 16 per cent ; Brick, 7 per cent ; Asphalt, 7 per cent ; Macadam, 6 per cent. Car*. Car bodies. 7 per cent ; Trucks. 8 per cent. Rolling- Stock. Armatures. 33 per cent ; Fields, 12 per cent ; Gear cases, 20 per cent ; Controllers, 4 per cent ; Com- mutators, 33 per cent. Wiring and other electrical equipment, 8 per cent. Eine Equipment. Iron poles, 4 per cent ; Wood poles, 8 per cent ; In- sulation, 12 per cent; Trolley-wire, 5 percent; Trolley insulation, 7 per cent ; Bonding, 8 per cent. All based upon renewals and per cent of wear. TRACK RETURN CIRCUIT. 499 CAR HEATIIG BY ELECTRICITY. Test on Atlantic Avenue Railway, Brooklyn. Cars. Temperature F. Watts Consumed. Doors. Windows. Contents, Cu. ft. Outside. Average in car. 2 12 850* 28 55 2295 2 12 850i 7 39 2325 2 12 808£ 28 49 2180 2 12 913£ 35 52 2745 4 16 1012 7 46 3038 4 16 1012 28 54 3160 TRACK RETITRI CIRCUIT. It goes without saying that the return circuit, however made, whether through track alone or in connection with return feeders, should he thehest possible under the circumstances. Few of the older roads still retain the bonds and returns formerly considered ample and good enough. Electrolysis and loss of power have compelled many companies to replace bonds and return circuits by much better types. The British Board of Trade paid especial attention to the return circuit in the rules gotten out by them (see page m67), and many American railroads would have been much in pocket to-day if such rules had been promulgated in the United States at the beginning of the trolley development. With few exceptions the practice of engineers has been to connect the rail joints by bomls, both rails of a track together at intervals, and both tracks of a double-track road together. To this has sometimes been added track return wires laid between the rails, and in other cases return feeders from sections of track have been run to the power-house on pole lines. The writer favors the full connection return with frequent insulated overhead return feeders where there may be danger from electrolysis of water and gas pipes ; in fact, ample return circuit has been proved time and again to be the only preventive of that trouble. Careful and continuous attention should be given to bonds from the moment cars are started on a line. Dr. Bell gives the following ratios of track return circuit to overhead sys- tem as being average conditions. Let Ry — resistance of track return circuit, and R =. resistance of overhead system ; Then Ry = .1 to .2R. Exceedingly good track and very light load. R, = .2 to .3R. Good track and moderate load. Ry = A to .6R. Fair track, moderate load. Ry = .2 to .3R. Exceptional track and large system. Ry — .3 to J1R. Good track, large system. Ry = .7 to I.O.K. Poor track, large system. In exceptional cases track resistance may exceed that of overhead system. It is sometimes assumed that R, = .25R, but this is rather better than usual. Under ordinary conditions R, = AR is nearer correct. If formula for copper circuit = cm. = then for Ry = 4R, the constant 11 should be increased to between 14 and 15 in order that copper drop may bear correct proportion to that of the ground return. Some forms of rail bond are shown on the following pages ; most of these are applied to the rail by pressure or hammer riveting, but some of our bet- ter road managements are now soldering all bonds by strong heat. 500 ELECTRIC STREET RAILWAYS. A few roads still use wire secured in the web of the rail by steel channel pins, which is about the easiest and cheapest, as well as the least efficient form of bonding. As copper bonds have a high value as junk, many of the long type are now stolen from suburban railways, and the tendency is strongly in favor of the concealed or protected bond which is so designed as to go in the space back of the first piate against the web. For a time these protected bonds were made very short, and no very great attention paid to their flexibility but experience has proved that no bond of less tnan eight or nine inches' will last well, no matter how flexible. Solid conductor bonds are only available for the outside of fish plates, and not less than two feet in length. In applv- ing tin copper bonds to the rails, it is necessary to apply them immediately after drilling the web, unless holes are made at the rail mill and carefully oiled, in which case the oil should be very carefully lemoved before apply- ing the bond. Bonds are best applied by a medium using heavy pressure, either by screw or hydraulic pressure, rather than by hammer riveting. On many of the systems, in large cities, rails are made practically continu- ous now by use of electrically welded joints or cast weld joints. In the electrically welded system a piece of wire about nine inches long, two inches wide and an inch thick is welded across the joint on each side of the rail web by means of a heavy current of electricity applied by special machinery, taking its power from the trolley system. After the straps are welded in position, the tops of the rail ends are carefully ground to an even surface. Contrary to the ordinary ideas of the results of expansion and contraction, but little trouble is experienced by broken joints or bent rails, and in most places, where the method is in use, it has been quite successful. The system is controlled by the Johnson Steel Co. of Cleveland, Ohio. The cast weld joint is simply a bunch of cast iron cast about the joint after it has been cleaned and prepared by placing a mold under it. The Falk Company of Milwaukee makes a specialty of bonding street railway systems in this manner, and the results seem to have been good. Several forms of plastic bond have been devised and used to some extent. They all consist of some form of plastic metal held in position between the fish plate and the rail Aveb, the surfaces of both being treated chemically or otherwise, so as to remove scale and oxide so that the plastic material may be applied directly against the wire. Solid Bonds. — This type is simply a heavy copper bar, say No. 0000 B. & S. gauge, with the ends compressed to form a collar, and bent to fit the holes in the rails, and their hammer riveted to place. A good example is that made by Messrs. Benedict and Burnham, and shown in Fig. 50 ; the first three cuts showing a side view of the bonds and rails, the next three cuts showing cross sections of rails with bonds applied. Benedict and Burnham Solid One-Piece Rail-Bond. Fig 50. Short thick Bond applied to " Tram " of Girder Rail, allowing constant inspection. O O O O O O I Fig. 51. Short thick Bond applied to Base of either Girder or T Rail. TRACK RETURN CIRCUIT. 501 O O O O O O Fig. 52. Solid long Bond clearing the Fish-plate in either Girder or T Rail. .Protected Bonds. — Good examples of these are exhibited in Figs. 53, 54, 55, 56, which show the type of protected bond sold by the Mayer & "Eriglund Co. of Philadelphia. They are applied by a special hydraulic press, and many variations of form are made to fit special cases. FlG. 53. Showing 7-inch Girder Rail, bonded with one Bond. Fig. 54. Showing 7-inch Girder Rail, double bonded with two Bonds, one on each side of rail. Electrical connection of 425,000 cm. Fig. 55. Showing 9-inch Grooved Girder Rail double bonded with two Bonds, one in each chamber and both on same side of rail. Electrical connection of 425,000 cm. 502 ELECTRIC STREET RAILWAYS. Fig. 56. Showing 9-inch Girder Rail quadruple bonded with four Bonds, two in each chamber, on both sides of rail. Electrical connection of 850,000 cm. Another form of this type of bond is that shown in Fig. 57, as made by the Forest City Electric Co. of Cleveland. ^TrY ^ Fig. 57. Still another form of concealed bond is shown in Fig. 58, and made by I. M. Atkinson & Co., Chicago. Rail Bond of J. WE. Atkinson & Company, Chicag-o. 1P§ Fig. 58. Applied either single or double under fish-plate. TRACK RETURN CIRCUIT. 503 In some types of bond the plug has a hole through it, and after placing it in the hole in the web of the rail a steel mandrel is driven through to expand the copper outwardly to fill the hole. forest City Electric Company Short Bond. This bond is applied underneath the fish-plate, and secured by a special tool. , 504 ELECTRIC STREET RAILWAYS. In numerous tests of rail bonds, Mr. W. C. Burton, of the J. G. White Co., says it was found that where the copper plug was well pressed home the resistance of the joint between rail and bond did not exceed that of three- eighths inch of the bond itself, even after a year or more of use : and that short bonds, especially those that could be covered by the 'fish-plate made rail-joint resistance a very small percentage of the total track resist- ance. He had never found tinned copper any better than the bare metal, and when pressed tight had not noticed any effect whatever from local action. Table Showing- Sectional Areas of Various Rails, tSae Equivalents in Circular Mils, and tlie Equivalent Cir- cular Jlils of Copper living- Same Conductivity. (Figures on rails are for one side of a single track.) Weight Per Yard. Area of Single Rail. Sq. in. Circular Mils of Single Rail. Equivalent Circular Mils of Copper for Same Conductivity. 45 50 56 60 65 70 80 4.4095 4.8994 5.4874 5.8794 6.3693 6.8592 7.8392 5,614,400 6,238,200 6,986,700 7,485,800 8,109,600 8.733,400 9,981,100 997,200 1,108,000 1,241,000 1,329,500 1,440,400 1,551,200 1,772,800 Area in Cir. Mils : _ 1,000,000 x wgt. per yard 10.2052 X .7854 Area in cir. mils Equivalent Cir. Mils of Copper Mr. W. C. Burton, of J. G-. White Co., found a very considerable difference in rail resistivity, and numerous tests of modern steel rails showed the spe- cific resistance to be from six to twelve times that of copper, where six has been the factor frequently used. In his own practice Mr. Burton uses a factor dependent upon the chemical properties and the physical treatment of the rail in the rolling-mill. BOABD ©E TRADE REGULATIONS. For Great Rritain. Regulations prescribed bv the Board of Trade under the provisions of Section of the Tramways Act, 189—, for regulating the emplov- ment of insulated returns, or of uninsulated metallic returns of low resist- ance ; for preventing fusion or injurious electrolytic action of or on o-as or water pipes, or other metallic pipes, structures, or substances ; and for min- imizing, as far as is reasonably practicable, injurious interference with the electric wires, lines, and apparatus of parties other than the company, and the currents therein, whether such lines do or do not use the earth as a return. Definitions. In the following regulations : — The expression " energy " means electrical energy. The expression " generator " means the dynamo or dynamos or other electrical apparatus used for the generation of energy. BOARD OF TRADE REGULATIONS. 505 The expression "motor" means any electric motor carried on a car and used for the conversion of energy. The expression "pipe" means any gas or water pipe, or other metallic pipe, structure, or substance. The expression "wire" means any wire apparatus used for telegraphic, telephonic, electrical signaling, or other similar purposes. The expression "current" means an electric current exceeding one- thousandth part of one ampere. The expression " the company " has the same meaning or meanings as in the Tramways Act. 189—. Reg-iilations. 1. Any dynamo used as a generator shall be of such pattern and con- struction as to be capable of producing a continuous current without appre- ciable pulsation. 2. One of the two conductors used for transmitting energy from the gen- erator to the motors shall be in every case insulated from earth, and is hereinafter referred to as the " line"; the other may be insulated through- out or may be insulated in such parts and to sucb extent as is provided in the' following regulations, and is hereinafter referred to as the " return." . 3 Where any rails on which cars run, or any conductors laid between or within three feet of such rails, form any part of a return, such part may be uninsulated. All other returns or parts of a return shall be insulated, unless of such sectional area as will reduce the difference of potential be- tween the ends of the uninsulated portion of the return below the limit laid down in Regulation 7. .„..'.« f * * 4 When any uninsulated conductor laid between or within three feet of the rails forms any part of a return, it shall be electrically connected to the rails at distances apart not exceeding 100 feet, by means of copper strips having a sectional area of at least one-sixteenth of a square inch, or by other means of equal conductivity. 5. When any part of a return is uninsulated it shall be connected with the' negative terminal of the generator, and in such case the negative termi- nal of the generator shall also be directly connected, through the current- indicator hereinafter mentioned, to two separate earth connections, which shall be placed not less than twenty yards apart. Provided that in place of such two earth connections the company may make one connection to a main for water supply of not less than three inches internal diameter, with the consent of the owner thereof, and of the person supplying the water ; and provided that where, from the nature of the soil or for other reasons, the company can show to the satisfaction of an inspecting officer of the Board of Trade that the earth connections herein specified cannot be constructed and maintained without undue expense, the provisions of this regulation shall not apply. The earth connections referred to in this regulation shall be constructed, laid, and maintained so as to secure electrical contact with the general mass of earth, and so that an electromotive force not exceeding four volts shall suffice to produce a current of at least two amperes from one earth connection to the other through the earth, and a test shall be made at least once in every month to ascertain whether this requirement is complied with. No portion of either earth connection shall be placed within six feet of any pipe, except a main for water supply of not less than three inches in- ternal diameter, which is metallically connected to the earth connections with the consents hereinbefore specified. 6. When the return is partly or entirely uninsulated, the company shall, in the construction and maintenance of the tramway (a), so separate the uninsulated return from the general mass of earth, and from any pipe in the vicinity ; (b) so connect together tbe several lengths of the rails ; (c) adopt such means for reducing tbe difference produced by the current be- tween the potential of the uninsulated return at any one point and the po- tential of the uninsulated return at any other point ; and (cl) so maintain the efficiency of tbe earth connections specified in the preceding regulations as to fulfill the following conditions, viz.: 506 ELECTRIC STREET RAILWAYS. (1.) That the current passing from the earth connections through the in- dicator to the generator shall not at any time exceed either two amperes per mile of single tramway line, or 5 per cent of the total current output of the station. (2) That if at any time and at any place a test be made by connecting a galvanometer or other current indicator to the uninsulated return, and to any pipe in the vicinity, it shall always be possible to reverse the direction of any current indicated by interposing a battery of three Leclanche cells connected in series, if the direction of the current is from the return to the pipe, or by interposing one Leclanche cell, if the direction of the current is from the pipe to the return. In order to provide a continuous indication that the condition (1) is com- plied with, the company shall place in a conspicuous position a suitable, properly connected, and correctly marked current indicator, and shall keep it connected during the whole time that the line is charged. The owner of any such pipe may require the company to permit him at reasonable times and intervals to ascertain by test that the conditions specified in (2) are complied with as regards his pipe. 7. When the return is partly or entirely uninsulated, a continuous record shall be kept by the company of the difference of potential during the work- ing of the tramway between the points of the uninsulated return furthest from and nearest to the generating station. If at any time such difference ) of potential exceeds the limit of seven volts, the company shall take imme- diate steps to reduce it below that limit. 8. Every electrical connection with any pipe shall be so arranged as to admit of easy examination, and shall be tested by the company at least once in every three months. 9. Every line and every insulated return or part of a return, except any feeder, shall be constructed in sections not exceeding one half of a mile iii length, and means shall be provided for insulating each such section for purposes of testing. 10. The insulation of the line and of the return when insulated, and of all feeders and other conductors, shall be so maintained that the leakage cur- rent shall not exceed one-hundredth of an ampere per mile of tramway. The leakage current shall be ascertained daily, before or after the hours of running, when the line is fully charged. If at any time it should be found that the leakage current exceeds one-half of an ampere per mile of tram- way, the leak shall be localized and removed as soon as practicable, and the running of the cars shall be stopped unless the leak is localized and removed within twenty-four hours. Provided, that where both line and return are placed within a conduit this regulation shall not apply. 11. The insulation resistance of all continuously insulated cables used for lines, for insulated returns, for feeders, or for other purposes, and laid be- low the surface of the ground, shall not be permitted to fall below the equivalent of 10 megohms for a length of one mile. A test of the insulation resistance of all such cables shall be made at least once in each month. 12. Where in any case in any part of the tramway the line is erected over- head and the return is laid on or under the ground, and where any wires have been erected or laid before the construction of the tramway, in the same or nearly the same direction as such part of the tramway, the com- pany shall, if required to do so by the owners of such wires or any of them, permit such owners to insert and maintain in the company's line one or more induction coils, or other apparatus approved by the company for the purpose of preventing disturbance by electric induction. In any case m which the company withhold their approval of any such apparatus, the owners may appeal to the Board of Trade, who may, if they thmk fit, dis- dispense with such approval. 13. Any insulated return shall be placed parallel to, and at a distance not exceeding three feet from, the line, when the line and return are both erected overhead, or 18 inches when they are both laid underground. 14. In the disposition, connections, and working of feeders, the company shall take all reasonable precautions to avoid injurious interference with any existing wires. 15. The company shall so construct and maintain their systems as to secure good contact between the motors, and the line and return respec- tively. BOARD OF TRADE REGULATIONS. 507 16. The company shall adopt the best means available to prevent the oc- urrence of undue sparking at the rubbing or rolling contacts in any place, and in the construction and use of their generator and motors. 17. In working the cars the current shall be varied as required by means of a rheostat containing at least twenty sections, or by some other equally ifftcient method of gradually varying resistance. 18. Where the line or return or both are laid in a conduit, the following Conditions shall be complied with in the construction and maintenance of iuch conduit : [a) The conduit shall be so constructed as to admit of easy examination of, and access to, the conductors contained therein, and their insulators and supports. 'b) It shall be so constructed as to be readily cleared of accumulation of dust or other debris, and no such accumulation shall be permitted to remain. (c) It shall be laid to such falls, and so connected to sumps or other means of drainage as to automatically clear itself of water without danger of the water reaching the level of the conductors. (d) If the conduit is formed of metal, all separate lengths shall be so jointed as to secure efficient metallic continuity for the passage of electric currents. Where, the rails are used to form any part of the return, they shall be electrically connected to the conduit by means of cop- per strips having a sectional area of at least one-sixteenth of a square inch, or other means of equal conductivity, at distances apart not ex- ceeding 100 feet. Where the return is wholly insulated and contained within the conduit, the latter shall be connected to earth at the gen- erating station through a high resistance galvanometer, suitable for the indication of any or partial contact of either the line or the return with the conduit. (e) If the conduit is formed of any non-metallic material not being of high insulating quality and impervious to moisture throughout, and is placed within six feet of any pipe, a non-conducting screen shall be interposed between the conduit and the pipe, of such material and dimensions as shall provide that no current can pass between them without traversing at least six feet of earth ; or the circuit itself shall in such case be lined with bitumen or other non-conducting damp- resisting material in all cases where it is placed within six feet of any pipe. (/) The leakage ciirrent shall be ascertained daily before or after the hours of running, when the line is fully charged, and if at any time it shall be found to exceed half an ampere per mile of tramway, the leak shall be localized and removed as soon as practicable, and the running of the cars shall be stopped unless the leak is localized and removed within 24 hours. 19. The company shall, so far as may be applicable to their system of working, keep records as specified below.' These records shall, if and when required, be forwarded for the information of the Board of Trade. Daily Records. Number of cars running. Maximum working current. Maximum working pressure. Maximum current from earth connections (vide Regulation 6 (1) ). Leakage current (vide Regulation 10 and 18/.). Fall of potential in return (vide Regulation 7). Monthly Records. Condition of earth connections (vide Regulation 5). Insulation resistance of insulated cables (vide Regulation 11). Quarterly Records. Conductance of joints to pipes (vide Regulation 8). 508 ELECTRIC STREET RAILWAYS. Occasional Records. Any tests made under provisions of Regulation 6 (2) ). Localization and removal of leakage, stating time occupied. Particulars of any abnormal occurrence affecting the electric working of the tramway. Signed by order of the Board of Trade this day of 189 Assistant Secretary, Board of Trade. OVERHEAD SYSTEM FOR EIEC1RIC STREET RAILROAD!. 1. Ladder system, shown in the following cut, formerly somewhat used on small roads, where both feeder and trolley wire of the same size would carry the load. Feeder in this case is simply an enlargement of the trolley wire, and as used might have better been one large trolley wire. TROLLEY WIRE 2. A modification of the above system is the folloAving. In this second system the trolley wire is cut into sections, and while losing the extra con- ductivity of the continuous trolley, by placing fuse and switch at the junc- tion of each sub-feeder Avith the main feeder, each such section may be cut out in case of trouble without depriving the remainder of the system of current. FIG. 61. Both above systems are now somewhat out of date, although occasion- ally used on the smaller roads. 3. The system shown in the following cut is more of a real feeding system than either of the previous two. The trolley wire is connected directly to the dynamo, but is also fed at various points, as at a, b, c, by larger wires tapped into it. A load at d would thus receive current from both feeders b and c, and the pressure can be more evenly maintained than by either of the previous methods. By making the trolley wire of larger cross-section than is usual in the previous systems, it is possible to have fewer sections and yet main- tain a fairly even voltage. OVERHEAD SYSTEM. 509 4. An obvious modification of the above is shown in the following cut. In this system the trolley wire is again divided into sections, but each sec- tion is supplied from its own separate feeder, the size of which may be so calculated as to keep a very even pressure at all points on the line, especially so if the trolley wire be not too small and the sections not too long. It is of course, subject to the objection that the sections receive no help from the remainder of the circuit, but has the advantage that each section maybe, (controlled by switch and circuit-breaker at the station, and if at any part of [the system, as at d, there is a heavy grade or a heavy massing of cars, cross ^connection can be made to the feeder c, either by switch or by permanent tie. Another method of tying that has been used in some localities is that of connecting the ends of trolley sections together with a small copper wire, say No. 12 B". & S., and thus getting part current both ways ; and in case of heavy overload oi short circuit on a section the tie-wires will burn off, leaving all other sections free as before. This method is said to be of consider- able advantage. 5. The following cut shows a combination of the previous methods, such as results from experience in operating larger systems of roads. The principal feeder C is tapped at intervals to feed the short and long sections, and in order to maintain even voltage at its distant end, is reinforced at d and e by the feeders E and F, while the still farther distant trolley-line sections are fed by the long feeders G and H, which can be joined as at/, if the circumstances call for it. _ As mentioned above, this method is the result of actual experience on a line after it has been run, and the loads have developed the points where current is most needed. While systems of overhead lines are always laid out with more or less care, traffic often takes the most erratic changes in direc- tion, and changes its call for load to such an extent that feeders often have to be run to new points, sections have to be joined or new divisions made, or feeders have to be tied ; and this cut shows the general result of such actual experience. As a general thing it is not good practice to cut the trolley into any more sections than necessary for safety ; and even then a separable line, that is, one that can be cut into sections by switches, is bet- ter than separate sections. 510 ELECTRIC STREET RAILWAYS. 6. For long roads the system shown in the following cut may be used with advantage, as, with heavy trolley wire such as should always be used on long lines, the trolley wire can be reinforced by the feeders as shown, Fig. 65. as to maintain a fairly constant pressure, and advantage be taken of all the conductivity of the system. On double-track roads all the trolley system should be united and at frequent intervals, so that advantage may be taken of the full conductivity installed. 7. A system sometimes used on small single-track lines, A\rhere feeders are not entirely necessary, but a single trolley Avire may be too small, is to run two trolley Avires side by side, and at all sidings the Avire nearest the siding is run around it, and the cars can pass and the trolleys follow each its own wire without troublesome switches. CAECUI^TIMCJ THE COlfDITCTIH-G- SYSTEM. Dr. Louis Bell gives the following steps as the best to be followed in entering upon the calculation of the conducting system of a trolley road : 1. Extent of lines. 2. Average load on each line. 3. Center of distribution. 4. Maximum loads. 5. Trolley wire and track return. 6. General feeding system. 7. Reinforcement at special points. It must be said at once that experience, skill, and good judgment are far better than any amount of theory in laying out the conducting system of any road. Much depends upon the character of the load factor, i. e., the ratio of average to maximum out-put ; and this, varying from .3 to .6, can only be guessed at by a study of the particular locality, the nature of its industries and working people, the shape of the territory, and the nature of the sur- rounding country. CALCULATING CONDUCTING SYSTEM. 511 1. Map out the track to scale, noting all distances carefully, and dot in any contemplated extensions, so that adequate provision may be made in the conducting system for them. Note all grades, giving their length, gra- dient, and direction. Divide the road into sections such as may best sug- gest themselves by reason of the local requirements, but such as will make the service under ordinary conditions fairly constant. 2 The average load on each section will depend, of course, upon the number of cars, and the number of cars upon the traffic. This can only be arrived at by a comparison with similar localities already equipped with street railway and even then considerable experience and keen judgment of the general nature of the towns are necessary in arriving at anything like a correct result. ' . 3 If the road has been correctly laid out as to sections, the load on each will he uniform and may be considered as concentrated at a point midway in each section. Now, if a street railway were to be laid down on a perfectly level plain where the cost of real estate was the same at all points, and wires could he run directly to the points best suited ; then it would only be necessary to locate the center of gravity of the entire system, and build the power station at that point, sending out feeders to the center of each sec- tion. Unfortunately for theory, such is never the case ; and cost of real estate, availability of the same, convenience of fuel, water, and supplies will govern very largely the selection of a location for the power-house. Evenwhen all the above points necessitate the placing of the power-house far from the center of gravity of a system, it may be possible to use such center as the distributing point for feeder systems, and even where this is not possible, it is well to keep in mind the center, and arrange the distribut- ing system as nearly as possible to fit it. All this relates, however, to preliminary determinations for the system as determined at the time, and in large systems will invariably be supplemented by feeders, run to such points as the nature of the traffic demands. A base- ball field newly located at some point on the line not known to the engineer previous to the installation, will require reinforcement of that particular section ; and often after a road has been running for some time, the entire location of traffic changes, due to change in facilities, and feeder systems then have to be changed to meet the new conditions, so that after all, loca- tion of the center of distribution depends largely on judgment. 4. The predetermination of the maximum or average load is another mat- ter for experienced guessing, as it will depend altogether upon the nature of the traffic, how many people patronize the line, and how often the cars are run. If the weight of the car and its load be known it is an easy matter to de- termine the power necessary to propel it ; and tables will be found in this section showing the tractive effort necessary, and all other data for such determination . Bell gives the following formula for the horse-power necessary at the wheel of a ear. Let P = total horse-power. W= weight of car and load in tons. .43 — h.p. per ton required at wheel at 20 lbs. per ton for a speed of 8 miles per hour. G = per cent grade. ?hen P = W(.43 + A3G). This applies to straight tracks only, and at a speed of 8 miles per hour, which is often exceeded. The same authority also states that allowing an efficiency between trolley and car- wheel of 66f per cent, and voltage at the car of 500, 1-i amperes per ton plus 1J amperes per ton for each per cent of grade will be approximately correct. This means an average of about 15 amperes per car, throughout the day, for the ordinary car and road. Long double-truck cars will take nearer 25 amperes, and in the writer's judgment this last is a good average to use for all traffic on ordinary street railways. The maximum current will rise to four or five times the average where but one or two cars are in use ; will easily be three times the average on 512 ELECTRIC STREET RAILWAYS. roads of medium size, while on very large systems it may not be more than double the average. If speeds are maintained on heavy grades the maxi- mum is still further liable to increase. Another point to be considered in connection with maximum load, is the location, not only of heavy grades, but of parks, ball-grounds, athletic fields, cemeteries, and other such places for large gatherings of people that are liable to call for heavy massing of cars, many of which must be started practically at the same time, and tor which extra feeder, and in some cases extra trolley capacity, must be provided. Having determined the average current per section of track, the maximum for the same, and the extraordinary maximum for ends, park locations etc as well as the distances, all data are obtained necessary lor the determinal tion of sizes of feeders. 5. The selection of the proper size of trolley is somewhat empirical, but the size may be g< .yerned by the amount of current that is to be carried. It is obvious tliat with given conditions the larger the trolley wire the fewer feeders will be necessary, and yet with few feeders the voltage is liable to vary considerably. In ordinary practice of to-day No. 0 B. & S. and No. 00 B. & S. gauge, hard-drawn copper are the sizes mostly in use, the latter on those roads having heavier traffic or liable to massing of cars at certain localities. On suburban roads using two trolley wires in place of feeders, 0000 B. & S. gauge will probably be best. Track return circuit has been treated fully in a previous chapter; and all that is needed to say here is, that some skill in judgment is necessary in settling on the value of the particular track return that may be under con- sideration, in order to determine the value of the constant to be used in the formula for computing the size of wire or overhead circuit. In ordinary good practice this value may be taken as 13, 14, or 15, according as the bond- ing and rail dimensions are of good type and large. 6. Feeder-points should, in a general way, be so located as to allow no drop in a section of trolley wire exceeding 5 per cent or 25 volts under nor- mal load. This drop is easily determined by the regular formula : Let D = distance from feeding point to end of the trolley section, cm. — circular mils of the trolley, E = drop in volts, 13 = constant for circuit in connection with a well bonded heavy track, /= current required per car, usually taken as 15 amperes under running conditions, but more safely taken as 25 amperes. Then _ cm. E —'~l3~T » and if the trolley wire selected be No. 00 B. & S. cm. = 133,600, and as the • •-, , • +1 * ii • • or 1+ ;n 133,000 X 25 maximum drop permissible in the trolley wire is 25 volts D = g — — longest section of trolley wire for one car, or 10,231 feet. If two cars are bunched at the end of the section the drop will be twice as great, or the length of section can be but 5,115 ft.; for 3 cars the lengtb Avill be 3,410 ft.; for 4 cars the length will be 2,558 ft.; and for five cars the length will be 2,046 feet. The above calculation will be correct for level roads and where the load is well distributed ; but the trolley-wire sections must necessarily be shortened up for grades or at such points in the line as heavy massing of cars is liable to take place, as at ball-parks, etc., where people all want to get home at once, and all available cars are started from that point. In such cases it will probably be safe to allow 50 amperes per car for the section of trolley wire on which the park is located, and the result is then D — ' A = 5,115 ft. for one car, and for n cars the greatest length of section would be 5,115 -j- n. CALCULATING CONDUCTING SYSTEM. 513 Having calculated all the points on the trolley line at which it should be fed, it remains to calculate the size of feeder for the purpose. As to the allowable drop in feeders, it is not well to have over 100 volts total drop at the car and 75 volts total drop is better under maximum load, as low voltage at the motors tends to over-heat them to a dangerous degree. Much of the regular drop can be overcome by over-compounding the gene- rators for a rise of potential of about 50 volts. It is decidedly better practice to make feeder determinations based on the maximum load, as the average load will easily care for itself, but during times of extraordinary crowds, or snow-storms, if the line has not been cal- culated for such heavy loads, all the motors will heat, and much trouble is liable all along the line. The writer considers 75 volts drop in feeders under maximum load condi- tions a safe basis, together with 35 amperes per car for all those liable to be on the section at once. Over-compounding will make up for 50 volts of the drop at the motors at times of heaviest distributed load, so there will be no danger. Feeder calculation will then be . . , 13 x D X 35 » cars cm. of feeder = =- to It is quite obvious that the current-carrying capacity of the feeder must be taken into consideration, in spite of any determination of drop ; and this can be found in the chapter on Conductors. Sizes of conductors are also governed to some extent by convenience in handling, and it is found that 1,000,000 c. m. is about the largest that can be safely handled for under- ground work, while anything larger than 500,000 c. m. for overhead circuits is found to be difficult to handle. 7. In cases where it is necessary to feed the trolley wire in short sections, in order to reinforce the trolley wire for heavy grades, sub-feeders are often used, the main feeder being tapped into its center, or at such point in its length as will give the best distribution, as shown in the following cut. SUB-FEEDER Fig. 66. For lines having parks at the end, or in fact for any such section, it is perhaps best to run a feeder nearly to the end of the section, then take the trolley line to the feeder at various points comparatively short distances apart, as shown in the following cut; and if the loads are at times especially heavy, the next feeder can be made to assist by cross-connecting, as at d. a BALL PARK AT END OF LINE TRACK RETURN CIRCUIT Fig. 67. In this connection it must be remembered that heavy loads from parks, as well as on grades, do not often come at the same time as heavy loads on other sections, and therefore that the over-compounding may not be but a 514 ELECTRIC STREET RAILWAYS. part of the full-load rise, and it is best under the circumstances to calculate the sizes of such feeders for a smaller drop, say 50 volts maximum instead of 75. In general it may be said that it is now tbe usual practice to use a few standard sizes of feeder wire, such as 100,000 cm., 200,000, 250,000, 500,000, and so connect them as to produce the required results, rather than to carry a large number of various sizes of wire in stock. In fact, this same practice is now carried out in large lighting installations as well, and in those constant pressure is much more needed than in railway circuits. Special HKethods of Distribution. For cases requiring excessively large currents carried a considerable dis- tance, or for ordinary currents carried excessive distances, it is usually economy to adopt some special method ; and among those most commonly mentioned are : the three- wire system, the booster system, the substation system. Three-Wire System. This system, patented some time ago by the I General Electric Company, has been seldom used, and where used has met with little success, owing to the difficulty met in keeping the system bal- anced. The diagram below will assist in making the method plain. Two 500-volt generators are used, as in the lighting system of the same type. The rail return is used as the neutral conductor; and if both trolley wires could be made to carry the same loads, and to remain balanced, then the rail return 7 ° 1 J ; i THREE WIRE SYSTEM Fig. 68. Three-Wire System. would carry no current, and no trouble would occur from electrolysis. The overhead conductors could also be very much smaller, as currents would be halved, and the full voltage would be practically 1000. The Booster System.— Where current must be conveyed a long distance, say five to ten miles, and be delivered at 500 volts, it is hardly good economy to install copper enough to prevent the drop; and if the volt- age of the generator be raised sufficiently to deliver the required voltage, the variations due to change of load will be prohibitive. In such cases a "booster" can be connected in series with the feeder, and automatically keep the pressure at the required point, as long as the generator delivers the normal pressure. The "booster" is nothing more than a series-wound dvnamo, connected so that all the current of the feeder to which it is attached flows through both field and armature coils, and the voltage produced at the armature terminals is added to that of the line, and as the voltage so produced is in proportion to the current flowing, it will be seen that the pressure will rise i^^HH CALCULATING CONDUCTING SYSTEM. 515 and fall with the current. This is now used in many instances, both in lighting and for railway feeders, and especially in feeding storage batteries, and has met with entire success. The following cut is a diagram of the connections. 6 TO 10 MILES I MOTOR \ TO DRIVE generatofl\eoostep. OVERHEAD RETURN BOOSTER SYSTEM Beturn Feeder Booster. —Major Cardew, Electrical Engineer for the Board of Trade, some time ago devised a method of overcoming exces- sive drop in track return circuits by the use of insulated return feeders, in series with which he placed a booster. The booster draws current back toward the station, adding its E.M.F. to that in the feeder. Cardew used a motor generator, the series field of which was separately excited by the outgoing feeder for the same section of road. Thus the volts "boosted" were indirect proportion to the current flowing. H. F. Parshall, in adopting the return feeder booster for some of his work in England, used a generator in place of the motor-generator of Major Cardew, exciting the field by the current flowing out on the trolley feeder, thus producing volts in the armature in proportion to the current flowing. The following diagram shows Parshall's arrangement. GENERATORS Fig. 70. Modification of Major Cardew's System of Track Return Booster for Preventing Excessive Drop in Bail Return Circuits. 516 ELECTRIC STREET RAILWAYS. Sul»-Station System. — Where traffic is especially heavy, and a rail- way system widespread, it is now the practice to use one large and very economical power station with high-pressure generators, now invariably polyphase alternators, and to distribute this high-pressure alternating current to small sub-stations centrally located for feeding their districts, and there changing the current by means of static and rotary transformers into continuous current of the requisite pressure, in the case of railways 500 or 550 volts. Such systems have already been mapped out for the Man- hattan Elevated Railway, and for the Metropolitan Traction Company of New York, and are now in operation, as well as on the Central Underground Railway of London. The following diagrams will assist in making the system plain. 6UBSTATION| no. 1 DISTRIBUTION FROM SUB-STATIONS Fig. 71. TESTS ©JF STREET RAILWAY CIRCUITS. The following tests are condensed from an article by A. B. Herrick in the Street Railway Journal, April, 1899. The following instruments will be required : A barrel water rheostat to take say 100 amperes. A voltmeter reading to GOO volts. A voltmeter reading to 125 volts. An ammeter reading to say 150 amperes. A pole long enough to reach the trolley wire, with a wire running along it having a hook to make contact. Use one generator at the station, and have the attendant keep pressure constant. Test for Drop and Resistance in Overhead lines an;3 Returns. The car containing the above equipment of instruments is run to the end of the section of conductor which it is desired to test, where a line circuit- breaker divides the sections. The instruments are then connected as shown in Fig. 72. It is clear now that if the switch G be closed, current will flow through the rheostat and be measured by the ammeter. We now have the trolley and feeder B for a pressure wire back to the station, and the readin<>- of voltmeter C therefore gives the drop between the station and the point A ^HHiH^ TESTS OF STREET RAILWAY CIRCUITS. 517 518 ELECTRIC STREET RAILWAYS. in the feeder and trolley carrying the load. Voltmeter D shows the drop across the rheostat ; and if the sum of readings C and!) be deducted from the station pressure, the difference will be the drop in the ground return. Fig. 73. The station pressure can be taken by changing the lead of voltmeter C down to F as shown by the dotted line. The drop on A and its resistance having been found, the trolley-pole can be swung around and the same data be determined for the circuit B. To Stead the Ground Return Drop Directly. Open the station switch on that feeder that is being used as pressure wire, and ground the feeder to the ground bus through a fuse for safety. Connect the instruments as shown in the following cut ; then when the switch G is closed and current flows, the drop from A to F read on voltmeter C will be the drop in the ground return from F to X. j FEEDER 1 A B TROLLEY TROL c LEY p. < w D p SiMMETE Ft -AW RHEOSTAT \>G o RAML V > ^ TESTS OF STREET RAILWAY CIRCUITS. To Determine Drop at JEnd of line. 519 For use on double-track lines only, unless a pressure wire can be run to the end of line from the last line circuit-breaker. Break all cross connections from feeder to trolley-wire for one track, as at n ; connect this idle trolley to the next one back toward the station, as at C, then make the tests as in the two methods described above, connections being shown in the following cut. FFEDER TROLLEY TROLLEY A TROLl EY n C TROLLEY S^^ B "V s 1 ^ / / Vi: Iw. > o ■*^G / M . / RHEOSTAT RAIL 1 To Determine the Condition of Track Donding-, and the Division of Return Current throug-h Mails, Water or Gas Pipes, and Ground. The cut below shows the connections for this test as applied to a single track, or to one track of a double-track road. Ground the feeder A at the station, or rather connect it to the ground bus through a fuse. Then connect the track at C to A by the pole E through the ammeter M. The drop between points F and D will be the drop through the rail circuit between C and D, due to the current flowing. If connection be made to a hydrant, or other water connection, and to a gas-pipe, as at X, still retaining the rail connection at C, more current will 520 ELECTRIC STREET RAILWAYS. flow through ammeter M, due to providing the metallic return through A for the water-pipe, and the first reading of the ammeter M is to the second reading as the resistance of the water-pipe is to that of the rail return, and the current returning to the station will distribute itself between the two paths in proportion to the readings mentioned. If ammeter G be read at the same time, the difference between its reading and the sum of the other two readings will be the amount of current returning by other paths than the rail and water-pipe. If C is near the station it may be necessary to break the ground connection between rails and bus, so that all current may return over the metallic circuit A. To determine condition of bonds, move the contact C back towards D, and the decrease in drop as shown by the vm. will be very nearly proportional to the length of track, except where a bad or broken bond may be located, when the change will be sudden. TESTING RAIL BOIDS. It is not commercially practicable to measxire the exact resistance of rail joints, as such resistance is small under ordinary circumstances, and all the conditions vary so much as to prevent accurate measurement being made. The resistance of rail joints is therefore measured in terms of length of the rail itself, and there are numerous instruments devised for the purpose, nearly all being based upon the principle of the wheatstone bridge, the resistance of the rail joint being balanced against a section of the rail, as in the following diagram. Fig. 77. Diagram of Method of Testing Rail Joints. A "Weston or other reliable milli-voltmeter, with the zero point in the mid- dle of the scale, is the handiest instrument for making these tests. The points b and c are fixed usually at a distance of 12 inches apart, the point a is then moved along- the rail until there is no deflection of the needle when both switches are closed. The resistance of the joint or the portion between the points b and c is to that of the length, x, inversely as the length of the former is to that of the latter, all being in terms of the length of rail, or, Let x rr distance in inches between points a and c, y = distance between the points c and b, v =r resistance of joint in terms of length of rail, TESTING RAIL JOINTS. 521 and if x — 36 inches and y z then 12 inches, Another scheme for testing rail joints is pointed out by W. N. Walmsley in the " Electrical Engineer," December 23, 1897. In the following cut, the instrument is a specially designed, double milli- voltmeter, both pointers having the same axis, and indicating on the same scale. DOUBLE MILIVOLTMETER

"Ui:T SYSTEMS OF ELECTRIC RAIIWATi. Previous to 1893 hundreds of patents were granted on conduit and other sub-surface systems of carrying the conductors for electric railways, and hundreds of experiments were carried on ; but it has been only since that year that capitalists have had the necessary courage to expend enough money to make a really successfully operating road. The work was put into the hands of competent mechanical engineers, who perfected and im- proved the mechanical details, and the electrical part of the problem was by that means rendered very simple. 532 ELECTRIC STREET RAILWAYS. The Metropolitan Street Railway Company of New York, and the Metro- politan Railroad Company of Washington, decided, in 1894, that,.hy build- ing a conduit more nearly approaching cable construction, the underground electric system could be made a success. The former contracted for its Lenux Avenue line, and the latter for its Ninth Street line. The New York road was in operation by June, 1895; the Washington road by August of the same year ; and they continue to run successfully. While modifications have been made in some details since these roads were started, yet the present construction is substantially the same. These roads were the first to avoid the almost universal mistake of spending too little and building unsubstantially where new enterprises are undertaken. The history, in these particulars, of the development of overhead trolley and conduit roads is to-day repeating itself in the third-rail equipment of branch and local steam roads. The Metropolitan Railroad, in Washington, used yokes of cast iron placed on concrete foundations, and carrying the track and slot rails. The slot rails had deep inner flanges, with water lips to prevent dripping on con- ductors. The conductor rails were T bars 4 inches deep, 13 feet 6 inches long, 6 inches apart, and were suspended from double porcelain corrugated insulators filled with lead and mounted on cast-iron handholes. A sliding plow of soft cast iron collected the current. During the first few months of its operation there were but few delays, mostly due to causes other than electrical defects. Some trouble came from short-circuiting of plows, which was remedied by fuses on plow leads, and a water rheostat at the power- house. The flooding of conduits did not stop the road, although the leakage was 300 to 550 amperes. Under such circumstances the voltage was reduced from 500 to about 300. The average leakage on minus side, when tested with plus side grounded, was one ampere over 6,500 insulators. The positive side always showed higher insulation than the negative, possibly due to electrolytic action causing deposits on the negative pole. The Lenox Avenue line of the Metropolitan Street Railway was the first permanently successful underground conduit line in the United States. The cast-iron yokes were similar to those used on their cable lines, placed 5 feet apart. Manholes were 30 feet apart, with soapstone and sulphur ped- estal insulators located under each, carrying channel beam conductors, making a metallic circuit. At first the voltage was 350, but it was gradually raised to 500. The pedestal support was afterwards abandoned, and sus- pended insulators used every 15 feet, at handholes. At one time iron-tube contact conductors were tried, but they proved unsatisfactory. The details of track construction for underground or sub-surface trolley railroads are essentially of a special nature, and are determined in every case by the local conditions and requirements. They belong to the civil en- gineering class entirely, and will not be treated here in any way other than to show cuts of the yokes and general construction. The requirements' of the conduit for sub-surface trolley conductors are first, that it shall be perfectly drained, and second, that it be so designed that the metallic conductors are out of reach from the surface, of any- thing but the plow and its contacts. Another requisite is that the conduct- ing rails and their insulated supports shall be strong and easily reached for repairs or improvement of insulation. The conducting rails must be secured to their insulating supports in such a manner as to provide for expansion and contraction. This can be done by fastening the center of each section of bar solid to an insulated support at that point, and then slotting the ends of the bar where they are supported on insulators. The ends of the bars will be bonded in a manner somewhat similar to the ordinary rail bonding. The trolley circuit of the sub-surface railway differs from the ordinary overhead trolley system in that while the latter has a single insulated con- ductor, and return is made by the regular running rails, the former has a complete metallic circuit, local, and disconnected in every way from track return. The contact rails must be treated like a double-trolley wire, and calculations for feeders and feeding in points can be made after the methods explained for overhead circuits and feeders earlier in this chapter. Feeders and mains are usually laid in underground conduits for this work, and the contact rails may be kept continuous or may be divided into as many sections as the ser- vice may demand, taps from the mains or feeders being made to the contact CONDUIT SYSTEMS OF ELECTRIC RAILWAYS. 534 ELECTRIC STREET RAILWAYS. FlG. 92. Drainage at Manhole of Conduit, Metropolitan Railroad, Washington, 1895. PLAN OF CLIP Fig. 93. Clip and Ear for Conduit. Metropolitan Railroad, Washington, 1895. rails at such points as may be determined as necessary. All the insulated conductors should be of the highest class ; may be insulated with rubber or paper, but should in any case be covered with lead. Especial care should be taken in making joints between the conducting rail and copper conductor so that jarring will not disturb the contact. Other than the above few general facts it is difficult to say much regard- ing this type of electric railway, for it is so expensive to install that it can be used in but a few of the largest cities, and in every case will be special, and require special study to determine and meet the local conditions. The reader is referred to the files of the street railway journals for complete descriptions of the few installations of this type of electric railway. CONDUIT SYSTEMS OF ELECTRIC RAILWAYS. 535 Following are a number of cuts showing the standard construction of electric conduits as designed and built by the Metropolitan Street Railway Company, of New York. The system of railway may be said to use all the latest methods, including wire-carrying conduits along side or under the tracks, as will be seen by the next cut. The porcelain insulator here shown for supporting the contact rails is very substantial in design and construction, and by its location at a hand- hole is easily reached for cleaning, repairs, and replacement. The jjIoiv has also received careful attention, and those now used as standard by the Met- ropolitan Company leave little to be desired. Fig. 94. Section of Conduit, Metropolitan Street Railway, New York. - Standard Work, 1897-98. Section, Side and End Elevation of Plow, Metropolitan Street Railway, New York. — Standard Work, 1897-99. 536 ELECTRIC STREET RAILWAYS. Fig. 96. Plan and Elevation of Plow Suspension from Truck, Metropolitan Street Railway, New York. — Standard Work, 1897-98. Fig. 97. Section and Elevation of Insulator, Metropolitan Street Railway, New York. — Standard Work, 1897-98. SVRFACi! CONTACT Oil EJLECOrKO-lflACHtfETIC SYSTEMS. The development of surface contact systems began even earlier than the use of the overhead-trolley wire, and many patents have been issued on the WESTINGHOUSE SYSTEM. 537 same. Most of these failed through ignorance of the requirements, and timidity of capital in taking up a new device answers for others. The Westinghouse Electric and Manufacturing Company and the General Electric Company finally took the matter up, and being equipped Avith vast experience of the requirements, and the necessary engineering talent and apparatus, have each developed a system that is simple to a degree, and is said to cost hut half as much to install as the conduit system, and to offer advantages not known to that or other systems. I quote as follows from a bulletin issued by the Westinghouse Electric and Manufacturing Company. Some Advantages of the System. No poles, overhead wires, or troublesome switches are employed. The streets, yards, and buildings are left free of all obstructions. The facility with which freight cars can be drilled in yards and through buildings, without turning the trolley whenever the direction of a motor car or locomotive is reversed, and the absence of the necessity of guiding the trolley through the multiplicity of switches usually found in factory yards and buildings, is of great advantage, permitting, in fact, the use of electric locomotives where otherwise electricity could not be used. The only visible parts of the system, when installed for street railway work, are a row of switch boxes between the tracks, flush with the pave- ment, and a double row of small contact buttons which project slightly above the pavement, and do not impede traffic in any way. This system can be used in cities where the use of the overhead trolley is not permitted, and if desired the continuation of the road in the suburbs can be operated by the cheaper overhead system. It would only be neces- sary to have a trolley base and pole mounted on the car, the pole being kept down when not in use. There are no deep excavations to make. The system can be installed on any road already in operation without tearing up the ties. The cost is only about one-half that of a cable or open conduit road. The insulation of all parts of the line, the switches, and the contact but- tons is such that the possibility of grounds and short circuits is reduced to a minimum. The system is easy to install, simple in operation, and reliable under all conditions of track and climate. Finally, the system is absolutely safe. It is impossible for anyone on the street to receive a shock, as all the contact buttons are " dead " except- ing those directly underneath the car. Requirements. In devising this system the following requirements of successful working were carefully considered. The insulation must be sufficient to prevent any abnormal leakage of current. The means for supplying the current to the car must be infallible. The apparatus must be simple, so that inexperienced men may operate it without difficulty. The system must operate under various climatic conditions. Finally, absolute safety must be assured. WESTOGHOVSE SYSTEM. This system includes the following elements. First. Electro-magnetic switches, inclosed in moisture-proof iron cases. Each switch is permanently connected to the positive main or feeder which is laid parallel to the track. Second. Cast-iron contact plates or buttons, two in each group, placed between the rails and electrically connected to the switches. A separate switch is provided for each group of buttons. Third. The conductor forming the positive main or feeder. This is com- pletely inclosed in wrought-iron pipe, and is connected to the various switches. 538 ELECTRIC STREET RAILWAYS. Fourth. Metal contact shoes or hars, suspended from the car trucks ; two bars on each car. Fifth. A small storage battery carried upon the car. The operation of the system is described as follows, and is illustrated by cuts making plain the text. H— SHUNT COIL I— SERIES COIL Fig. 98. Diagram of Switch Connections. CAR WIRING D =. STORAGE BATTERY' Fig. 99. Diagram of Car Connections. Electro-magnetic switches, Xl5 X2, X3., inclosed in water-tight casings, are installed at intervals of about 15 feet along the track to be operated. Each switch is provided with two windings, I and H, which are connected by the wires N and M to two cast-iron contact buttons, 1 and 2, which are mounted on suitable insulators and placed between the rails. Each car to be operated on this system is provided with two spring- mounted T steel contact bars, Q3 and Q2, and a few cells of storage battery in addition to the usual controllers and motors. The contact bars are mounted at the same distance apart as the contact pins, 1 and 2, so that as the cars advance along the track the bars will always be in contact with at least one pair, as the length of the bar exceeds the distance between any two pairs by several feet. Suppose a car is standing on the track over the switch X2, the contact bars, Qx and Q2, being then in connection with the buttons 1 and 2 respec- tively. The first step is to "pickup" the current, i.e., render the buttons 1 and 2 alive. Switch A is first closed ; this completes the circuit from the storage bat- tery, D, through the wiring, R, contact shoe, Q15 button No. 1, and shunt coil, H, to the ground. The current passing through H magnetizes the core, S, which in turn attracts the armature, P, closing the switch and es- tablishing connection between the 500-V main feeder K, and button No. 2, through the contacts, JJ, coil I, and wiring N. Switch C is noAV closed and switch A opened ; the switch X, is kept closed, however, by the current flowing from button No. 2 through bar Q2, connection T, resistance L, con- nection R, bar Qlt button No. 1, connection M, coil H to ground. The car now proceeds on its way, current from the main passing through connection T, to the controller and motors. When the car has advanced a short distance the contact bars make connection with the pair of buttons connected to switch X3. Current then passes from bar Q, through the shunt coil of this switch. The operation described above is then repeated. As soon as the bars leave the buttons 1 and 2, current ceases to pass through the coils I and H of switch X2, and this switch immediately opens by grav- WESTINGHOTJSE SYSTEM. 539 ity, leaving the buttons connected to it dead and harmless. As connection with the main has already been established through switch X3, there will be a continuous flow of current from the feeder, and no flash will occur either at the button or the switch. It will be observed that all the current passing to the car from the main through switch contacts J J passes through the series coil, I, holding the switch firmly closed and precluding all possibility of its opening while cur- rent is passing through the contacts, even should the circuit through coil H be interrupted. Although the act of "picking up the current " requires some time to describe, it takes in practice only a few seconds. Two separate switches, A and C, are shown in the diagram; but in practice one special switch of circular form is provided, and the necessary combina- tions required for " picking up the current " are made by one revolution of the switch handle. The battery need only be employed to lift the first switch; for after that has been closed, the contact shoes bridge the main voltage over from one set of pins to another, as described, thus closing the successive switches, with- out further attention from the motorman. The battery is charged by leaving switches A and C closed at the same time. The Switch. Fig. 100 shows the general arrangement of switch, bell, and pan. The switch and magnet are mounted upon a marble slab, which is secured in the bell by means of screws to the bosses, B B. The switch magnet, M, is of the iron-clad type. It is secured to the upper Fig 100. Section of Switch, Bell, and Pan. side of the marble base, and is provided with a fine (shunt) winding for the " pick up " current, and a coarse (series) winding through which the work- ing current passes. When magnetized the poles attract an armature attached to a bridge piece, J, each end of which carries a carbon disk, N. R, R, are guides for the bridge piece, J. Directly above each of the carbon disks, N, is a stationary disk, O, mounted upon a marble base. One of the disks, O, is permanently con- nected by means of one of the contact cups, G1} as explained later, to the positive main cable, and the other, through the series coil and cup, G2, to the positive contact button. 540 ELECTRIC STREET RAILWAYS. The pan, C, is provided with four bosses, S, to support the vertical split pins, F, which are insulated from the pan. These pins slide into recepta- cles, G, on the switch base. The pins, F, are provided with connectors, I, for the purpose of making connection with the several cables, H, which pass through the holes in the under side of the pan. The pan is completely filled with paraffine after the connections are made, thus effectually keeping out all moisture. The object of the bell, A, and the pan, C, with the split pins, F, and the cups, G, is to provide a ready means of examination of the switch without disconnecting the wires. The bell can be lifted entirely free of the pan. In replacing it, it is only necessary to see that a lug, T, on the side of the cover, fits into a slide, U, on the frame. When in this position the split pins make connections with their corresponding cups, G. The bell, A, is provided with lugs, L, to facilitate handling ; and also a double lip, W. The inner portion of this lip fits into and over the annular groove, D, of pan C. This groove is filled with a heavy non-vaporizing oil. The outer portion of lip, W, prevents Avater from entering the groove. The object of the groove, D, and the lip, W, is to make a waterproof joint to pro- tect the switch and cable terminals without the necessity of screw joints or gaskets. The bells are all tested with 25 pounds air pressure ; they may be entirely submerged in several feet of water without affecting the operation of the system. Xlie Contact Kuttons are made of cast iron. They are about 4J inches in diameter, and, when installed on paved streets, project about five-eighths of an inch above the pavement and offer no obstruction to traffic. This is sufficiently high to enable the collector-bars to make contact, and at the same time to entirely clear the pavement. For open-track installations they are substantially mounted in a combination unit as described below. Fig. 101. Section of Combination Unit. The Combination Units. The bell and pan are entirely inclosed in a cast-iron switch-box. This box and the contact buttons are made into a complete unit as shown in Fig 20. Each unit consists of three separate castings. The cylindrical cast-iron box, which incloses the switch, bell, and pan, is bolted into a recess provided for that purpose in the bottom of the spider-like structure, Avhich is a sep- arate casting, consisting of box rim, receptacles for the button insulators, and supporting arms. The removable lid is the third casting. The insulators, A, Fig. 100, are made of a special composition, and are ce- mented into the tapered cups, B, and supported by the iron plates, C. The contact buttons, E, are mounted on top of these insulators and stand, when installed, about one inch above the rail. The four arms, G, are secured to the ties by means of the bosses, H, thus reducing to a minimum the labor of leveling the boxes and avoiding the necessity of special ties. WESTINGHOUSE SYSTEM. 541 Mains and Wiring-. The positive main or feeder is incased in a 1^-inch iron pipe, and passes directly through each switch-box, and a tap is made to each switch, the switch-boxes being all connected by the iron pipe, as per cut below. ruu u u uuu-u Fig. 102. Track Equipped for Track Return Circuit. No additional wires are used to interconnect the coils or contacts of ad- jacent switches. The Contact Bars are of steel, of ordinary T section. They are sup- ported from the car trucks by two flat steel springs and adjustable links. These bars are inclined at the ends so that they may readily slide over the buttons and over any ordinary obstacle. Insulated Return line. In case it is considered best not to use the rails as the return line, insu- lated mains for this purpose may be included in the system. It is only necessary to install another row of contact buttons, another collecting bar, n r i r n r n r pi, j§) Jc qI @ Jq T - i _J ■■ 'i ;■ ^®Y 'wWf U L U L U L Fig. 103. Track Equipped for Insulating Return Circuit. and to use double-pole switches. Fig. 103 illustrates an installation of this kind. For all ordinary work, however, the ground return is satisfactory. Modifications of the System. The description given on the preceding pages applies to the system as in- stalled for yard and similar work. Modifications can be made and detail matters arranged according to the requirements of each case. Street Railway Work. The foregoing description applies to installations where the track is open (unpaved),and where it is unnecessary to make provision for traffic crossing the tracks except at certain points. For street railway work, the switch- boxes are preferable installed outside the track, Avhile the buttons are placed between the rails and mounted on a light metal tie, as shown in Fig. 542 ELECTRIC STREET RAILWAYS. The operation of the system is exactly the same as in open-track work. Connecting wires pass from the buttons under the tie to the switch-boxes. For double-track work the switches are installed between the two tracks, and the boxes may be built to hold two switches, one for each track. NE or_PAVING CHANNEL IRON Fig. 104. Section of Track Equipped for Street Railway Service. When, as is sometimes necessary, the buttons are placed in a single row, it is necessary that the "pick-up" current should be of the same voltage as that of the main circuit, and consequently the car-wiring indicated in Fig. 96 is used, instead of that shown in Fig. 99. Fig. 105. Diagram of Car-Wiring. Referring to Fig. 105, the method of "picking up" the current is as fol- lows : Switch A is first closed ; this completes the circuit from a storage battery D, through a small 500-volt motor-generator F, which immediately starts. As soon as it is up to speed, which only requires a few seconds, switch B is closed ; current then passes from F through the wiring R, to contact shoe Q, and then through the switch magnet, as explained on page 530. Switches A and B are then opened, thus stopping tbe motor-generator, which need only be used to operate the first switch. The successive switches are closed, as described on page 527. This arrangement of a high-voltage " pick-up " may also be used advan- tageously with two rows of buttons where the track is liable to be obstructed by mud or snow. Sectional Mail Construction. For suburban railway or similar service two light rails may be substituted for the two rows of contact buttons, as shown in Fig. 90. The cars are then equipped with contact shoes instead of bars. These rails are insulated from the ground, and may also be insulated from each other wherever desirable, thus breaking them up into sections, which are each controlled by a single switch. The sections may be made of any desired length to suit the conditions. For example, between stations they may be 500 or more feet long, while near stations or crossings, where anyone is liable to come in contact with the rail, the length of a section may be reduced to 50 feet or less. The electrical operation of two-rail installations is the same as when two rows of buttons are used. The sectional switches along the tracks are entirely under the control of the motorman, and the rails may be rendered " dead" at any moment should occasion arise. GENERAL ELECTRIC COMPANY. 543 The Westinghouse Company use a system of surface contact all oyer its large works at East Pittsburg, and another plant has been in operation for some time at Indian Head, Md. Fig. 106. Sectional Rail Installation. «EXEHAI ELECTRIC SITSTJEM: OI1 S1J1IFACE CONTACT RAILWAY. Following is a description of the surface contact system, as developed by the General Electric Company, and practical application of it has been made at Monte Carlo, and at the company's works at Schenectady. The description is from a report made by W. B. Potter, Cf . Eng. of the Railway Department, and written by Mr. S. B. Stewart, Jr. In the operation of electric cars, by tbe closed conduit surface plate con- tact system of the General Electric Company, the current is collected for the motor service by means of two light steel shoes carried under the car, making contact with a series of metal plates, introduced along the track between the rails, automatically and alternately energized or de-energized by means of switches grouped at convenient places along the line ; the method of the switch control being such that in the passage of the car, in either direction, it is impossible for any plate to become alive except when directly under the car body. In ordinary street car practice, the contact plates are spaced approxi- mately ten feet apart, positive and negative plates being staggered, as shown in Fig. 106, which admits of but three plates ever being covered at any one time by the shoes, which are so designed as not to span more than two plates of the same polarity. In grouping the switches it is customary to locate them either in vaults constructed between or near the tracks, or in accessible places along the side of the street, the location and spacing of groups and number of switches in each group being based upon a comparative cost between the style of vault or other receptacle, and the amount of wire with ducts be- tween the contact plates and their corresponding switches. The main generator feeder is carried to each vault or group, and auxiliary feeders from it are distributed to each switch, the track rail being utilized for the return circuit. 544 ELECTRIC STREET RAILWAYS. The operation or performance of this system can be readily traced out bv reference to Fig. 107. It will be seen that the current in its passage to the motor from the positive generator conductor passes to contact A of switch No. 2 through the carbons on its magnet armature (which has been lifted by the energized coilG) to contact plates B and C, through the contact shoe D to the controller and motor, coming out at contact shoe E to the contact plate F, when it passes through the coil of the automatic sAvitch G, ener- gizing it and returning by the track-rail H ; thus maintaining contact at switch No. 2 armature carbons as long as the shoes remain on the contact plates C and F. It should now be noted that contact plate B is energized MOTOR <^n /^^ -|QIIMil|^ Fig. 107. Diagram of Connections for Surface Contact Railway Plate System, General Electric Co. as stated above. As the car proceeds, the shoe D spans the plates B and d thereby keeping the coil of switch No. 2 energized after shoe has left plate C, and until shoe E comes in contact with plate J, which immediately ener- gizes coil No. 1, thus making the preceding contact plate energized, prepara- tory to the further advance of the car. It will be noted in the above description of the performance of the system, that we have assumed switch No. 2 on Fig. 107 as closed; it should therefore be understood that an aux- iliary battery circuit is necessary in starting or raising a first switch, pre- paratory to its armature being held in contact position by the generator current, which current energizes the preceding contact plates consecutively as described above. The battery current is brought into the automatic switch circuit momen- tarily during the period of first movement of handle of the controller in starting a car, the transition of the controller cylinder also bringing the generator current in connection with the battery for a short period of time, thus replenishing the elements sufficiently to operate the switches. The battery is also used to supply current for lighting the car, the generator circuit being disconnected while the car is at rest. Surface Cobs tact .Plates. The surface contact plates are made of cast iron, with wearing surfaces well chilled, designed to be leaded into cast-iron seats in such "a manner that they are thoroughly secure, but can be readily removed by special tongs for the purpose. The seat is imbedded in a wooden or composition block set into a cast-iron box, the latter being spiked or screwed to the tie. A brass terminal is fastened to the seat for the reception of the connecting wire from the switch. See Fig. 106. GENERAL ELECTRIC COMPANY. 545 As stated above, the plates are usually located 10 feet apart for straight line work, but somewhat closer on curves, depending upon the radius of the curve and length of contact shoe. The negative and positive contact plates are staggered with a uniform angular distance between them, situated not less than 10 inches from the track rails. Fig. 108. Plan and Section of Track, Monte Carlo, Europe. General Electric Company's Surface Contact System, 1898. Surface Contact Switch. The automatic switches are constructed on the solenoid principle, the armature or core of which is employed in closing the contacts as shown in Fig. 109. Automatic Switch for Open Conduit, Surface Plate Contact System. 546 ELECTRIC STREET RAILWAYS. Fig. 109. The end of the armature core is provided with a pressed sheet- steel carbon-holder, for the purpose of supporting the carbon contacts which are held in place by bronze clips and cotter pins which can easily be re- moved. The pressed-steel carbon-holder can also be detached with little trouble by removing the end holding it to the core. Copper plates are se- cured to the slate base for contact surfaces and the attachment of feeder- wires. The wire of the solenoid is wound on a copper spool and placed in a bell-shaped magnet frame, and a pole-piece, slightly recessed to receive the end of the armature core when the switch is in a closed position, is at- tached to the top cover, and extends part way down through the winding. The recess in the armature increases the range of the magnet, making the attraction uniform except at the point of contact where the power increases rapidly, thus securing an excellent contact. A blow-out magnet coil is con- nected in series with the feeder current, and so situated that the influence of its poles is used to rupture any arc that might be formed while the switch is opening ; however, this blow-out magnet is used simply as a precaution- ary device, as under ordinary conditions there is no arcing, the succeeding automatic switch closing the circuit before it is opened by the preceding one. Each vault or group of switches should be provided with cut-outs or an automatic circuit breaker to protect them in the event of short circuits. Surface Contact Shoes. The contact shoes are made of " T " steel of light section, the suspension for which is an iron channel beam extending longitudinally with the truck frame directly under the motors, with a substantial wooden cross-arm at- tached to each end for the shoe-supporting casting, the shoes being attached to these supporting castings by a spring equalizing device for maintaining the shoes at the proper height, and also for making them flexible enough to meet any slight variations in the contact plates and track rails. The shoes when in their correct position should never drop over one-fourth inch below the surface contact plates, and are designed so that they may raise three- fourths of an inch or more above them. See Fig. 109. Fig. 110. Collecting Shoes, Monte Carlo, Europe. General Electric Company's Surface Contact System, 1898. A screw adjustment is provided to lower the shoes as they wear away, or to take care of any other discrepancies due to wear of parts, etc. ; if they are allowed to drop too low they will interfere with rail crossings, causing short circuits. Storagre Batteries. _ It requires for closing the first automatic switch when starting, and for lighting the car approximately, ten storage battery elements capable of 35 amperes rate of discharge for five hours. GENERAL ELECTRIC COMPANY. 547 The batteries are only slightly exhausted in making the initial connec- tions through the automatic switch, as it only takes approximately 15 am- peres momentarily to perform this work, the battery is immediately recharged by current which has passed through the motors. The battery serving as a rheostatic step, this momentary cnarging does not represent any extra loss of energy. The circuit connections of the battery are accomplished in the controller, and require no attention on the part of the motorman. Car JLigliting-. The amount of recharging derived from the motor circuits is sufficient to operate the automatic switches, but where lighting of the car is done from the same battery, an additional recharge is required. Assuming that 10 20-volt lamps are used for lighting a car, the batteries will need to be recharged every night about five hours, at an approximate rate of 25 amperes. It is customary to run leads from both the positive and negative terminals of the batteries to charging-sockets attached to the under side of one of the car sills in a convenient place for connection to the charging-wire. A small generator of low potential (30 volts) driven by a motor or other method is required for supplying current for recharging the batteries where the desired low-potential current is not accessible, and the wiring from the charging source should be run to a location in the car-house most convenient for connections to the battery sockets. These locations may be fixed either in the pits or on posts at the nearest point to Avhere the cars will be sta- tioned, and there should be flexible lead wires attached to plugs for connect- ing to the battery circuit on the car. In wiring the car-house for the battery connections, it would be found convenient to designate the polarity of the various wires either by different colored insulation or tags, and the plugs at the ends of the flexible leads should be marked plus and minus to avoid mistakes in making connections with the car battery receptacle. Motors and Controllers. The motor and controller equipment used with the surface plate contact system is standard apparatus as ordinarily employed for electric car service, with the exception that provision is made in the controller for cutting in and out the storage battery while starting the car. Care of Apparatus. As success in the operation of the contact plate system depends largely on the care of the apparatus, a few general remarks on the subject will not be out of place here. Care should be taken that the contact plates are kept clean, and they should be frequently inspected, the roadbed being well drained. Any small quantity of water temporarily standing over the tracks, however, would do little harm, as the leakage through the water would not be sufficient to create a short circuit, although this condition should not be allowed to exist any length of time. The automatic switches should be carefully inspected and all cast-iron parts thoroughly coated with heavy insulating paint, and a test for insula- tion or grounds be made frequently, and all the parts kept clean and free from moisture. The contact shoes, in order to prevent leakage, should have their wooden supports well protected with a coating of an insulating paint, and should also be occasionally cleaned. The storage batteries should be properly boxed and should have the cus- tomary care which is necessary to keep them in good working order. TRANSMISSION OP POWER. The term " Transmission of Power," as used by electrical engineers, lias come to have a conventional meaning which differentiates it from what must be considered its full meaning. Any transmission of electric current, for whatever practical purpose, whether for lighting, heating, traction, or power-driving, must of course be a transmission of power ; but the conven- tional meaning of the term as now used by electrical engineers and others eliminates many of these objects, and is held to mean simply the trans- mission of electric current from a more or less distant point or station to a center from which the power is distributed, or to power motors at different points in a factory or other installation. While the distances over which electric current is transmitted for arc lighting in some large cities and in many small places far exceeds the length of line of the ordinary or average power transmission, yet the former is never alluded to as transmission of power. The same condition obtains with traction, the transmission of cur- rent covering miles of territory, and yet it is only alluded to as power transmission when the current is transmitted from a central point to vari- ous sub-stations from which it is distributed. The engineering features of transmission of power will all be found treated under the separate heads in their respective chapters, and the fol- lowing is a short resume, of the subject matter. Building- : Structural conditions and material. JHotive Power: "Water power : Turbines, etc. Steam power : Boilers and appliances. Engines and appliances. Shafting and pulleys. Belting and rope drive. Generators: Dynamos : direct current. alternating current, double current. Transmitting- Appliances : Switchboards. Transformers, step up. Botaries. Cables and pole lines. Conduits, etc. Distributing- Appliances; Sub-stations and terminal houses. Transformers, step down. Switchboards, high tension and secondary. Botary converters. Direct current motors. Synchronous motors. Induction motors. Frequency changers. Distributing circuits. 548 DISTRIBUTING APPLIANCES. 549 Much has been written regarding the relative values of the different methods of transmitting power, and comparison is often made between the following types, i.e., a. Wire rope transmission. b. Hydraulic transmission, high pressure. c. Hydraulic transmission, low pressure. d. Compressed air transmission. e. Steam distribution for power. /. Gas transmission. g. Electrical transmission. All of the first six methods listed have so many limitations as to distance, efficiency, adaptability, elasticity, etc., that electricity is fast becoming the standard method. The matter of efficiency alone is 'one of the best argu- ments in its favor, and I take from Dr. Bell's book, " Electric Power Trans- mission" the following table of the efficiencies such as have been found in practice. System. Per Cent Efficiency ; Wire rope Hydraulic high pressure Hydraulic low pressure Pneumatic Pneumatic reheated, virtual efiiciency Electric 45 50 50 65 For short distances out of doors, transmission by wire rope is much used both in the United States and Europe, and where but few spans are neces- sary, say less than four, the efficiency is very high. Hydraulic transmission is in considerable'use in England, but except for elevator (lift) service is in little use in the United States. Pneumatic transmission is in wide use in Paris, but not so for general distribution in the United States, although for shop transmissions for use on small cranes and special tools is making good progress. Electrical transmission is so elastic and so adaptable to varied uses, and has been pushed forward by so good talent, a not small factor, that its progress and growth have been simply phenomenal. In one place alone, that of traveling cranes for machine shops, it has revolutionized tbe handling of material, and has cheapened the product by enabling more work to be done by the same help. Electric Power Transmission may be divided into two classes, i.e., long distance for which high tension alternating current is exclusively used ; and local or short distance transmission for which either direct current or polyphase alternating current are both adapted, with the use of the former largely predominating, owing perhaps to two factors, a, the much earlier development of direct current machinery, and b, to the fact that a large number of manufacturers are engaged in the building of direct current machinery. Both types of current have their special advantages, and engineering opinion is, and will probably remain, divided as to which has the greater value. Long distance transmission is now accomplished by both three-phase three- wire, and by the two-phase four-wire systems, with the former predoni- 550 TRANSMISSION OF POWER. mating for the greatest distances, owing to economy of copper. Each sys- tem lias certain advantages over the other, and both have strong advocates among engineers. For the distribution of very large amounts of power the three-phase system presents a strong point in its economy of copper, and another in simplicity of switching appliances. Every case of electric transmission presents its own problem, and needs thorough engineering study to decide what system is best adapted for the particular case. It is, therefore useless to enter into any detailed discus- sion here, as all the engineering details are treated of elsewhere in the book under the respective departments. The economic discussion does not enter into the engineering problem except in the preliminary study, which has presumably been satisfactory before reference is necessary to this book. Limitations of Voltage.— While 10,000 volts pressure was used with some distrust for a time previous to 1898, since that time 15,000, 20,000, 25,000, and 40,000 volts have been and are still in use with substantial satisfaction. Properly designed glass or porcelain insulators, made of the proper material, and tested under high pressure conditions, cause little trouble from puncture or leakage. The latter is its own cure, for the reason that the leakage of current over the surface of the insulator dries up the mois- ture. Dry air, snow, and rain-water are fairly good insulators, and offer no difficulties for the ordinary high voltages. Dirt, carbon from locomotive smoke, dust from the earth, and such foreign material that may be lodged on the insulators, are sure to cause trouble. In the West and some sections of the East many insulators are broken by bullets fired by the omnipresent marksman. Oil insulators have proved worse than useless, as dirt and dust, to say nothing of bugs, are gathered by the oil, and produce very bad results'; they were introduced in the United States in some of the early high-voltage installations, but after a short time the cups holding the oil had to be broken off. Glass makes the surest insulator, as the eye can make all necessary tests ; but it is so fragile that porcelain is more commonly used. It is not safe to accept a single porcelain insulator without a test with a pressure at least twice as great as that to be used. The interior of the porcelain should be perfectly vitreous, and should not absorb red ink so that it can- not be wiped off perfectly clean. A convenient way of testing such insulators is to invert a number, say a dozen, in a pan of salt water ; fill the pinhole with more water of the same kind. Connect the pan with one terminal of a high potential transformer, and use as the other terminal a piece of metal, say a spike or old battery zinc pencil which will be connected to the opposite terminal of the trans- former, and inserted in the pinhole of each insulator. A double-pole switch should be used to open and close the low-pressure side of the testing trans- former. Under these conditions one insulator is tested at a time, and good porcelain will stand very high pressure before a breakdown. Heavy sea-fog is about as bad a condition as can be assumed for high voltage trans- mission. Mr. Ralph D. Mershon of the Westinghouse E. & M'fg. Co. made a long series of tests at Telluride, Col., on the high-pressure lines in use there. At 50,000 volts there will be a brush discharge or leakage from one wire to the next that can be seen at night, and makes a hissing noise that can be heard a hundred feet or more. This brush discharge begins to show at about 20,000 volts, on dark nights, and increases very rapidly, as does also the power loss at 50,000 volts and higher. This loss depends upon the distance apart of the conductors and their size. Above 50,000 volts the losses become serious, the discharge dis- posing of a large amount of energy. For these reasons, wires should be kept well apart~and be of as small size as other properties will allow. The wave form of E. M. F. used also influences the brush discharge, being the least in effect for sine wave curves of E. M. F., and being much in- creased by the use of the sharp, high forms of curve. Line inductance, capacity, and resonance, unbalancing of phases, etc., have caused little trouble in practice, although they should be given serious consideration, especially for lines carrying heavy currents. In regard to the frequency to be adopted for power transmission, one has to be governed by the case in hand, and the commercial frequencies avail- able at economical cost. LIMITATIONS OF VOLTAGE. 551 Since the success of the Niagara plant the frequency used there, 25 per second, has become a standard for power transmission purposes, but should be avoided if much arc or incandescent lighting is to be done. Other fre- quencies, such as 30 and 60, are in common use, the latter being the favorite for plants having a mixed output of power and lighting. It must be remembered that the higher the frequency, the more trouble- some are the rotary converters that may be connected to the system. Induction motors and synchronous motors of the revolving field type are now almost perfection, and are useful to counteract each other's effects on lines, and both give their best results at low frequencies. Alternating arc lamps cannot be used with any satisfaction on a frequency less than 40. STORAGE BATTERIES. ELECTRIC STORAGE BATTERIES. Partly condensed from articles by Joseph Appleton in " Electrical Engineer." An electric storage battery, or accumulator, is a combination of cells, each of which is a unit. In the ordinary lead, sulphuric acid type, a cell is made up of three parts — the jar, or box, the plates, and the electrolyte. Thenar, or containing-box , may be of any good non-conducting and acid- proof material of sufficient strength and rigidity to support the plates and the electrolyte. In the smaller stationary types the jar is of tenest made of glass or of hard rubber, the latter especially for portable cells where lightness is of moment. Portable cells are now often made of hard wood lined with lead. Large cells for central-station work are made of heavy planks, well jointed, and lined with five-pound sheet lead. Stationary cells should always be supported upon some well-designed in- sulator, such as porcelain, so constructed as to have a retaining-cup of oil, in order to maintain a high degree of insulation. They are also generally set up from the floor a short distance, most often on stringers of well dried and filled hard wood. The plates are of two kinds, positive and negative, arranged alternately, there always being one more negative than positive, A set or group of these plates is commonly known as an element. All positive plates are connected together, as are also all negative plates, but the positives and negatives are separated from each other by insulating strips of some kind. The electrolyte used with all lead batteries — and no others are in exten- sive use at the present time — is sulphuric acid diluted with water to a s.g. of 1.15 to 1.30 according to the type. The acid must be free from impurities, such as arsenic, nitric or hydrochloric acid, and the water must be distilled. Storage or secondary batteries of the ordinary lead, sulphuric acid type may be divided into two classes, the Plants and the Faure. Both are lead elements in dilute sulphuric acid, but are formed differently. The Plante type is constructed of lead plates so designed as to present a large surface area to the action of the electrolyte, the active material being formed on the plates, either electrically, by charging and discharging, com- monly called " forming," or chemically. In the Faure, commonly known as the pasted, type the active material is applied mechanically to a lead conducting-plate or grid. Tbe material may be active Avhen applied, or may be such that it can be converted into active material by electrical or chemical formation. Tlates. The positive plate is of lead, upon which a coating of peroxide of lead has been formed or mechanically applied. The negative plate is of pure lead, the surface of which is spongy or porous in its formation. The peroxide and spongy lead are the portions of the plates which are sub- jected to the chemical action, and are called the active material, the lead body of the plates serving practically as a support for the active material. The chemical condition of the plates and acid differs when charged and discharged. At full charge the positive plate has a dark brown coating of peroxide of lead, the negative plates having the porous or spongy condition above described, of dark slate color, and the electrolyte being of full specfic gravity and strength. In this condition, when the positive and negative poles are connected through an external circuit an E.M.F. is set up in the cell, a current flowing through the circuit from the positive plate. When discharged, the positive plates have a chocolate, and the negative a light slate color. A drab color on the positive indicates sulphating or an over dis- charge. 552 ELECTRIC STORAGE BATTERIES. 553 Chemical Action. The chemical action taking place during charging is as follows : the cur- rent enters at the positive pole, passing through the acid to the negative. Both plates contain sulphate of lead, due to the preceding discharge, and the net result of the passage of the current is to decompose this sulphate, and at the same time to transfer all the oxygen from the negative to the positive. At the completion of the charge, the negative is entirely free from oxide, and the positive contains no oxide lower than the peroxide, though it may still contain some sulphate. The reduction of the sulphate of lead forms free sulphuric acid, and, of course, increases the density of the elec- trolyte. The complete account of the chemical reactions in charging is too extensive to be given here. If charging is continued after all the active material has been converted to peroxide of lead and spongy lead, oxygen and hydrogen gas will be given off in bubbles. In discharging, the sulphur radical in the acid combines with the active material on both plates, forming sulphate of lead, the specific gravity of the electrolyte being reduced. When all the active material has been acted upon, the cell is discharged, as an equilibrium has been created between the positive and negative plates, and the E.M.F. set up by the chemical action has been reduced to zero. In practice the E.M.F. is never allowed to fall below 1.8 volts. The chemical reactions are given as follows, by Frankland. If the buff lead salt be the active material of the battery plates, the fol- lowing equations express the electrolytic reactions taking place in the cell : — I. In charging — (a.) Positive Plates. S3Pb,014 + 30H2 + 05 = 5Pb02 + 3S04A2. Buff load Water. Lead Sulphuric Water. Peroxide. Acid. (b.) Negative Plates. S3Pb5014 + 5H2 = 5Pb + 3S04A2+20H2. II. In discharging — (a.) Positive Plates. 5Pb02 -f 3S04H2+5H2 = S3Pb5014 + 80H2. (b.) Negative Plates. 5Pb + 3S04H2 + 05 = S3Pb5014 + 30H2. If the red lead salt be the active material, then the following equations express the same electrolytic reactions : — I. In charging — (a.) Positive Plates. S2Pb,O10 + 02 + 20H2 = 3Pb02 + 2S04H2. Red lead Lead Sulphuric Salt. Peroxide. Acid. (b.) Negative Plates. S2Pb3O10 + 4H2 =: 3Pb + 2S04H2 + 20H2. II. In discharging — (a.) Positive Plates. 3Pb02 + 2S04H2 + 2H2 = S2Pb3O10 + 40H2. (b.) Negative Plates. 3Pb + 2S04H2 + 202 = S2P3O10 + 2H2. 554 STORAGE BATTERIES. It is, however, very questionable whether these salts play any important role in the normal reaction of the cell. The various oxides of lead are as follows : — Plumbous or sub-oxide Ph>0. Plumbic oxide, litharge PbO. Triplumbic oxide, or red lead miniuin Pb304. Diplumbic oxide Pb203. Monoplumbic dioxide, or peroxide PbU2. CALCriATIOlf OF E.M.F. OF STORAGE BATTERY, Streintz. Let JE=E.M.E. required. S = Specific gravity of the electrolyte. s = Specific gravity of water at the temperature of observation. Then JS= 1.850+ .917 (S — s). Wade. 7F=work in joules. Q= coulombs of electricity passed through the electrolyte. H= number of calories liberated by the recombination of a unit weight of one of the decomposed ions. e = its electro-chemical equivalent. c = its chemical equivalent. h = electro-chemical equivalent of hydrogen = -00001038. J = Joule's coefficient = 4.2. E = E.M.~F. required. Then W— QE— QJeH. E = JeH. e = hc. E = JhcH=4t.2x. 0000 038 cH= .0000436 cH. heat of formation valency .0000436 X heat of formation valency .0000436x46,000 CALCFIATIOir OF THE CAPACITY OF A STORAGE BATTERY IW AMPEfiE HOURS. The current in ampere hours maintained by the consumption of any given chemical substance varies with the change of valence and inversely with the molecular weights of the transforming substance. The combustion of liberation of 1 pound of hydrogen corresponds to 12,160 ampere hours. The theoretical capacity in ampere hours may be calculated as follows : — F= change of valence of the ions. JF=the sum of the molecular weights affected. 12,160 = capacity per pound of hydrogen. „. n , 12,160 X V Then Capacity per pound = In lead-lead-sulphuric acid cells the above formula gives 40.24 ampere hours as the capacity per pound of lead sulphate. The above formula is based on the supposition that the entire material of both plates is transformed into lead sulphate. This is never accomplished, and Fitzgerald gives as a safe rule : .53 oz. lead peroxide and the same Aveight of spongy lead per ampere hour for a 10-hour rate of discharge, .62 oz. for a 5-hour rate, .70 oz. for a 3-hour rate, 1 oz. for a 1-hour rate. All above for the ordinary thickness and an electrolytic density of 1,200. CALCULATION OF E.M.F. OF STORAGE BATTERIES. 555 IHH Hl'DRO^ETEll. The hydrometer is an instrument for determining the density of liquids. It is usually made of glass, and consists of three parts: (1) the upper part, a graduated stem or fine tube of uniform diameter ; (2) a bulb, or enlarge- ment of the tube, containing air ; and (3) a small bulb at the bottom, con- taining shot or mercury, which causes the instrument to float in a vertical position. The graduations are figures, representing either specific gravities or the numbers of an arbitrary scale, as in Beaume's, TwaddelFs, Beck's, and other hydrometers. There is a tendency to discard all hydrometers with arbitrary scales, and to use only those which read in terms of specific gravity directly. This ten- dency is all the more to be indorsed, as there are considerable discrepancies in the different tables professing to give the Beaume scale, the following one being, perhaps, as much quoted as any. Deaume's Hydrometer and Specific Grat Sties Compared. Liquids Liquid s • Liquids Liquids TS2-C Liquids Liquids Heavier Lighte r % d Heavier Lighter than Heavier Lighter * = than than £ § than than than rcr- Water, Water iacce Water, Water, CX- Water, Water, 2,'A sp. gr. sp. gr • AS sp. gr. sp. gr. £:: sp. gr. sp.gr.. 0 1.000 19 1.143 .942 38 1.333 .839 1 1.007 20 1.152 .936 39 1.345 .834 2 1.013 21 1.160 .930 4!) 1.357 .830 3 1.020 22 1.169 .924 41 1.369 .825 4 1.027 23 1.178 .918 42 1.382 .820 5 1.034 24 1.188 .913 44 1.407 .811 fi 1.041 25 1.197 .907 46 1.434 .802 7 1.048 26 1.206 .901 48 1.462 .794 8 1.056 27 1.216 .896 50 1.490 .785 9 1.063 28 1.226 .890 52 1.520 .777 10 1.070 1.000 29 1.236 .885 54 1.551 .768 11 1.078 .993 30 1.246 .880 56 1.583 .760 12 1.086 .986 31 1.256 .874 58 1.617 .753 13 1.094 .980 32 1.267 .869 60 1.652 .745 14 1.101 .973 33 1.277 .864 65 1.747 15 1.109 .967 34 1.288 .859 70 1.854 16 1.118 .960 35 1.299 .854 75 1.974 17 1.126 .954 36 1.310 .849 76 2.000 18 1.134 .948 37 1.322 .884 Streng-th of IMlute Sulphuric Acid of Different Densities at 15° C. (50° F0. (Otto.) Per Cent. Specific Per Cent. Per Cent. Specific Per Cent. of H2S04. Gravity. of S03. of H,S04. Gravity. of S03 100 1842 81.63 23 1167 18.77 40 1306 32.65 22 1159 17.95 31 1231 25.30 21 1151 17.40 30 1223 24.49 20 1144 16.32 29 1215 23.67 19 1136 15.51 28 1206 22.85 18 1129 14.69 27 1198 22.03 17 1121 13.87 26 1190 21.22 16 1116 13.06 25 1182 20.40 15 1106 12.24 24 1172 19.58 14 1098 11.42 Ordinarily in Accumulators the densities of the Dilute Acid vary between 1150 and 1230. 556 STORAGE BATTERIES. Conducting* Power of Dilate Sulphuric Acid of Various Strengths. (Matthiessen). Sulphuric Relative Specific Acid in Temperature. Resistances. Gravity. 100 parts C.° Ohms per by Weight. cub. centim. 1.003 0.5 16.1 16.01 1.018 2.2 15.2 5.47 1.058 7.9 13.7 1.884 1.080 12.0 12.8 1.363 1.147 20.8 13.6 .960 1.190 26.4 13.0 .871 1.215 29.6 12.3 .830 1.225 30.9 13.6 .862 1.252 34.3 13.5 .874 1.277 37.3 .930 1.348 45.4 17.9 .973 1.303 50.5 14.5 1.086 1.492 60.6 13.8 1.549 1.638 73.7 14.3 2.786 1.726 81.2 16.3 4.337 1.827 92.7 14.3 5.320 1.838 100.0 Conducting- Power of Acid and Saline Solutions. Copper (Metallic) at 66° F 100,000,000. Sulphuric Acid 1 Measure "^ Water 11 Measures lQQ n „^„^4™„+Q (Equal to 14.32 parts by weight of Acid fy8,u approximate. in 100 parts of the mixture), at 66° F. . .J Sulphate of Copper, saturated solution at ) a . .. 66° F | b'* Chloride of Sodium, saturated solution at \ R 1 ,, 66° F J bl Sulphate of Zinc, saturated solution at ) QK n ,, 6$F }35<0 (\ MOTAtlLATlOM AID CARS. Fig. 2. Standard Hydrom- eter. 8§ inches long. In small batteries, in which the cells are small enough to be handled Avhen assembled, the cells may all be assembled before placing. Large cells have to be assembled in place, as they will seldom permit change of position without considerable incon- venience. The battery-room should be dry, well lighted and ventilated, and of moderate temperature, as the evaporation of electrolyte is apt to be troublesome in heated rooms. All exposed iron work should be painted with an acid-proof paint; in fact, all metal work exposed to the acid fumes should be painted for its protection. The floor of the battery room is preferably of brick, tile, or cement, laid so it will drain easily to some common outlet. Wooden floors should never be used unless protected by lead trays to catch any stray acid. The battery room should preferably be located as near the power-house as possible, thus reducing the cost of connecting conductors, and possibly using the same attendants. INSTALLATION AND CAKE. 557 Cells should be arranged so as to be easily accessible for examination and repairs. Large cells are seldom placed in more than one tier, but the smaller ones can be erected in two or three tiers. Where cells are of glass they may be conveniently set in trays on a bed of sand, and the trays be set on insulators. Wooden tanks are set directly on insulators, as they are always built of sufficient strength to support their weight and contents. Cell Connections. In small cells the plates of one polarity are usually connected by a lead strap that is cast on the plates in a bunch, the strap of one cell being con- nected to that of the next by a bolt or screw clamp or weld. All battery con- nections should be of ample sectional area to avoid loss, and, as lead is the metal mostly used for such purposes, and as compared with copper has about seven times the resistance, it is especially important that its area be large. The best method of connecting the positive group of plates to the adjacent group of negative plates in the next cell is to bum or weld the two to a lead strap of large cross-section ; and, in case of very heavy currents, a copper conductor may be embedded in this lead strap. .Lead-Burning- Apparatus. The hydrogen flame has the special property of not oxidizing, or otherwise soiling the lead, and is therefore used for melting together two lead surfaces, notably that between cells and the sheet lead lining of the tanks. Hydrogen gas is generated in a vessel from sulphuric acid and zinc. The gas is collected and passed through a water bottle to a burner, where it is mixed with air that has been forced into the burner by a pump or bellows, the mixture being ignited for the welding. The use of this burner requires some skill and practice, especially in join- ing the edges of sheet lead, as it is very apt to burn away. All plate ter- minals, and all lead connections of any kind, must be scraped clean before connecting up. Acid. Sulphuric acid made from pyrites is not suitable for storage battery use ; only that made from sulphur should be used. Ordinary sulphuric acid con- tains many impurities that are apt to be injurious to the plates, notably, copper, iron, arsenic, nitric and hydrochloric acids. The acid should only be diluted with pure distilled water, and the acid should always he poured into the water, and never vice versa. Mix carefully, as much heat is generated. Tests for Impurities. Conner and Arsenic. —To a fresh solution of hydrogen sulphide, H S add an equal quantity of the diluted electrolyte, which must be diluted far enough so that no white precipitate is thrown down A black precipi- tate generally shows presence of copper, although it may be lead, if the acid has already been used in batteries ; a yellow precipitate shows presence of "iron -To a small quantity of the diluted electrolyte add a few drops of nitric acid, HNO„ and heat; when cold add a few drops of solution of potas- sium -sulphocyanide, KCNS; the presence of iron will be shown by a deep red color. . „„,„„> no fril Citric Acid. -Make up a solution or diphenylarmne. * R_ (C, ;H3), as , to£ lows : h gm. NH(CRH02, 100 cc. strong sulphuric acid. H2S04, 20 cc. ot water H,0; to a small quantity of this solution, in a test tube, add a small quantity of the diluted electrolyte, which must not have been in use; the pie»ence of nitric acid will be indicated by the appearance of a blue color. Hydrochloric Acid. — To a small quantity of the proposed diluted electrolyte add two or three drops of nitric acid, HNO„ heat this in a test tube then let it cool; now add two or three drops of nitrate of silver, AgNOv The presence of hydrochloric acid will be indicated by precipitated or cloudy appearance. 558 STORAGE BATTERIES. JFirst Charge. Charging current should always be ready for application when the electro- lyte is put in the cells, as it injures plates to stand in the acid without being charged. The first charge should be carried on for a much longer period than any of the subsequent or working charges, as it virtually completes the forma- tion of the plates. See that the positive pole of the charging dynamo is connected to the posi- tive pole of the battery. The voltage of charging commences at about 2 volts per cell, and rises to 2.6 volts at the full charge while taking current at the normal rate shown on the maker's lists. The curves in Fig. 3 show the voltage of a cell during charge and discharge at the normal rate. Continue the first charge for at least 10 consecutive hours, and 20 or 30 would be preferable. The first charge is usually about twice the capacity of a battery, and should be made at the normal rate. This cut shows the general forms of the charge and discharge curves at any rate; but in commercial use cells are almost always worked at much higher rate than shown in the cut, and give lower efficiencies. For exam- ple, a five-hour rate of discharge is quite usual, and in some cases even higher rates. Some of the larger users of the Electric Storage Battery Company's cells usually overcharge 15 per cent. So the ampere efficiency is 85 per cent, and the watt efficiency about 72 per cent. The specific gravity of the electrolyte will drop during the first few hours of the first charge, but will rise again, as the process continues, until its maximum point is reached at full charge. If the s. g. be 1.000 at the start it will decrease to about J .180, and rise again to about 1.210 at full charge. As the charge nears completion, bubbles of gas will rise from both plates, and the charging current should then be reduced, as the active material is becoming fully formed, and cannot take up all the gas set free from the de- composition of the acid. As the amount of gas li Iterated is in proportion to the current flowing gasing will decrease as the current is decreased. It is especially important with the pasted plates that charging be com- menced immediately after the electrolyte is put in, as the plates are apt to sulphate otherwise, sulphating being the formation of a coating of sulphate of lead between the grid and the active material, which practically insulates the two from each other, and is very difficult to reduce. Sulphating will also occur with pasted plates if discharged too low. The plante form of plate is not so susceptible to injury from sulphating. INSTALLATION AND CARE. 559 It will take 20 or 30 discharges to fit a new battery to give its full ca- pacity, and it is well to charge for 25 per cent longer time at normal rate for the first few months. In ordinary work the battery will retain its nor- mal condition with a charge of 10 per cent in excess of the discharge. General Charging-. During ordinary charging of the battery keep in view the following points : — Charge at normal rate, or lower, except in emergency. Under normal charging conditions 2.5 volts may be considered full charge, although it can be charged higher than this on an over-charge. The specific gravity of the electrolyte is a good indication of the condition of the cell; but care must be taken that it is of uniform density throughout, as during charging the electrolyte at the bottom of the cell will become denser unless agitated, as the sulphuric acid liberated from the active mate- rial falls to the bottom. The water in the electrolyte will evaporate, exposing the top of the plates, unless replaced, Avhich should be done through a hose or tube reaching to the bottom of the cell, as water added- otherwise will stay on top, being lighter than the acid. The specific gravity of its electrolyte is the best possible guide to the con- dition of a cell, as it may appear fully charged by gasing and by the voltage, and yet its condition be such as to cause these appearances when only partly charged. As the hydrometer measures the density of the liquid in the upper part of the cell only, care must be taken that the electrolyte be stirred up so that the density will be the same throughout the cell, or nearly AMPERE HOURS AMPERE HOURS : '".""".".; ' a::. A / y ^ A ' Z / /- Vi A- -V 2- A~ /- z ± 2 X / 4 ± ■ y \ / A X- A X- J f * ? / .■" / yS* n - J-*.*—- - . 0 ■* ii O CO <0 rf. oiocoto^- «gm co^caococo SPECIFIC GRAVITY Fig. 4. Curve of Specific Gravity at Charge and Discharge. so : of course the difference will be greater in the deeper cells. As the den- sity of the electrolyte is due to the sulphuric acid in it, and the sulphuric acid is liberated from the active material in proportion to the charge given, the s. g is always a true indication of the condition of the cell as to its charge. Although not always the most economical, the highest efficiency and longest life are obtained when the battery is charged slowly, never exceeding the normal rate. Conditions of plant operation will determine the most economical method for each installation. 560 STORAGE BATTERIES. Each cell should be tested with a voltmeter and hydrometer once a week. Any cell found with voltage low should he examined thoroughly for any foreign substance that may have short-circuited it. This Avill be indicated by low specific gravity and lack of gas given off, and voltage rising slowly at the end of a charge, when it should rise quickly. Always reduce charging current near the end of charging, so as not to waste energy by escape of gas. 4000 «=T ^ 4i! ;:: : ■ ,. .. - R-- a. --T- : ■ : O 3000 \\ ■ >-~ dr.: .™i ::: <2000 ;; 0 .- >-■-■: ..,_ - ":■ 1 ^ zz "-■: AMPERES 1000 1200 1400 1600 AMI CAPACITY CURVE Fig. 5. Curve of Variation of Capacity. When discharging at normal rates, never discharge a battery below 1.8 volt. In discharging at high rates 1.8 volt will be reached before the bat- tery is discharged to the same condition as at normal discharge owing to the internal resistance, producing a greater fall of potential in accordance with the IR law. Capacity at Different Kates of IMscharg-e. The output capacity of a battery will decrease as the rate of discharge increases; but the efficiency will not, as commonly supposed, decrease in the same degree, as the decrease in capacity is due to the fact that at high discharge rates the point is soon reached where the cell is unable to maintain the rate of discharge. But when apparently completely exhausted at a high rate, a cell will still furnish current at a lower rate, and on re- charging it will be found that only the amount taken out, plus the usual excess, is necessary to recover the full capacity. The internal losses, how- ever, are greater at high rates, which reduces the efficiency to some extent. If cells are given short periods of time to recuperate, during excessive dis- charge, they will give practically the same capacity as at normal discharge. The General Electric Company is now making a recording wattmeter, es- pecially adapted for storage batteries, that will show at all times the amount of energy in the battery, as its reading will decrease with discharge and in- crease as a charge is put in. Never allow a battery to stand without charge ; even half charge is better than none, and full charge is much the best. SOME OF THE ADVAMTACJES OF STOWAGE BATTERIES. For Central Station. The chief points of advantage are : (1.) Reduction in coal consumption and general operating expenses, due to the generating machinery being run at point of greatest economy while in service, and being shut down entirely during hours of light load, the bat- tery supplying the whole of the current. (2.) The possibility of obtaining good regulation in pressure during fluc- tuations in load, especially when the day load consists largely of elevators, and similar disturbing elements. (3.) To meet sudden demands which arise unexpectedly, as in the case of UNPACKING, SETTING UP, AND USING. 561 darkness caused by storm or thunder showers ; also in case of emergency due to accident or stoppage of generating plant. (4.) Smaller generating plant required where the battery takes the peak of the load, which usually only lasts for a few hours, and yet where no battery is used, necessitates sufficient generators, etc., being installed to provide for the maximum output Avhich, in many cases, is about double the normal output. All the above advantages apply quite as well to batteries in the power- house of street railways, and for maintaining the voltage at or near the end of a branch they are of inestimable benefit. They can be so installed as to take care of both railway and lighting load, as is done at Easton, Pa. For Uarg-e Office XKuilding-s. Many of the same advantages mentioned in the above paragraphs apply quite as well to large isolated plants ; some of those in the modern office- building being much more extensive than a large proportion of the central stations throughout the country. In many such plants the night operatives can be dispensed with, as the battery will take all the lighting load. The load-peak on most office buildings is pretty heavy between four and six o'clock in the winter afternoons, and will run up very rapidly if a shower comes up in summer, sometimes getting ahead of extra engines. The storage battery can always take the load until new generators can be started. Running the dynamos at a more even load is also more economical. for Sniall Isolated Plants. For country residences and the like,Avhere buildings are far from any cen- tral supply, a dynamo or two run by a gas or oil engine, with batteries used for storing the output, enables one to have all the advantages of the current, and with compai-atively little care, as the plant need be run but once or twice per week in order to keep the battery stored. This is of especial ad- vantage when there is a small water-power. Telephone and Telegraph. Many storage cells are now in use in telegraph and telephone work, where they have replaced many hundreds, if not thousands, of gravity cells. miscellaneous Uses. For the horseless or motor carriage storage batteries are well adapted, and are in considerable use. Train-lighting is done to a small extent by storage batteries. Launches for lakes and rivers are now often propelled by storage bat- teries. Street-cars are occasionally equipped with storage batteries, and in some localities have had a precarious success. :orSTIMTCTI©]¥!S FOR UUfPACKIIG, §ETTI]¥CJ UP, ATITD I7SIMG- STORAGE KATTEK IE*. (By the Electric Storage Battery Company.) 1. The elements are packed in the following way : one set of each positive and negative plates, i.e., a complete element, are packed together in posi- tion with sheets of paper and pieces of wood between the plates. A piece of string is tied around same to keep it compact and tight (see illustration, Fig. 6). Take the elements out of the packing cases carefully, and see that they are free from all dirt and foreign material. Place each element on a piece of wood, as shown in Fig. 7; cut the string and take out the paper and wood. Space the plates so that the separating rings can be placed in position on the positive plates, two to each positive plate. Be sure 562 STORAGE BATTERIES. that the containing jar is clean before placing the element in it. In setting up the larger elements it is advisable to tie a piece of string around the ele- ment after all the rubber separating rings are in position to prevent the plates and rings shifting while being placed in the contain ing-jar. The string must, of course, be removed as soon as the element is in the contain- ing-] ar. 2. Place cells in position on battery stands. 3. Scrape the lead lugs before connecting up, so that both surfaces pre- sent a bright metallic appearance. 4. See that all bolt connectors are well screwed up, otherwise resistance and consequent heating Avill result. Always be sure that the cells are con- nected up in series ; i.e., positive of one cell to negative of the next. Figs. 6 and 7. one or more negative plate than positive in every cell. The negative (pole) plates are of a grayish color, and the positives are gen- erally light brown when new. The free pole at one end of the series will, in consequence of this, be a positive, that of the other end being a negative. 5. When all the cells are connected up in this manner, the electrolyte may be added, provided the charging current is available. The electrolyte mm t never be allowed to stand for more than two (2) hours in new cells before the charging is started. To make .Acid. 6. " Oil of Vitriol "is of much higher specific gravity than that required for the cells, and must never be used unless diluted. It must be free from impurities, such as arsenic, nitric or hydrochloric acid, and must be diluted with pure water to a specific gravity of twelve hundred (1,200), or 25° Baume, as shown by the hydrometer at a temperature of 60° Fahrenheit. In mix- ing the electrolyte, the acid must always be poured into the water, and never the water into the acid. 7. Always see that the electrolyte is cold before pouring into the cells. It is advisable to mix it at least twelve (12) hours before using. 8. The initial charge must be commenced immediately the cells are filled at about one-third Q) of the normal rating for four (4) hours, then increased to the normal current, at which it should be continued for twenty (20) con- secutive hours, if not longer, until the positive plates are of a dark brown color, and the voltage of the cells are 2.6 volts per cell while charging at the normal rate. If possible do not stop charging at the above period, but continue at a lower rate, gradually reducing the charging current until one- fourth (I) of the normal rate is reached, at which rate it should be continued until the cells reach a voltage of 2.6 volts per cell. 9. In subsequent charges and in general use, it is only necessary to charge until the voltage is 2.5 per cell while charging. It is advisable to charge the cells once a week until the voltage per cell is 2.5 volts on about one- third (i) the normal charging rate. 10. The cells maybe discharged down to 1.8 volt per cell, on closed cir- cuit at normal rate ; but their efficiency and life will be improved if the discharge is not regularly carried to thi's point, but is stopped before the UNPACKING, SETTING UP, AND USING. 563 cells become so nearly emptied. The cells must never be allowed to stand idle if more than seventy-five (75) per cent of their capacity has been used. 11. If a battery is to remain idle for a long time, it should first he fully charged and then given a recharge, enough to bring it to a boil, at least once a week. If, for any reason, this weekly charge is impossible, the battery should be thoroughly charged ; then syphon the electrolyte from each cell, heing sure to refill each cell with water immediately thereafter. Then start discharging the battery at its normal rate, which will only last a few hours ; then decrease the resistance in the battery circuit until it is almost short- circuited. The battery should be in the water about thirty-six (36) hours, the acidulated water being then drawn off. 12. To put the cells in commission again, replace the electrolyte, and pro- ceed as per instructions for first charge. ' 13. The specific gravity of the electrolyte should be twelve hundred (1,200), 'or 25° Baume, when the'cells are fully charged. 14. Always see that the plates are well covered with electrolyte. 15. The cells should be individually tested at regular intervals with a low- reading voltmeter and a hydrometer. It is very essential that the voltage of each cell should be recorded at the end of every charge and discbarge. If any cell reads low, give it immediate attention, as otherwise serious results may ensue. Partial list of Manufacturers of Storag-e Batteries. United States. Electric Storage Battery Company, Philadelphia, Pa. Electro-chemical Storage Battery Company, New York, N. Y . American Battery Company, Chicago, 111. Willard Electric and Battery Company, Cleveland, O. Gould Storage Battery Company, Depew, N. Y. England. The Electrical Power Storage Company. Chloride Electrical Storage Syndicate. D. P. Accumulator Company. Crompton & Howell. Epstein Company. France. Societe Anonyme pour le Travail Electrique des Meteaux. Germany. The Tudor Company. Battery for Private Residence. The battery should have a capacity to supply one-half the lamps wired for eight or ten hours on one charge. The average use is much less, and the battery will supply ordinary calls for two or three days on a charge. The 'capacity of the engine and dynamo should be equal to that ol the bat- tery at the eight-hour discharge rate, so that on special occasions, when all the lamps are needed, both dvnamo and battery can supply current together. The best method of installation Avill be dictated by local conditions, but, up to 200 lamps capacity, a shunt-wound dynamo that will give 150 volts pressure is probably the best. The best method of regulating a plant of this small capacity is by counter E.M.F. cells, placed in series between the battery and lamps, being all in when the battery is fully charged, and cut out one at a time as the pressure falls. Counter E.M.E. cells are simply unformed lead plates, mounted in the same manner as are the regular plates, and placed in opposition to the regu- aThe use'of counter E.M.F. cells enables one to charge the battery at the same time that lights are being supplied from it, as the counter E.M.* . cells will absorb the extra pressure necessary for charging. 564 STORAGE BATTERIES. "Where it is desired to charge the hattery at the same time that lamps are operated, 18 counter E.M.E. cells are necessary; hut where the hattery can he charged when lights are not in use, as is easily done in the ordinary house, hut 7 counter E.M.F. cells are necessary. The cuts following show two methods of controlling the pressure, the first diagram being with the use of counter E.M.F. cells as described above, while VOLTMETER <1 T 6 6 DYNAMO RHEOSTAT Jl!L. A oL-96Q \ 0---0 l-^I O P ° O?" * B -°l£%o£>' ° ° L*sC — o 9 p^p— -p -oLiv6o o o o L*6D -- Q-? p 9 o- — e fW o ■; o>J 54^i! DIAGRAM of CONNECTIONS FOR THE PEQUOT LIBRARY, SOUTHPORT, CONN. THE E.S.B.Co. PHILA, PA. Fig. 8. BATTERY FOR PRIVATE RESIDENCE. 565 tlie second is done by cutting in and out the end cells. Both diagrams show the proper arrangement of all controlling and indicating appliances for a switchboard. D CELLS BATTERY The method of regulating by cutting in and out end cells is used only in plants large enough to afford an attendant, as the end cells are charged and discharged to different degrees, and need attention to keep in normal con- dition. Useful appliances for isolated batteries are underload switches, for auto- matically cutting out the battery when it has discharged as low as is safe, and overload switches for preventing discharge at greater than a safe rate, say in case of a short-circuit on the line. Both devices open the main bat- tery circuit and prevent trouble. 506 STORAGE BATTERIES. Storage Battery in I^arge Isolated Plants. A large isolated plant, such as is now used in large office buildings, is prac- tically a central station with a prescribed territory; and the battery is, in this case, an auxiliary, and used for furnishing the peak of the load, and in some cases all the load, during such periods of the run as it is within the capacity of the battery. Experienced judgment is necessary in properly proportioning a storage battery to any plant; and it is necessary to know a number of points regard- ing its particular features, such as the following ; viz.: — 1. Nature of load and duration. 2. Maximum, minimum, and average loads. 3. Size and type of generating units. 21000 20000 !- ._. ■ -"/■ \ 1 19000 13000 17000 16000 rn ;i i — 14000 13000 - -A - f "S 12000 ; ■; / t —A 1 0000 9000 8000 7000 t ■ 1 I p h ■ \7 '"7 / ' -1 '- 5000 4000 3000 2000 1000 - 0 1 / .. -■ -. : _■■:■ 0 ... '■ ■ i WEEK DAY LOAD Fig. io. Where it is possible to do so, a load diagram constructed from actual records of output is in all ways the best, as it will include the information necessary, excepting data as to generators and voltage. Even in new plants it is nearly always possible for the designing engineer to construct a load diagram that will serve well for proportioning the battery. 6000 ::: 1 ■ -1 — 1 5000 - — ^"T ! -b_: 4000 -'■ | ■■ ■ 1 ■ 1 it : 3000 -{- r~ | — ^ . 1000 -f — 8 ^ -'|_!l_ " 4 -rr '■■'■ 0 : : " ' \- ' ■ ' 1" ■ i i i 1 1< 5 6 1 0 SUNDAY LOAD Fig. 11. STORAGE BATTERY IX ISOLATED PLANTS. •567 Advantages of a, Battery in an Isolated Plant. 1. Generator capacity for the average load is all that is necessary, the bat- tery taking the peak ; and if the low load is within the capacity of the bat- tery, the generating plant may be run at economical loads only, and shut down entirely during the time of low load, providing the battery is then fully charged, thus saving fuel. 2. Lamps may be run on the same lines with elevators or other variable load, the battery providing instantaneous regulation. 3. Greater reliability of plant, and provision for quick supply in case of storms and other sudden calls. 4. Possibility of reduction in pay-roll due to use of battery instead of steam plant and generators. "jr BUS BARS « PROPOSED ARRANGEMENT FOR BELTED BOOSTER, WITH END CELL REGULATION THE ELECTRIC STORAGE B Battery Cliarg-e and Control. In the large isolated plant and in the central lighting station there are a number of methods in common use for operating the battery and controlling its output and pressure. 568 STORAGE BATTERIES. In such plants the dynamos are seldom designed with large enough range in voltage to permit of charging the battery direct to its full pressure, and recourse is then had to the " booster ; " a belt or motor driven dynamo, with its armature in the battery-charging circuit, and its fields being excited from the bus bars, which may be used to supply the excess pressure neces- sary to produce the proper rise of voltage in the line to overcome the counter E.M.F. of the batteries. The booster must have a capacity for the full charging current, and a range of pressure from ten to fifty volts. Following are a number of diagrams of arrangements of batteries in actual use, the diagrams showing relative location of all appliances for switchboards and battery. These diagrams are furnished by the courtesy of the Electric Storage Battery Company of Philadelphia, Pa. Belted Booster; End Cell Regulation. The preceding diagram, Fig. 12, is one of the simplest forms for a plant with no special complications, and explains itself. Belted Booster; Regulation Itj Counter J2.]fl.l?. Cells. The following diagram shows the relative location and arrangement of all controlling and indicating appliances for a battery using a belted booster, and the regulation being accomplished by counter E. M. F. cells as pre- viously described. Fig. 13. Diagram of Connections for Plant consisting of Storage Battery, C.E.M.F. Cells, Compound Wound Dynamo and Belt-driven Booster. TheE. S. B. Co. BELTED BOOSTER. 569 switch board panel for motor driven booster with end cell regulation, for storage battery in large public building note:- on fifteen point voltmeter switch points numbered 58, 59, 60, etc. connect with CORRESPONDINGLY NUMBERED POINTS OF END CELL SWITCH. ON END CELL SWITCH POINT8 NUMBERED 57, 58, 59, ETC. CONNECT WITH CORRESPONDINGLY NUMBERED POINTS OF END CELLS. Fig. 14. 570 STORAGE BATTERIES. Motor- BE riven Booster; End Cell Regulation. Tlie preceding diagram (Fig. 14) gives the layout of the switchboard and all connections for a storage battery in a large public building. DIAGRAM OF CONNECTIONS FOR BATTERY BOOSTER AND BOOSTER DYNAMO IN CONNECTION WITH C. E. M. F. CELLS AS AN AUXILIARY TO AN EXISTING SWITCHBOARD FOR COMPOUND WOUND DYNAMOS. FIG. 15. MOTOB-DBIVEN BOOSTER. 571 Motor-driven Booster; Counter B.M.F. Cell Regulation. The Dreceding diagram shows connections and relative location of appli- ances for the slitchloard for connection to an existing switchboard ; coun- terE m" F cells being used for regulation, with a motor-driven booster for ^NoSg-On Fifteen Point Voltmeter Switch Point s ^^^Mi^' etc connect with correspondingly numbered Points of End Cell Switch. On End Cell Switch Points numbered 57, 58, 59, etc., connect with corre- spondingly numbered Points of End Cells. Yacht Plant. Yachts cannot carry any surplus weight of machinery ; and in order to chlr|fthe Stery^t7s often cut in tWo and the twohalves charged in par- DIAGRAM OF CONNECTIONS OF SWITCHBOARD, FOR PLANT CONSISTING OF STORAGE BATTERIES WITH C.E.M.F. CELLS, AND SHUNT OR COMPOUND GENERATOR. BATTERY IN TWO PARTS, CHARGED AND DISCHARED IN PARALLEL. note: on c. e. m. f. cell SWITCH POINTS NUMBERED 1, 2, 3, 4, ETC. CONNECT CORRESPONDENT NUMBERED POINTS OF C. E. M. F. CELLS. Fig. 16. 572 STORAGE BATTERIES. allel from the regular lighting dynamos, counter E. M. F. cells being inserted to take up the extra voltage of the dynamo, and to be used for regulation when in use on the bus bars. For discharge the cells are again all connected in series, and run as usual. Note. — On C.E.M.F. Cell Switch Points numbered 1, 2, 3, 4, etc., Gonnect with correspondingly numbered points of C.E.M.F. Cells. .AMMETER ' AMMETER i "VOLTMETER AMMETER | j AMMETER | H^SFJ/ -^>'f", "^ a53e D|8CH"AR6E I CHARGE DISOHARS ™3 BBEQ9TM DYNAMO Fig. 17. Diagram of Connections of Storage Battery Switchboard Panel for Yacht " Niagara." The E.S.B. Co., Phila. Plant for Ifaclit Niag-ara. Preceding is the diagram for the connections of battery and switchboard for the above-named yacht. This battery is also charged in parallel and dis- charged in series, as was the last ; but rheostats are here used for equalizing the charging current to the different legs of the battery. FLUCTUATING POWER LOAD AID LIGHTS OUT THE SAME Ol'IAMO CIRCUIT. ky GENERATOR CONSTANT CURRENT BOOSTER, ADJUSTED FOR AVERAGE LOAD ON MOTORS AND ELEVATORS.. SHUNT BOOSTER WITH BEVERSED SERIES WINDING ' Fig. 18. Arrangement of Storage Battery and Booster for Circuits having a Widely Varying Power Load in Connection with Lighting. FLUCTUATING POWER LOAD. 573 "When electric elevators or other appliances taking current intermittently are connected to circuits furnishing current for incandescent lamps, there BATTERY END CELLS STARTING ao.x CONNECTIONS FOR BATTERY, DYNAMO AND BOOSTER FOR FLUCTUATING LOAD. E-S.BX0. Fig. 19. 574 STORAGE BATTERIES. will be a very considerable fluctuation in the ligbt unless means are fur- nished for preventing it. This does not permit of using one dynamo for both services unless a storage battery be connected as a regulator. The diagram on p. 572 (Fig. 18) shows the scheme of such a connection of battery; and the more complete diagram following that gives the actual con- nections and diagram of panel board for an existing plant now being worked in this manner. J 3 / 3' 0 STORAGE BATTERY 1 ( pi TROLLEY WIRE _^/° Hiiiiniiim STORAGE BATTERY O £= > r* 500 VOLTS o J I W M RAIL RETURN Fig. 20. Arrangement of Battery for Street Railway Circuits where Refine- ment of Regulation is not necessary. r [VOLTMETER] !\ > ST T CIRCUIT BREAKER STORAGE BATTERY REGULATION AT DISTANT POINT ON LINE. E.S.B. Co. STORAGE BATTERY FOR STREET RAILWAYS. SlORAfiE BATIERl AS AUXIIIARI FOR POWER PIMX FOB STREET RAILWAYS. Owing to great fluctuations of load on the power-plant of street railways, a storage battery of the proper size and properly connected can be made to assist greatly in the economy of the station. It will maintain a much evener pressure on the circuits. Will take on all overload ; and at the low demand between one and six o'clock a.m. will take all the load on all but special occasions, thus relieving the steam plant and attendant labor. On such occasions, as it may be necessary to shut down the power-plant for a short time, the battery Avill take the entire load for a short period. Battery used for Simple Reg-ulation. The two preceding diagrams illustrate the simplest form of application of a storage battery to street railway circuits. The first is when the battery is placed in the power-house, and in connection with a compound-wound gen- erator ; the two cells shown in shunt to the series winding are needed to prevent the main battery reacting on the generator. The second diagram shows the use of a battery at some distant point on the line where it acts as a regulator of pressure, and at the same time a regulator of load on the engine. Close Reg-ulation, with Battery and Booster. The following diagram is a sketch of an arrangment of a storage battery in connection with a differentially wound booster that will maintain a very close pressure on the lines at all times. With this arrangement, when a heavy load comes on the circuit the cur- rent through the series field of the booster increases the pressure from the battery to the line, thus compelling the battery to assist. As the load de- creases the series field is overbalanced by the shunt field, and the generator then feeds directly into the battery. GENERATOR IUNT FIELD SHUNT FIELD IS CONNECTED IN OPPOSITION TO SERIES FIELD, IN BOOSTER . Fig. 22. Differential Booster for Maintaining Constant Voltage on Rail- way Circuits. Battery for Reg-ulation of Pressure at tRe End of a Long1 Railway Feeder. The following diagram illustrates the use of a storage battery in main- taining a constant pressure at the end of a long railway line, as is done on one of the Philadelphia lines at Chestnut Hill. In this case the booster is located in the main power-house and charges the battery, which is located a number of miles away, through a special feeder at such times as the load is light and power is available at the power-house. 576 STORAGE BATTERIES. BATTERY STATION Fig. 23. Diagram Showing Application of Storage Battery to Electric Traction, Battery Located at a Distant Substation and Acting as a Load Regulator. Generator and Battery can feed the system either separately or in com- bination through main feeder No. 1, a special feeder No. 2 with Booster being used as an adjunct to main feeder, or for independent charging of Battery. The E. S. B. Co., Philadelphia, Pa. STORAGE BATTERY FOR CEXTRAI-STATIOUT USE. All the advantages recited in the preceding paragraphs relating to the use of batteries in small and large isolated plants, and in street railway power, apply equally well to. their use in central lighting stations ; and with some refinements not necessary in railway work, they have been found to make for increased economy of working in every case where they have been in- telligently applied. The Edison Illuminating Companies were the first to develop their use on this side the Atlantic ; and the growth of such use has been steady, and the capacity of batteries has increased to a very great extent since the first Tudor battery was installed in the station of the Boston Edison Company. Different methods of Application of Battery to Central Station Practice. Fig. 24. Circuits of Storage Batteries in Connection with Three-Wire System, Philadelphia Edison Station. STORAGE BATTERY FOR CENTRAL STATION. 577 The three diagrams, Figs. 24, 25, 26, illustrate the straight application of a storage battery to use in a central lighting station for all the regular uses of regulation of pressure and load, etc. The first is the sketch of connections of the plant used in the station of the Philadelphia Edison Company ; the second, that of the plant for the San Francisco Edison station ; the third, that of the recently installed plant of the Chicago Edison Company, the largest by far yet constructed. m Fig. 25. Storage Batteries in Connection with Three-Wire System as used at San Francisco Gas and Electric Co., San Francisco, Cal. The E. S. B. Co., Phila. J + Auxiliary Bus 1 -1- Charging Bus hrt ~: rtit — Charging Bus m Switches C^A. CJ^A <1 1 "jAmmJeters | Fig. 26. Diagram of Connections of Storage Battery for Chicago Edison Co. E. S. B. Co. The two diagrams, Figs. 27 and 28, show the circuits and connections of batteries in the two large substations of the New York Edison Company ; the first is the station at Bowling Green, and the second at 12th Street. The second of these substations is right in the heart of the city, and feeds in all directions into the heart of the network of conductors. The first-mentioned station, that at Bowling Green, is in the lower part of the city, and feeds a large district occupied by the large office buildings, and keeps up pressure at what was practically the lower end of the network. 578 STORAGE BATTERIES. Fig. 27. Battery, Booster, and Feeder Connections, Bowling Green Storage Battery Station. Fig. 28. Battery, Booster, and Line Connections of the 12th Street Station of the New York Edison Company. STORAGE BATTERY FOR CENTRAL STATION. 579 The diagram, Fig. 20, illustrates the method of connecting a storage bat- tery to a three-wire system with the dynamos of full pressure and connected directly across the outside conductors. This method has been in use abroad by the Siemens-Halske Company to some extent, and will make a satisfac- tory three-wire system from one dynamo or more. ] .£^1 J©-, i BOOSTER > ^ „ oQo-^sll-Or. = Fig. 29. Diagram of Connections Showing Application of Storage Battery to Three Wire System with Generators across Outside Wires Only. The E. S. B. Co., Phila., Pa. The diagram, Fig. 30, shows one of the newer applications of the storage battery for use in connection with long-distance transmission, and it is quite similar to the preceding application with the exception that in this case a rotary converter is used in place of the regular generator. The diagrams, Figs. 31, 32, of the Hartford Electric Lighting Company's plant, show a very clever method of using a rotary converter and storage battery on a three-wire direct current system. Fig. 30. Diagram of Connections for the General Electric Co.'s Exhibit, Omaha, Nebraska, Showing Applications of Storage Battery to Three Wire System with Generator across Outside Wires Only. The E. S. B. Co., Phila. The terminals of the direct current side of the rotary are connected to the outside wires of the three-wire circuits, and the neutral is carried back of the rotary, and connected to the middle of the secondary on each of the two or three static transformers. This method works well whether the battery is connected or not. TESTES^ STORAGE BATTERIES. Condensed and rearranged from Article by Carl Hering in "Electrical World." An intelligent test of storage batteries requires a considerable knowledge of such batteries, in addition to the mere capacity to make the proper con- nections and to read the instruments accurately. The conditions of the test are also highly important, and must be well understood if the results are to be reliable. Storage battery tests may in general be separated into two classes; viz. : — 580 STORAGE BATTERIES. FARMINGION RIVER POWER STATION 600 K.W. TWO PHASE 500-VOLT ALTERNATORS A. AMMETER V. VOLTMETER C..B, AUTOMATIC CIRCUIT BREAKER 8, .SWITCHES, Figs. 31 and 32. Connections of Machines and Circuits of Hartford Electric Light Company, showing Special Connection of the Storage Batteiy to Rotary Converters. a. To determine for a purchaser if the hattery fulfills the specifications under which it was furnished. TESTING STORAGE BATTERIES. 581 b. To determine for a maker or prospective investor all the qualities of a battery, including its capacity, efficiency, maximum, minimum, and normal or best rate of working, both as to charge and discharge. The first test should really be included in the second; or, when making it, it will he Avell to carry out as much of the routine of the second test as can be done without excessive cost to the client, and anyway as much as may be necessary to determine the prescribed results. In the second test the operator will necessarily have to determine the con- ditions; and it is therefore highly important that he fully understand the peculiarities of storage batteries and their behavior and working, especially so where two batteries of different makes are to be compared. Following are some of the points to be determined. 1. Whether the battery is for stationary or for portable purposes. 2. Weight of plates, of acid, of containing-cell, of one coupling. 3. Floor space, accessibility for inspection and repairs. 4. Size of plates. 5. Dimensions of containing-cell or box. 6. Rate of charge, — maximum, best, normal. 7. Rate of discharge, — maximum, best, normal. 8. Efficiency at all rates of charge and discharge. 9. Normal rate of charge per unit of plate surface. 10. Normal rate of charge per pound of plates, and per pound of cell total. 11. Normal rate of discharge per unit of plate surface. 12. Normal rate of discharge per pound of plates and per pound of cell total. 13. Curve of rise of voltage at different rates of charge. 14. Curve of fall of voltage at different rates of discharge. 15. Kilowatts capacity at different rates of charging. 1(3. Kilowatts capacity at different rates of discharge. 17. Curve of load value when charging at constant potential. 18. Curve of load value when charging at constant current. 1J. Curve of specific gravity of acid by hydrometer during charge and dis- charge. 1. The specifications of the manufacturer will essentially determine whether the battery is for stationary or portable purposes, except in trials of new ones, in which case the person making the test will be in position to say from his trials for which purpose the battery may be best adapted. Batteries for stationary purposes may, in general, be chosen regardless of weight and dimensions, but for portable purposes size and weight must, of a necessity, be the smallest commensurate with the service demanded. 2. A knowledge of the weight of plates, acid, containing-cell, and one coupling is useful in comparing output per unit of weight with other makes of battery. 3. The floor space required, and accessibility for repairs, often govern the selection of batteries for special purposes ; and good practice would dictate that the cell occupying the least space per unit of output, and the one that was repaired with the least trouble, be selected. 4. The size of plates will determine the output per unit of surface. 5. Dimensions of the containing jar or box must be known, in order that proper space may be laid out for its installation. 6. In order to adapt a battery to the purposes of its use it is highly impor- tant that the maximum and normal rate of charge be known, as the battery is most frequently charged during the idle time, or time of lowest output of some operating electrical plant. It is sufficiently obvious that where a plant is available for but a short time, a battery admitting of a high rate of char- ging is desirable, although not always the most efficient in all ways; whereas, if there is plenty of time, during which the charging may be done, then the battery may be charged at a slower and more efficient rate. 7. A full knowledge of the maximum and normal rates of discharge is of the very highest importance, as on this depends the capacity and good work- in? of the battery. The capacity of • all lead batteries is reduced by hastening the discharge, and this is especially so for batteries having the active material in thick masses, or so disposed that the acid has not free access to it. In batteries having the active material disposed in thin layers, and freely exposed to the action of the acid, the reduction of capacity is not so great. 582 STORAGE BATTERIES. While it may be true that a battery maybe constructed for less cost if made for low rates of discharge, the capacity is so much reduced when dis- charged at high rates, that it seems better policy to construct for high rates of discharge, in which case the battery may be equally well used for dis- charges at low rates, but will not hold a charge quite so long as will the slow discharge battery. Treadwell says 8 amperes per square foot of positive plate is a good rate of discharge. Theoretically, the capacity of a battery depends upon the amount of active material, while the rate of discharge depends upon the amount of surface acted upon by the acid. In most installations where a storage battery is used, it is essential that the battery be capable of a high rate of discharge for a short time, say an hour or two, and it is this fact that governs the selection rather than its capacity, although this latter condition must receive due attention after the rate of discharge is settled. In the United States it is now customary to designate the capacity of a , storage battery by a time rate ; viz., a given battery has a certain capacity, at a full discharge in three hours, and such a capacity at a discharge in five hours, etc., 8 to 19 inclusive. Nearly all these items are determined by cal- culations from the readings of the instruments in use, and need no further explanation here. The following named readings may be taken as the routine of a test. Charge. Time. Amperes input. Volts of charging circuit. Specific gravity of acid by hydrometer. Temperature of room. Temperature of acid. Statement of gasing. Discharge. Time. Amperes output. Volts at cell terminals. Specific gravity of acid by hydrometer. Temperature of room. Temperature of acid. Statement of gasing. General Conditions. Insulation resistance of cell from ground. Resistance of cell between terminals when fully charged and when fully discharged. If there is a storage battery recording wattmeter available it will be use- ful in connection with the readings mentioned above. §OVRCE§ OV CURBEflfT FOR CHARGOO. Current from a battery of storage cells will be found by far the best for testing a cell or cells. Where one cell is under test, four others of similar size connected, two in multiple and two in series, will be found to give good results. Tf current from public circuits, or from a dynamo, is to be used, it should be as steady as possible, of considerably higher voltage, and have a large resistance capable of carrying indefinitely the maximum current in series with the cell. Before starting a test, it is necessary to decide the points at which the battery may be considered charged and discharged, as overcharging and undercharging and light and full discharge make much difference in the results. It is difficult to predetermine a rate at Avhich the battery will be fully discharged in a certain time, and the only way is by trial rates. Even SOURCES OF CURRENT FOR CHARGING. 583 then, no rate can be taken as reliable unless it can be repeated under the same conditions, any variation in result showing that the battery had not recovered from its previous discharge. Charging too long at a high rate will injure the plates, but moderate over- charging with a small current is beneficial to the plates, though it, of course, reduces the efficiency. Charging too little results in increased efficiency but less capacity. Discharging too far increases the capacity, reduces the efficiency, and re- sults in great variations in voltage and a tendency to increase the destructive action on the plates. Discharging too little increases the efficiency but reduces the capacity. Destructive action on the plates determines the limits of charge and dis- charge and inside the safe limits the points of stopping charge and discharge will depend on whether high efficiency or high capacity is deemed the most desirable under the special conditions. The proper stopping point is deter- mined by a preliminary test for a curve of voltage, then the points may be selected between the points of rapid change in pressure. Slow discharge Avill take out more of the charge than a rapid discharge, the latter condition leaving some of the charge in the battery, which may show in the next discharge, and make the results erroneous. If a rapid discharge be followed by a slow one, the capacity for the second test will indicate higher than it ought, in some cases showing an efficiency ex- ceeding 100 per cent. If a slow discharge be followed by a rapid one, then the capacity of the second test will indicate lower than will be the correct result. Destructive action on the plates can only be determined by inspection, which will show other than normal colors, sulphating, buckling, loosening of the active material, etc. A number of discharges may be necessai'y to determine if a certain rate is deleterious. In stating the limiting voltages, it is most correct to state the rise or fall of voltage in percentage of the initial pressure, taking as such initial pressure the reading of voltage a short time after the start to charge or discharge, and when it has become constant. The percentage is not always the same for charge and discharge. For the sake of uniformity, especially in comparing cells, it is best to make all tests with continuous discharge without stop. It is considered best to charge, with constant voltage, but is very difficult to do, as the current varies greatly, starting in at a large amount and reducing to a small amount at the end of the charge. The current may vary through wide limits without much effect on the charging voltage. Varying the charging current by steps will be found to result in more nearly constant voltage, reducing to a lower value when the voltage indicates a ranid rise. Take the time of charge at each rate in order to compute the capacity of charge. It is best to make the discharge at constant current, as that more nearly approaches actual practice. If 'this is not practicable in the circumstances, then the best method is to discharge through a constant resistance. Discharge at a constant current will require the use of a rheostat that can be changed by very small increments, such as a Avater box or carbon plate resistance. The readings will then be the voltage at the cell terminals and the constant amperes, and with a proper rheostat the test is very simple. Discharge through a constant resistance, which, by the way, is seldom an actual condition, owing to heat variations, the calculations become tedious, as they have to be made for each reading, and a careful record kept of the time. A discharge at constant watts would be the most correct method for bat- teries that were to be used for traction, but the calculations and adjust- ments are so troublesome and difficult as to add to the liability to error. In comparing two cells connect them in series for charge or discharge, cut- ting out each one as its work is completed, measuring the voltage at the cell terminals. In a comparison of different cells it is necessary to base the comparison on some common factor, such as the following items, the selection depend- ing on the special conditions to be filled: — Ampere-hours per pound. Watt-hours per pound. Charge and discharge rate in hours. 584 STORAGE BATTERIES. Discharge in watt-hours per pound. Discharge in ampere-hours per dollar of cost. Discharge in watt-hours per dollar of cost. Readings of instruments will he governed as to time by the circumstances of the test and the quality of the apparatus. If the source of current or the rate of discharge is variable, many more readings will be necessary than if they are steady. If the instruments do not respond freely to changes of current many readings will also be necessary on that account. If all the conditions are favorable 15 to 25 readings will he sufficient to give a good average. Betore starting test, take the voltage of the cell on open circuit, as it is some indication of the condition of the cell. During test take occasional readings of voltage from which to calculate the internal resistance of the cell, as follows: first take the voltage of the cell while connected in circuit and working, then take the cell out of circuit and take voltage on open circuit. Connect voltmeter terminals to the lead terminals of the cell, not to the circuit or the couplers. Connect the amperemeter as close as possible to one terminal of the cell, so as to include any leakage. Leakage may he found by connecting one leg of the voltmeter to ground and the other to one terminal of the cell and then the other. The leak, if any, will be found nearest the terminal indicating the least deflection of the voltmeter. Where the circuit is merely switched from the charging source to the dis- charging circuit, it is necessary to reverse the ammeter leads. Calculate efficiencies for ampere-hours and watt-hours, and for mean' volts, as follows: — . . „ . „ Discharge in ampere-hours X 100 Ampere-hour efficiency % = -^ : : Charge in ampere-hours m- ,, , ,„. . ,„ Discharge in watt- hours X 100 Watt -hour efficiency % = 7^r- - : = Charge in watt-hours _„ . .. .. „ Mean volts of discharge x 100 Efficiency of mean volts % = ^ rr — * i_ Mean volts of charge „. „. „. . „ Mean volt efficiency x ampere-hour efficiency Watt-hour efficiency % = — -• Comparing ampere-hour efficiency with mean-volt-efficiency Avill show whether loss in watt-hours is due to polarization and internal resistance, or to leakage and gasing or lack of retaining power of the active material. SWITCHBOARDS. There are two general types of modern switchboards : (1) Those in which all the switching and indicating apparatus is mounted directly on switchboards. (2) Those in which the main current carrying parts are separate or at a distance from the controlling and indicating apparatus. Both of these can be further divided into Direct Current and Alternating Current, and there are numerous and distinct classes under these. Modern switchboards are made of slate or marble panels, each having a definite function. LAYOUT OF SWITCHBOARMS. In laying out buildings for central stations or isolated plants, the switch- board should be located in an accessible place, and have plenty of room both back and in front. In many cases the switchboard can be placed advantageously on a gallery overlooking the machinery. If due considera- tion be given ito the location of switchboard with respect to the machines and feeders which it controls, unnecessary complications and expense can be avoided. Switchboards are now standardized, covering a large range of D.C. and A.C. generators and feeders, although, of course, it is often necessary to meet special conditions, which, however, can be met usually by slight modi- fications from standard. Unnecessary complications and extra flexibility being at the expense of simplicity, are always to be avoided. It would seem unnecessary, for in- stance, in the great majority of cases to have more than one set of bus bars. Plainness, combined with neatness, and symmetry, is much to be preferred, and nothing should be placed on a switchboard which has no other function than ornamentation. If extensions to switchboards are expected, which is usually the case, panels controlling generators should be together at one end of the switch- board, and those controlling feeders at the other end. When total output panels are used, they are placed between the generator and feeder sections. Of course, where switches are controlled at a distance, this rule need not be followed ; but, on the other hand, it is often advisable, in order to sim- plify station wiring, and to save copper in the busses, to intermingle the generator and feeder switches. Even in this case it is desirable to group the generator controlling and indicating devices together and likewise those for the feeders. For ordinary D.C. switchboards 4 feet is little enough behind the panel. In any case, there ought to be a clear space between connections on panels and wall, of 2£ to 3 feet. For large work and most A.C. work it is very often necessary to have 6 to 8 feet behind panels. In the high-tension work of 5000 volts and above, the General Electric Company remove all high-tension apparatus from the face of the board ; the switches being placed in fire-proof compartments of brick or soapstone, and operated mechanically through bell cranks and levers by means of a handle on the panel, or electrically by means of a controlling switch. The instruments are connected to secondaries of current or potential transform- ers, which are placed in some convenient place in connection with the high- tension wiring. This, of course, necessitates more room than the ordinary switchboards require. The main current carrying apparatus can be placed directly behind the controlling board, below in a basement, or under a gallery ; or above in a gallery ; or, if switches are electrically or electro- pneumatically controlled, they can be placed in any convenient place. In locating switches and other appliances, it is usually assumed that dynamo leads come from below, and that feeder wires go out overhead, except in the case of underground feeders, which naturally go out below. COBflTBUCTIOI. Central station switchboards are usually composed of panels about 90" high and 1" thick, and varying in width from 167/ to 36". The panels are 585 5S(5 SWITCHBOARDS. generally in two sections ; the top varying from 60" to 65", and the lower from 25" to 30". The General Electric Company's Standard is 62" and 28" respectively for top and lower part ; the Westinghouse Standard is 65" and 25". The General Electric Company also makes panels 76" high, \\" thick for isolated plants. Each panel is beveled all around on the front edges with a \" to \" bevel. "Where a well finished switchboard is desired, black enameled slate is recommended for circuits of less than 1100 volts. The main current carry- 1. Method of Joining Adjacent Panels. Channel Foot for Switch- board Frame. ing parts are mounted directly on the panel. For higher voltages it is necessary to use marble on account of its higher insulating qualities. Plain slate can be used where a low-priced switchboard is desired for low voltages. There are several different varieties of marble used for switchboards, viz. : blue or white Vermont, pink or gray Tennessee, and white Italian. Marble being a natural product cannot always be matched in shade or markings. The colored marbles do not show so readily as white marbles the effect of oil or grease, and therefore are more suitable for switchboards. Of the colored varieties, the blue Vermont marble can be obtained in the most uniform color. Steel angle bars varying from 1\" x 1£" x \'' to 3" x 2" x \", are ordinarily used for supporting the panels, although in some cases for heavy work, steel channels, tees, or "I" beams are used. The angle bars stand on the floor, to which they are fastened by means of a small foot iron. The panels are bolted to the narrow web of angle bars, and adjacent angles are bolted together through their wide webs (Fig. 1). The panels should be set up on a level strip, which can be of either hard wood, " I " beams, or an inverted channel. The frame-work of all switchboards should be in- sulated from ground when used on circuits of 600 volts or less. In high tension A.C. systems it is neces- sary to ground all frame-work to carry off static discharge and in order to get rid of danger to the Fig. 3. Showing operator should he accidentally touch the frame- Method of Bracing work. For securing the structure in a vertical posi- Switchboard Panel tion, rods with turn buckles for adjustment of length to Wall. are run from the back wall to the angle frame, at or near the top. A " Y " connection can be made to straddle the two angles, and a bolt be put through the whole. The wall end can be secured by expansion bolts or other means. CONSTRUCTION. 587 Circuit breakers should, be placed, if possible, near tbe top of the panel, so that there will be no apparatus above them. Instruments should be placed within convenient view of attendant, and switches and rheostat hand wheels should be located within easy reach. It is recommended that illuminating lamps be left off of switchboards, and that instruments be illuminated from ligbts on the front of the switchboard. The copper bus-bars and connections on the back of switcb boards need careful laying out, with a view to carrying the current economically and without overheating, and above all, in order that there will be no undue crowding, and that they will present a neat and workmanlike appearance. The tendency has been of late to place the busses toward the top of panels, except in the case of small isolated plant switchboards. The switches, cir- cuit breakers, and instruments are connected to busses by means of bare copper strips or insulated wire, bent in the most convenient shape to suit the case. It is not recommended, as a rule, to have long studs on the appara- tus projecting out far enough to connect direct to busses, as the strain on the switch, due to weight of busses, is likely to affect the adjustment of switch contacts. Very often the connection strips are sufficient to rigidly support the busses, but in some cases it becomes necessary to provide insulated supports for carrying them. Copper bars, flat or round, are now practically universal on low-potential boards. Owing to the greater ease in making attachments and in adding capacity the flat bar is to be preferred, and a thickness of g", \", and \" ', with width according to the current carrying capacity required, is convenient. The size of copper bus-bars and connec- tion strips is usually figured on the basis of 1000 amperes per square inch of cross-section. By properly laminating the bars, it is safe to use this basis even for very heavy current. Contact surface should be figured anywhere from 100 to 200 amperes per square inch, according to the method of clamp- ing, bolting, or soldering. In clamping or bolting, steel bolts should be used. Herrick gives the following table as embodying the current practice for central stations, based upon a load factor not exceeding 50%. If figuring on a 100% load factor, the following amperes must be cut in half : — COPPER BAR DATA. From " Modern Switchboards," by A. B. Herrick. Dimensions. Amps. Cir. Mils. Sq. Mils. Ohms per Foot. Weight, per Foot. l xf 433 318,310 250,000 .0000336 .97 li x \" 530 397,290 312,000 .0000269 1.21 H x \" 626 477,465 375.000 .0000223 1.45 if x \" 725 556,400 437,000 .0000192 1.70 li x i" 676 596,830 468,750 .0000179 1.82 H x f" 798 716,200 562,500 .0000149 2.18 if x f" 916 835,600 656,250 .0000128 2.54 2 x t" 1035 954,930 750,000 .0000112 2.92 2i X §" 1154 1,074,300 843,750 .00000995 3.27 2 x¥/ 1222 1,273.240 1,000,000 .00000840 3.89 2J X \" % x f" 1500 1,591,550 1,250,000 .00000672 4.86 1715 1,989,440 1,562,500 .00000537 6.07 0000 B. & S. 257 211,600 .0000505 .64 \" round 305 250,000 .0000428 .76 \" round 426 390,625 .0000273 1.18 \" round 560 562,500 .0000190 1.71 \" round 861 1,000,000 .0000107 3.05 For the sake of securing the best conductivity, as far as possible, all switchboard connections should be worked out of rolled copper ; but it is 588 SWITCHBOARDS. occasionally necessary to use copper or brass castings, although their use should he avoided as far as possible, as the conductivity is always low, vary- ing from 12% to 60% according to mixture. Where necessary to use cast- ings, they should be made of new metal only, and care should be taken to insist upon a standard of conductivity in each piece if it is to be used where such a condition counts. A conductivity of 50% may be considered high and sufficient. The following table from " Modern Switchboards," by A.B. Herrick, gives percentages of mixtures with resulting conductivity as compared with 100% copper : — % % Conduc- % % Conduc- Copper. tivity. Copper. Tin. tivity. 98.44 1.56 46.88 98.59 1.41 62.46 94.49 5.51 33.32 93.98 6.02 19.68 88.89 11.11 25.50 90.30 9.70 12.19 86.67 13.33 30.90 89.70 10.30 10.21 82.54 17.50 29.20 88.39 11.61 12.10 75.00 25.00 22.08 87.65 12.35 10.15 73.30 36.70 22.27 85.09 14.91 8.82 67.74 32.26 25.40 16.40 83.60 12.76 100.00 27.39 100.00 11.45 All minor connections to bus-bars such as switch leads, feeder ends, or in fact any attachments whatsoever, whether bolted to, clamped against, or soldered, should have ample surface contact, not less than ten (10) times (and twenty (20) times is better), the cross-section of the smaller of the two conductors connected, and where the sub-connection is of round-section it should be cup-soldered or " sweated " into a flat lug having the proper amount of surface contact for bolting or clamping to the bus-bar. Cup-soldered conductors should enter the socket from two to three diameters. While all permanent joints of this nature should be soldered, it is sometimes necessary equalizer bus *° make joints that can be easily disconnected, in which case the old-style socket with binding screws may be used, but the con- ductor should be entered from four (4) to ten (10) diameters to make a secure connection. BUS EXCITES* DYNAMOS. The diagram ^ig. 4) and text on a method of ex- citing dynamos from the bus-bars, in starting, was published by W. B. Potter, in the " Electrical Engi- neer." Besides being a very simple method of bus- connecting for excitation, if the equalizing switch, E.S., and positive switch, P.S., are left closed all the time, which can be done without harm excepting when some repairs or changes may be wanted in the dynamo, all equalizing connections are left in circuit all the time, and any dynamo that may be Fig. 4. Excitation of Generators. *i BUS EXCITED DYNAMOS. 589 Fig. 5. Connections of Generator Panels for Direct Current. 300-1800 Amp. G. E. Co. 590 SWITCHBOARDS. t-1 Fig. 6. Switchboard Panel for One Three-phase Alternating Current Gen- erator, to 2500 volts. G. E. Co. wh — 1 0=3 m 1 (fl II w~~ HI ORMER I tia ~r Fig. 7. Switchboard Panel for One Single-phase Alternating Current Gen- erator, to 2500 volts. G. E. Co. BUS EXCITED DYNAMOS. 591 | IffjOjl § 5s I Fig. 8. Diagram of connections for switchboard of main power station Manhattan Railway Co., L. B. Stillwell, Cons. Engr. 592 SWITCHBOARDS. running will then take its proper amount of current through its series coils and will, therefore, compound more nearly as it was designed to do, than if all the load is on the series coil of the running dynamo. If greater sim- plicity is desired, the equalizing switch, E.S., and positive switch, P.S., can be one double-pole switch, and the negative switch, N.S., a single pole. Leave the double-pole switch closed all the time, and throw the machine in and out with N.S. Mr. Potter says : — By reference to the accompanying diagram, it will be seen that by closing the positive switch, F.S. (the equalizer switch, E.S., being closed), the series coil of the generator to be started is connected in parallel with the series coils of generatoi's in operation, thus separately exciting the field of the generator to be started. The field switch, F.S., being closed, the voltage is then adjusted by the field resistance to correspond with that of tbe bus, and the more easily so, as by this method there is secured a variation of voltage corresponding to that due to changes of load on the over-compounded generators in operation. This method also insures the polarity being at all times the same as the other generators. On closing the negative switch, N.S., and reducing the resistance in the shunt field, the generator takes up its load smoothly and without the violent fluctuation usually caused by connecting the series coils after the full potential has been developed by the shunt field only. It is not necessary to show here all the standard forms of switchboard, or the appliances that are used with them, as changes take place so often that any article pictured or described is apt to be out of date in a very short time. A few diagrams showing standard arrangements that are not subject to much change follow. 1 have included the diagram of general arrange- ment of switchboard connections of the great plant of the Manhattan Elevated Railway of New York, as being very simple and of considerable interest. -A.MC SWITCHSOAIlI>S. This line of switchboards represents an entirely different construction from that of ordinary switchboards. Extra flexibility makes it desirable, and small currents make it possible, to use plug connections instead of the ordinary type of switches. The function of arc switchboards is to enable the transfer of one or more arc light circuits to and from any of a number of generators. This trans- ferring is sometimes accomplished by means of a pair of plugs connected with insulated flexible cable; sometimes by plugs without cables, which bridge two contacts back of the board, or by a combination of cable plugs and plugs without cables. The type using plugs without cables is pref- erable, because danger is eliminated, which would otherwise be possible to attendant, due to contact with exposed or abraded cables carrying high- potential current. Below is a cut of the G. E. Co. Standard Carrier bus type of Arc Board with description. The accompanying illustration shows an arc switchboard of the General Electric panel type, arranged for three machines and three circuits. The vertical rows of sockets are lettered and the horizontal numbered. The ends of the vertical bars are connected to the machines and circuits. Each of the bars is broken in three places, and the machine may be connected to its circuit by plugging across these breaks, thus making the bar continuous ; by removing any pair of plugs the machine may be disconnected. Cll, Ell and Gil are ammeter jacks, and are used in connection with two plugs connected with a twin cable, for placing an ammeter in. the circuit. The six horizontal bars are for the purpose of transferring a machine or a feeder to some circuit other than its own. Each horizontal bar is pro- vided, at one side of the panel, with a socket (A3, A4, A5, A7, A8, and A9) by means of which it can be connected with the horizontal bar on the adjoining panel. All ordinary combinations can be made by means of the bars and plugs ; but cable plugs are provided with each panel, so that when necessary, machines and feeders can be transferred without the use of the bar. These plugs and cables are intended for use only in case of an emergencv. To run machine No. 1 on feeder No. 1, insert plugs in BIO, CIO, B6, C6, SWITCHING DEVICES. 593 B2, and C2. To shut down machine No. 2, and run feeders Nos. 1 and 2 in series on machine No. 1, insert a plug at C5, D5, C7, and D7, and remove plugs at C6 and DG ; this leaves two circuits and two machines in series. Short circuit machine No. 2 by inserting the plug at E7. Cut out machine No. 2 by removing the plug at D10 and E10. Take out plug at D7. SWITCHOG DEVICES. Switching devices in connection with switchboards can be divided gener- ally into two classes, viz. : 1. Switches. 2. Automatic circuit breakers. 594 SWITCHBOARDS. Pig. 10. Gen. Elec. Oil Break Switch, 5000 volts, 300 amcs Opened and Closed by Hand. <*IUPS., CASE REMOVEI Fig. 11. Gen. Elec. Co. Oil Break Switch Opened and Closed by Hand, SWITCHBOARD DEVICES. 595 Switches for low voltage and small current are of the same general form, though differing in details. In the main they consist of a blade of copper hinged at one end between two parallel clips, the other end of blade sliding into and out of two parallel clips. The clips are joined to copper or brass blocks to which the circuit is connected. There seems to be little uniformity among manufacturers regarding the cross-section of metal and surface of contact to be used. Perhaps a cross-section of metal of one square inch per 1000 amperes of current capacity is as near to the common practice as any, and a contact surface of at least one inch per 100 amperes or ten times the cross-section of metal is also common practice, but will depend somewhat on the pressure between surfaces. Auxiliary breaks are demanded by the National Code for currents exceeding 100 amperes at 300 volts, and "quick-break" switches are now quite common for pressure as low as 110 volts, although not in any way more necessary for that pressure than should be " quick-make" switches. 596 SWITCHBOARDS. The rules on switch design issued by the National Code cover the require- ments well, and they must be followed in order to obtain or retain low insurance rates ; all switches must meet the requirements. See index for " National Code," and refer to section on " Switches." Blades, jaws, and contacts should be so constructed as to give an even and uniform pressure all over the surface, and no part of the surfaces in contact should cut, grind, or bind when the blade is moved. The workman- ship should be such that the blade can be moved with a perfectly uniform motion and pressure, and the clips and jaws should be retained so perfectly in line that the blades will enter without the slightest stoppage. For pressures above 1000 volts, practice varies among the different manu- facturers. The General Electric Company makes a switch in which the cir- cuit is ruptured in oil. In the type designed by the Westinghouse Co. de- pendence is placed upon the arc being ruptured in open air by drawing it through a wide break. The Stanlay Co. has devised a switch which is designed to rupture the arc by means of a sliding shutter, Avhich intercepts i the arc when the contact is broken. For non-inductive loads of small power and up to 2500 volts, any good form of quick-break switch can be satisfactorily used. Attached are shown a few types of high-potential switches. AUTOMATIC CIMCXJIT BMEAKEIl§. Automatic breakers are devices which have as an integral part an auto- matic trip which opens the circuit when the flow of current exceeds a pre- AMPERES A B | C D E -^d^oir- -£- ~~"~C J:i_ 28 2]4 + '« Fig. 13. One Form of Circuit Breaker. 1800 to 10000 Amperes. G. E. Co. determined limit. Many types are now made, some with carbon secondary breaks ; but a very successful type is one early introduced by the G. E. Co., with the magnetic blow-out principle applied to extinguish the arc. Illus- trations follow of one of the main sizes and a table for the various adjust- ments of the same. For mean high potential circuits the Westinghouse Electric & Mfg. Co. has devised the instrument shown in the following cuts and diagrams (Figs. 15 and 16) : — The circuit-breaker consists of two hardwood poles, one being longer than the other, mounted upon a marble base, to which are secured the terminals to which the main leads or wires are connected. The poles are connected by a hinge, so that their extremities are in line at the upper end. On the upper end of each pole is mounted a copper sleeve supporting a round carbon contact block with a hole through its center. The longer pole is provided with spring jaws or clips so that it may be quickly and easily attached to, or detached from, the terminals on the marble base. The short pole has a flexible wire running through its interior ; this wire is connected to the copper sleeve at the upper end of the short pole and to the lower clip terminal on the long pole. The sleeve at the upper end of the long pole is AUTOMATI C CIRCUIT BREAKERS. 597 Amperes. AVide Open. Closed. When See. Contacts Touch. A 2 § b'o ! TlT 8 32 7B!T 77 D E r3e F D 150- 2000 ftof 1 i7sto£ 1800- 3000 11 Stoi 1 2000- 6000 1* |to* 1 2000-10000 I Mi H 1 to | 1 NOTE — B is dimension when parts are new. First, Adjust E. Second, Adjust Brush Tension. Third, Adjust C. FIG. 14. Dimensions for Adjusting Mlv Circuit Breakers. connected to the upper clip terminal. Thus, these connections practically make the sleeves at the upper ends of the two poles the terminals of the apparatus. The poles being removed from the base, a wire is inserted through the hole in the carbon tip at the upper end of the short pole, and secured to the High Potential Circuit Breakers, Made by Westinghouse Electric and Manufacturing Company. Fig. 15. ) to 15000 Volts. Fig. 16. 20000 to 40000 Volts. copper sleeve by a screw and washer. The other end of the fuse is p„. through the carbon tip on the long pole, and secured to the copper sleeve by a cam-shaped lock. The length of the fuse should be from 6 to 10 inches. The poles, after being fused, are placed in position by taking hold of the lower end of the long pole. When the fuse blows, the short pole is released by the action of the spring at the lower end, and falls away from the station- 598 SWITCHBOARDS. High Potential Circuit Breakers Made by Westinghouse Electrical and Manufacturing Company. SK, '•A ^ a ^rv. ~:- M ;«y J PRINCIPAL DIMENSIONS, 6000-15000 VOLTS Figs. 17 and 18. Figs. 19 and 20. ary pole, thus making a very long break'. The lock cam has a long string attached to it, by means of which the fuse can be released if desired, thus causing the short pole to drop in the same manner as when the fuse blows. This feature permits the device to be used as a switch. REVERiffi CUK«E1¥T CIRCUIT BREAKER§. For large installations of electrical transmission, where it is highly im- portant that continuity of service shall be maintained, it is good engineering to use two separate lines of conductors. In such cases it is usual to keep both circuits connected so that in case of trouble on one of them its fuses or circuit breaking devices will cut it out, leaving the clear line to carry the load. An examination of the following diagram Avill explain the utility of the reverse current circuit breaker. Let a and ax be circuit breakers at the dynamo end of the two lines, and b and bY reverse current circuit breakers at the far end of the same. Should a short circuit occur as at x on the main line, it is plain that current will rush in both directions from the dynamo, by way of the main line and by way of the auxiliary line and the far end of the main line, in which portion the direction of the current will be the reverse of what it was ordinarily. Under this condition it is obvious that all the circuit opening devices would operate, and the auxiliary line would be of no effect in maintaining continuity of current. Now, if circuit breakers of such a design that they will open on a reversal of the direction .-..k REVERSE CURRENT CIRCUIT BREAKERS. 599 b , [ 7> -Q-J -B— Pig. 21. Diagram Showing Use of Reverse Current Circuit Breaker. of the current through them, be placed at the far end, as at b and bx then the main circuit breakers, a, a, will open, as Avill also the reverse current circuit breakers, b, b, thus leaving the auxiliary line intact. Of course a short circuit on tbe auxiliary line will operate in a similar manner. The following diagram shows the connections of the reverse current circuit breaker at Buffalo as designed by the General Electric Co. An Pig. 22. The Circuits of a Reverse Current Circuit Breaker Set Showing How a Direct Current Motor is Used with Alternating Currents to Distinguish between Power Passing in One Direction and Power Passing in the Other Direction in the Line. Fig. 23. The Circuits of a Time Element Relay Circuit-break- ing Set. 600 SWITCHBOARDS. ordinary fan motor is introduced by means of a transformer into the line, and acts to operate a relay on the shunted circuit breaker, a reversal of the current reversing the motion (or pull) of the fan motor armature, and closes the relay contacts as shown. Inline El«*iii«*Mt for Circuit J$r«>ak<»rs. — Where circuits are loaded with large synchronous or induction motors and other devices liable to produce short circuits on the system when out of step or started too sud- denly, it is not only necessary to protect the local or feeder circuit with circuit breakers, but in order to prevent the operation of all the protecting devices between the one in trouble and the dynamo itself, it is found advis- able to introduce a time element or adjustable delay on all the main circuit breakers. This device must allow the circuit breakers farthest from the station to be adjusted so they will open first, and all the intermediate devices must have variable or graduated adjustments, say for opening after three seconds, and the main circuit breaker at the power house itself will open last of all, say in five seconds. Mr. L. B. Stillwell devised an instrument for this purpose, and it has been widely adopted. Both the Westinghouse Co. and General Electric Co. have adapted this time element device to the circuit breakers in use at Niagara Falls, and the following cut shows the arrangement by General Electric Co. diagramatically. The instrument is composed of a simple clock move- ment, the wheels of which are prevented from turning by a pawl which may be lifted out of place by either one of two relay magnets connected by transformer in the main line. The lifting of the pawl allows the clock wheels to revolve and close a relay circuit connected with the circuit breakers which promptly open. The clock movement can be adjusted to close the local circuit in any desired time ; and in case the clock is started on a short circuit, which is off before the lapsing of the time period, the pawl drops, and the movement returns to its original position. LIGHTNING ARRESTERS. IIGHTIU'U AKHESTEM« IX GEItERAI. (From pamphlet by Westinghouse Electric & Manufacturing Company.) Tlie Timction of liig-litiiiitg* Arresters.- — The function of a lightning arrester is two-fold. It should provide a path to earth offering the least possible resistance to the passage of static discharges, and it should avoid interruption of the service. The latter, though a negative function, is one of primary importance. In the early days of electrical industry it was found that lightning dis- charges from overhead wires would pass more readily to ground over a small air gap than through coils or even long lengths of straight wire. Numerous arresters based upon this principle were constructed and placed in practical use. The simplest form of these is the old saw-tooth spark-gap arrester which is still used for protecting telegraph and telephone lines. But a great difficulty arose with gap arresters when used on electric lighting, railway or power circuits, owing to the fact that the dynamo cur- rent followed the lightning discharge, establishing thereby a short circuit which would melt the dynamo fuses" and thus interrupt the service. With the object of overcoming this trouble various arresters were der vised that would automatically interrupt the dynamo short circuit. At first this interruption was accomplished by simply placing fuses in the lightning arrester circuit, thus making it necessary to renew the fuses after each discharge. This method was obviously unsatisfactory. Arresters were then devised which would automatically interrupt the arc and then immediately adjust themselves for another discharge by means of moving parts ; the latter, however, proved to be the cause of considerable annoy- ance, and experience demonstrated that the arc rupturing arresters were uncertain in action and hence unreliable. Recognizing the importance of the problem the Westinghouse Electric & Manufacturing Company undertook a series of extensive theoretical and practical investigations, with the object of devising arresters which would offer a low resistance path to ground for disruptive discharges, and at the same time operate automatically and repeatedly without moving parts and without interrupting the service. — PI A- ,-p, n — , 11111111 ' y LINE iiiiiiii; u GROUND Of LINE Fig. 1. Diagram Showing Electrical Connections for A. C. Lightning Arresters. The results of these investigations, which extended over a period of sev- eral years, are embodied in the Wurts Non-arcing Lightning Arresters. With a non-arcing arrester the dynamo current does not continue to fol- low the discharge ; the apparatus is not left unprotected for an instant ; the instrument does not deteriorate ; it is entirely automatic in action, and will handle frequent and persistent discharges with perfect facility. For systems of distribution, with their various motors, converters, and other appliances, a liberal allowance of line arresters judiciously distributed over the lines is essential for securing adequate protection. Much, how- 601 602 LIGHTXIXG ARRESTERS. ever, depends upon the local conditions, such as the character of the soil with reference to the ground connections, and severity of lightning dis- turbances, the grade of insulation to be protected, the voltage of the circuit and the surroundings with reference to telegraph and telephone wires. EiG. 2. Double-Pole Non-Arcing Metal Lightning Arrester. Type " A." (For Station Use.) Fig. 3. Unit Lightning Arrester, Type THE ]\OI¥. ' C," Showing Cylinders in Place. -arcihtg metal ARRE§fEll. The non-arcing metal lightning arrester for alternating current circuits is based upon the discovery made by Mr. A. J. Wurts that an alternating current arc cannot be maintained over a short air-gap when the electrodes consist of certain metals and alloys thereof. Types " A " and " C " arresters, described below, are of the non-arcing metal type. THE NON-AROIXG METAL LIGHTNING ARRESTER. 603 The Type "A" Arrester. — The construction of this arrester can be best understood by reference to Fig. 2. It will be noted that there are seven independent cylinders of non-arcing [metal placed side by side and separated by air-gaps. The cylinders, which are mounted on a marble base, are knurled, thus presenting hundreds of confronting points for the discharge. The dynamo terminals are connected I I Figs. 4, 5. Double-Pole Non-Arcing Metal Line Arrester — Type " C." to the end cylinders, and the middle cylinder is connected to the ground. The arrester is, therefore, double pole, that is, one arrester protects both sides of the circuit. When the lines become statically charged the dis- charge spark passes across between the cylinders from the line terminals to the ground. The non-arcing metal will not sustain an arc or become fused by it ; hence with an arrester constructed of this material all possibility of vicious arcing and short circuits is avoided. There are no moving parts, no coils to impede the passage of the lightning discharge, and in fact nothing that requires either adjustment or inspection. These arresters are made in units for 1000 volts ; for 2000 volts two units are connected in series, and for 3000 volts three are connected in series, as indicated in the diagram, Fig. 7. GROUPS— Fig. 6. Lightning Arrester for 15,000 Volt Circuit — Type " R 604 LIGHTNING ARRESTERS. Tlie Type " C " Arrester. — This is similar to type "A," but instead of being mounted on marble it is inclosed in a weather-proof iron case for line use. The cylinders are placed in porcelain holders, as shown in Figs. 3 and 4. The arrester complete in the iron case is shown in Fig. 5. The! method of connecting type " C " arresters to circuits of different voltage is also shown in Fig. 1. The Type " JR. " Arrester. — Our standard form of arresters for pro- tecting alternating current high potential power transmission circuits is shown in Fig. G. A diagram illustrating the method of connecting the arresters and choke coils for various voltages is given in Fig. 7. FIG. 1 DYN. LINE Fig. 2 I 3,000 VOLTS 1 5,000 VOLT8 -=±=- Fig. 5 DYN. LINE o r^ r^ o o o u u o i Fig. 3 DYN. LINE FlG. 4 DYN. LINE i1 ■? Y ■? 8,000 VOLTS i 10,000 VOLTS FlG. 7. Diagram Showing Pyramidal Form of Connecting Lightning Arresters and Choke Coils for Various Voltages. Explanatory Note — Each circle represents a choke coil. Each rect- angle represents one unit (type " C ") non-arcing metal lightning arrester. . CHOKE COILS FOR A. C. CIRCUITS. 605 Sub-Fig. 1, four coils in series and one and one-half unit arresters between line and ground. Sub-Fig. 2, live coils in series and two and one-half unit arresters between line and ground. Sub-Fig. 3, six coils in series and four unit arresters between line and ground. Sub-Fig. 4, six coils in series and five unit arresters between line and ground. Sub-Fig. 5, six coils in series and seven unit arresters between line and ground. Plan View of Lightning Arrester Racks, Showing Unit Lightning Arresters and the Connections for Each Voltage. CHOKE COE3LS JOB A. C. CIMCUITi. A lightning discharge is of an oscillatory character and possesses the property of self-induction ; it consequently passes with difficulty through coils of wire. Moreover, the frequency of oscillation of a lightning dis- charge being much greater than that of commercial alternating currents, a coil can readily be constructed which will offer a relatively high resistance to the passage of lightning and at the same time allow free passage to all ordinary electric currents. Any coil will afford a certain amount of impedance to a disruptive dis- charge. Experience has shown, however, that there is one form which offers at once the maximum impedance to the discharge with the minimum resistance to the generator current. Choke coils of this type connected in the circui*, when used in connec- tion with non-arcing lightning arresters, offer a very reliable means of pro- tecting well-insulated apparatus against lightning. This arrangement is particularly suited for protecting station apparatus in power transmission systems. Coils can, however, be used to advantage on the line for the pro- tection of the more expensive translating devices. 606 LIGHTNING ARRESTERS. Tests made under actual working conditions indicate that for ordinary commercial voltages effective protection is obtained with four choke coils in series in each wire, with four lightning arresters intervening, as shown in Fig. 10. This diagram also shows the method of connecting the coils and arresters to one end of a three- wire transmission system. Fig. 9. A. C. Choke Coil. Fig. 10. One end of a 2000-Volt 3-Wire Power Transmission System Showing Bank of Choke Coils and Lightning Arresters. GROUND CONNECTIONS. 607 AMBEITEBS JOB ». C. CIBCUIT§. The non-arcing metal arresters described above are not suitable for use on D. C. circuits, but a non-arcing D. C. arrester has been devised bv Mr. . J. Wurts. The principles upon which this arrester is designed are based upon the following facts : — First. A discharge will pass over a non-conducting surface, such as glass or wood, more readily than through an equal air-gap. Second. The discharge will take place still more readily if a pencil or carbon mark be drawn over the non-conducting surface. Third. In order to maintain a dynamo arc, fumes or vapors of the elec- trodes must be present ; consequently if means are provided to prevent the formation of these vapors there will be no arc. 0 ^-ttv e • •e o • o 9 o fill o o • • o 9 J 0 m i 0 b Fig. 11. Non-Arcing Railway Lightning Arrester, Type " K." (For Station Use.) TJie Type " BL " .Ai'restier. — The illustration, Fig. 11, shows the type "K" arrester for station use on D. C. circuits up to 700 volts. The instrument is single pole, and consists of two metal electrodes mounted upon a iiguum-vitge block, flush with its surface. Charred or carbonized grooves provide a ready path for the discharge. A second lignum-vitse block fits closely upon the first block, completely covering the grooves and electrodes. Disruptive discharges will pass readily between the electrodes over the charred grooves, which act simply as an electrical crack through the air, providing an easy path. The resistance between the electrodes is more than 50.000 ohms, so that there is, of course, no current leakage, but it should not be understood that the lightning discharge passes through this high resistance — it leaps over the surface of the charred grooves from one electrode to the other exactly as it would if there were but a simple air-gap. The presence of the charred grooves simply makes the path easier. There being no room for vapor between the two tightly fitting blocks, no arc can be formed, hence the arrester is non-arcing. GROMD COMECTIOIS W&M. JL. C. JkJ¥l> I>. C. Too much importance cannot be attached to the making of proper con- nections from the arrester to ground, which should be as short and straight as possible. It is obvious that a poor ground connection will render inefficient every 608 LIGHTNING ARRESTERS. 3rs to drive the static elec- t that we not only should sction, but also thoroughly latural conditions. 11 lightning arresters may 'le six feet square directly s been reached ; second, ushed coke or charcoal of No. 1G tinned copper No. 0 copper, securely ver the ground plate effort made with choke coils and lightning tricity into the earth. It is, therefore, im understand how to construct a good groum appreciate the necessity of avoiding unfavorable n A good ground connection for a bank of station he made in the following manner : First, dig a ho] under the arrester until permanently damp earth h cover the bottom of this hole with two feet of < (about pea size) ; third, over this lav 25 square fee plate; fourth, solder the ground wire, preferabl across the entire surface of the ground plate; fift with two feet of crushed coke or charcoal ; and sixth, till in the holewith earth, using running water to settle. The above method of making a ground connection is simple, and has been found to give excellent results, and yet, if not made in proper soil, it would prove of little value. Where a mountain stream is conveniently near, it is not uncommon to throw the ground plate into the bed of the stream. This, however, makes a poor ground connection, owing to the high resist-' ance of the pure water and the rocky bottom of the stream. Clay, even when wet, rock sand, gravel, dry earth and pure water are not suitable materials in which to bury the ground plate of a bank of lightning arresters. Rich soil is the best. It is therefore advisable before installing a bank of choke coils and lightning arresters to select the best possible site for the lightning arrester installation, with reference to a good ground connection. This may often be at some little distance from the station, in which case it is of course necessary to construct a lightning arrester house. Where per- manent dampness cannot be reached, it is recommended that water be sup- plied to the ground through a pipe from some convenient source. jLicjHTnriarc} arresters eom direct cerrest. (From pamphlet by General Electric Company.) Some years ago Prof. Elihu Thomson devised a lightning arrester based on the principle that an electric arc may be repelled by a magnetic field. In this device, the air-gap, across which the lightning discharges to reach Fig. 12. Type "A" Arc Station Arrester. the ground, is placed in the field of a strong electro-magnet. When the generator current attempts to follow the high potential discharge, it is instantly repelled to a position on the diverging contacts where it cannot be maintained by the generator. LIGHTNING ARRESTERS FOR DIRECT CURRENT. 609 The magnetic blow-out principle has been employed in the construction of a complete line of lightning arresters for all direct current installations, and in more than ten years of service magnetic blow-out arresters have always been effective in affording protection to electrical apparatus. In designing lightning arresters for the protection of bigh-voltage alter- nating current circuits, however, different conditions have to be met, since high-voltage arcs are not readily extinguished by a magnetic blow-out. In a recently designed lightning arrester for alternating current circuits, metallic cylinders with large radiating surfaces are found to so lower the temperature of the arc that volatilization of the metal ceases and the arc is extinguished. Fig. 13. Type " AA" Arc Station Arrester. The variety of these lightning arresters provides for the protection of all forms of electrical apparatus and circuits. The Type "A" Arrester is manufactured for the protection of arc lighting circuits, and is in extensive use throughout the world. Its construction includes a pair of diverging terminals mounted on a slate base with an electro-magnet connected in series with the line. The magnet windings are Fig. 14. Type "A," Form " C," Lightning Arrester, in Iron Box for Line Use 610 LIGHTNING ARRESTERS. of low resistance, and therefore consume an inappreciable amount of energy with the small current used for arc lighting, although they are always in circuit. The single Type "A" Arrester is suitable for circuits of any number of series arc lamps not exceeding seven ty-nve. For circuits of higher voltage, a double arrester known as the type l,AA" is made by mounting two arresters on one base and connecting them in series. One arrester should be installed on each side of the circuit, as shown in the Diagram of Con- nections. For use in places exposed to weather, the Type "A" Arrester is furnished inclosed in an iron case, and designated Type "A," Form " C." Connections for Type 'A" Arresters. Fig. 16. Type " B " Incandescent Station Arrester 300 Volts or Less. The construction of the Type " B " Arrester is similar to that of the Type "A," but its magnet windings are excited only when a discharge takes place across the air-gap. A supplementary gap is provided in the Type "B" Arrester, in shunt with the magnets, thus providing a relief for the coils from excessive static charge without affecting their action upon the main gap. The magnet coils, carrying current only momentarily, allow the same arrester to be used on circuits of large and small ampere capacity. The Type "B" can also be furnished with weatherproof case similar to that used with Type "A." FlG. 17. Type " MB " for Direct CurrentCircuits up to 850 Volts. Fig. 18. Connections for Type " B " Arresters. LIGHTNING ARRESTEES FOR DIRECT CURRENT. 611 The Type " MD" Lightning Arrester has been designed for use on direct current circuits up to 850 volts. While similar to Type"M," Form " C " Arrester, it is considerably smaller, and is inclosed in a compact porcelain box measuring 1\ inches x 5 inches x 4£ inches. For street car and line use, Fig. 19. -n r— ' MD" Lightning Arrester in "Wood Box. the arrester is furnished in an additional box of iron or wood, as shown by Fig. 19. This arrester has been adopted as standard for railway and all direct current 500-volt circuits. It has a short spark gap, a magnetic blow-out, and a non-inductive resistance. CONNECTIONS OF MAGNETIC BLOW-OUT LIGHTNING ARRESTERS TYPE MD. FOR DIRECT CURRENT CIRCUITS UP TO 850 VOLTS. CONNECTIONS FOR LIGHTING OR POWER CIRCUITS. (METALIC CIRCBITS) T CONNECTIONS FOR RAILWAY CIRCUIT (ONE SIDE GROUNDED) REACTANCE COIL JS COMPOSED 25 FT. OF CONDUCTOR WOUND IN COIL OF TWO OR MORE TURNS AS CONVENIENT. Fig. 20. Connections of Magnetic Blow-out Lightning Arresters, Type " MD " for Direct Current Circuits up to 850 Volts. 612 LIGHTNING ARRESTERS. LIGHTSIMCJ ARKESTERM FOR AJLTFRUrATIXC* CUIIRE1IT. The G. E. Alternating Current Arresters bave been designed to operate properly with very small gap spaces. The arrester for 1000-volt circuity has two metal cylinders 2 inches in diameter and 2 inches long, separated by a spark gap of about J-5 inch. One cylinder is connected to the overhead line and the other cylinder to the ground, and a low non-inductive graphite resistance is placed in circuit. The large radiating surface of the metal cylinders combined with the effect of the non-inductive resistance prevents heating at the time the lightning discharge passes across the gap, and the formation of vapor which enables the current to maintain an arc is thus avoided. ALTERNATOR ALTERNATOR 10000 V. ARRESTER CONSISTS OF FOUR 2000 V. O.P. ARRESTERS CONNECTED IN SERIES. - GROUND 15000 V. ARRESTER CONSISTS OF SIX. 2000 V. D.P. ARRESTERS CONNECTED INSERIES. FIG. 21. Connections of Wirt or G. E. Alternating Current Short Gap Lightning Arresters, 5000 to 10,500 volts. The arrester under normal action shows a small arc about as large as a pin-head between the cylinders. The arrester for 2000-volt circuits is designed with two gaps of approxi- mately 3V inch each and a low non-inductive resistance. The G. E. Arresters are now furnished by the General Electric Company for use on all alternating current circuits at practically any potential. For circuits above 2000 volts, the standard 2000-volt double-pole arrester has been adopted as a unit, and several of these are connected in series to give the necessary number of spark gaps. LIGHTNING ARRESTERS FOR ALTERNATING CURRENT. 613 FlG.22. G. E. Alternating Current Lightning Arresters. ALTERNATOR FIG. 24. Connections of Wirt or G. E. Alternating Current Short Gap Lightning Arresters, 1000 to 3000 Volts. ALTERNATOR GROUND 1000 V.S. P. 1000 V.D. P., GROUND 2000 V.S..P, 2000 V.D. P. 614 LIGHTNING ARRESTERS. Fig. 25. THE GARIODT ARREiTER. In Fig. 25 a cross-section view is shown of the Garton Arrester. The discharge enters the Arrester by the bind- ing post A, thence across non-inductive resistance B, which is in multiple with the coil F, through conductors imbedded in the base of the Arrester, to flexible cord C, to guide rod D and armature E, which is normally in contact with and rest- ing upon carbon H, thence across the air-gap to lower carbon J, which is held in position by bracket K. This bracket also forms the ground connection through which the discharge reaches the earth. We have noted that the discharge took its path through the non-inductive resistance in multiple with the coil. This path is, however, of high ohmic resistance, and tbe normal cur- rent is shunted through the coil F, which is thereby energized, drawing the iron armature E upward instantly. This forms an arc between the lower end of the armature and the upper carbon H. As this arc is formed inside the tube G, which is practically air-tight, the oxygen is consumed, the current ceases', and the coil loses its power, allowing the armature to drop of its own Aveight to its normal position on the upper carbon. The arrester is again ready for another discharge. ELECTRICITY METERS. Meters for measuring the amount of electrical energy furnished to cus- tomers are commercially called wattmeters or recording wattmeters, whereas they are really measurers or meters of watt-hours. The Edison chemical meter, in which a shunted definite portion of the current supplied to the customer is made to deposit zinc upon an electrode of an electrolytic cell, is properly a coulomb meter, or ampere hour meter, which becomes a watt-hour meter if the pressure be maintained constant. This last meter is rapidly going out of use. The Thomson watt-hour meter, which is replacing it, can be used upon either direct or alternating circuits. It consists of a motor whose armature is connected in series with a resistance to the two mains, and whose field coils are in series with the supply circuit. The armature in rotating moves a recording mechanism. The rapidity of rotation is regulated by a copper disk connected to the armature shaft and moving between the poles of adjustable permanent magnets. It is made for use on two or three wire circuits, arc circuits, single phase or three phase a. c. circuits, and for recording input and output of storage batteries. The following diagrams show some of the principal uses to which it is put with the scheme of the connections to the circuits. There are many other purposes to which it is put, but the reader is referred to the instruction books accompanying the meters for further information on the subject. Fig. 1, Two-wire Meter. (Small Capacity.) 615 Fig. 2. Two-wire Meter. (Large Capacity.) 616 ELECTRICITY METERS. Fig. 3. Three-wire Meter. (High Efficiency Type). Fig. 4. Primary Meter. Fig. 5. Arc Circuit Meter. Fig. 6. Station Arc Meter. ELECTRICITY METERS. 617 Fig. 7. Balanced Three-phase Secondary Meter. Eig. 8. Balanced Three-phase Primary Meter. 618 ELECTRICITY METERS. Fig. 9. Two- wire Meters from 75 Amperes to 1200 Amperes. Fig. 10. Two Meters on Mono- cyclic System. Fig. 11. Balanced Three-phase Meter. Fig. 12. Three-wire High Efficiency Meter. PNT TUBE Fig. 13. Arc Circuit Meter. Fig. 14. Single-phase Primary Meter ELECTRICITY METERS. 619 Fig. 15. Large Capacity Station Meter Form G,. Fig. 16. Station Arc Meter. Fig. 16a. For Storage Battery 25 and 50 Amperes. 100 volts. OEIXRAI HTOT£§ CO!¥CI!Jl]¥I]¥« THOMSON In case a new jewel is inserted in the meter it is advisable to put in a new shaft end, as the point on the old one will probably be injured, more particularly if the meter has been running on the broken jewel. Just before inserting a new jewel in a meter, it is well to place a drop of fine watch oil on the jewel. Oil must not l»e used in the top bearing under any circumstances. Oil or dirt on the commutator will cause the meter to register less than the correct number of watt hours. If no " constant" is marked on the dial, the meter reads directly in watt hours. See that the disk and armature move freely, and that no dirt collects on the magnets in such a way as to touch the disk. Install the meter in a dry place, as far away from any heavy vibration as possible. When it is necessary to install a meter near a railroad, or in any place where the vibration is sufficient to cause sparking at the brushes, the ten- sion of the brushes upon the commutator should be slightly increased. This will do away with the sparking, and ensure greater accuracy. In case of severe jar, it is advantageous to place a number of soft rubber washers under the heads of the screws which bind the meter to the wall and between the meter and the wall itself at each screw. The disk will always rotate to the right when the meter is properly connected. 620 ELECTRICITY METERS. Testing- of Thomson Recording' Wattmeters. Most companies find it desirable to test meters on their lines from time to time, not so much to check the accuracy of the meters as to be able to state to the customer how the meter is operating. If only a rough test be required, it can be made by turning on a specified number of lights, multiplying the number of lights by the average watts per lamp, and using the following standard formula : — 3600 x Constant (if meter has one; „ , . .. _,/, ,,_ ^ . =r Seconds per revolution of disk. Watts m use ^ By using a stop watch, meters can be tested in this way, and the only in- accuracy is the difference between the estimated and actual watts per lamp. If a more accurate test be required, there are two methods, both of which are simple, and obviate the necessity of taking down the meter. A portable indicating wattmeter may be connected in series with the meter to be measured. The portable instrument will read directly in watts, and with the above formula give an absolute test. Another method is to have half a dozen high candle-power lamps, which have been tested at the station so that their wattage at all voltages is abso- lutely known. These lamps can be connected as the only load on the meter. By reading the voltage at the point of test with a portable voltmeter, and noting the watts recorded by the meter for the group of lamps, a direct comparison can be made. Calibration of Thomson Recording* "Wattmeters. Meters which have been neglected, misused, or very much worn, should be taken down, and brought into the station for repair and recalibration. In modern meters the speed can be increased or retarded about 16% by moving the magnets. On older meters having only one movable magnet, the variation obtainable by moving the magnet is considerably less. Meters which cannot be properly calibrated by moving the magnets can be roughly corrected by changing the resistance in series with the armature. Meters which are slow on light loads can be speeded without affecting the accuracy on high loads by increasing the shunt field coil, which is the fine winding. Meters Avhich show a tendency to creep, that is, to move slightly without any load, have too many turns in the shunt field coil. Creeping is almost invariably traceable to vibration, which aids the meter to overcome friction on very light loads. It can be corrected by removing turns from the shunt field coil until the meter disk just barely fails to move on no load. ALTERIATIHG CURRENT METERS. In addition to the Thomson watt-hour meter, which is used on either a.c. or d.c. circuits there is a class of induction meters used only on the a.c. circuits. The Schallenberger meter is of this type, and is made by the Westinghouse Electric and Manufacturing Company in several designs, such as Watt-hour meters, ampere-hour meters, and the first mentioned are also made in two- and three-phase meters. All of these meters depend in some way on the rotating of a disk or cyl- inder by means of induction coils properly placed in relation thereto. The Duncan integrating meter is another of the class, and one formerly made by the Fort Wayne Electric Corporation was very similar to the Schallenberger ampere-hour meter. Some of these meters are regulated as to speed by small fans placed on the armature shaft, and are hardly as accurate as those having a retarding disk betAveen magnets. THE STAIIEY METER. The Stanley manufacturing Company has recently (January, 1899) brought out an a.c. meter that is sealed and Avarranted to remain accurate within a very small percent for a period of 3 years, provided it is properly installed and the seals are not broken. This meter is of the induction type, and the disk upon which the coils act is held in suspension, and at the same time retarded by tAvo permanent magnets. The disk is so adjusted as to remain suspended midAvay between the poles of the magnets, and there is no other gearing for friction. THE STANLEY METER The following two cuts show its construction : — 621 Figs. 17 and 18. Directions for Installing- Stanley Meters. Place the instrument on a secure support in as nearly a vertical position as can be judged by the eye. Open one of the mains in the circuit to be metered, and connect the heavy black terminal of the meter to the main leading to the transformer or current generator, and connect the white ter- minal toward the lamp circuit or current consuming device. Connect the small shunt wire directly across the mains to the opposite side of the circuit so that the shunt connection of the meter will receive the full working pressure of the circuit at approximately the voltage indicated on the case cover. See cuts No. 19 and No. 20 for diagrams of connections. Figs. 19 and 20. Directions for Reading*. Kilo-watt hours are recorded directly on the dial without the use of a con- stant, unless otherwise marked on the case cover. The first right-hand pointer on the dial indicates 1,000 Watt hours, or 1 K. W. H. for one com- plete revolution of the pointer, and each unit indicated by this pointer rep- resents 100 Watt hours. The other pointers, taken in order from right to left, record successively 10 K. W. H., 100 K. W. H., 1,000 K. W. H., and 10,000 K. W. H. for one complete revolution of the pointer. 622 ELECTRICITY METERS. DIAGRAMS OF CO!¥l¥FCXIOI¥S OF SHAILE]». JIERttER II¥XEGRAXII^ WATTMETERS XO VARIOUS STYIES OF CIRCUITS Fig. 21. Connections for Single-Phase Circuits ; Current not exceeding 100 Amperes, Potential not exceeding 500 Volts. The illustration above shows the method of connecting a meter to a single- SKSSKr™* not exceedin§ io° ~ and at a ^ Fig. 22. Connections for Single-Phase Circuit; Current exceeding 100 Amperes, Potential not exceeding 500 ^olts. The illustration herewith shows the method of connection to a single- P^lV^rnn ^J^? a ^rrent exceedi"§" 100 amperes at a potential not exceeding 500 volts. In this case a series transformer is used, the current to be measured passing through the primary coil of the transformer while the meter receives from the secondary coil of the transformer current bear- ing a fixed ratio to the primary current. DIAGRAMS OF CONNECTIONS. 623 Fig. 23. Connections for Single-Phase Circuit ; Potential exceeding 500 Volts. The illustration shows the method of connecting the meter to a single- phase circuit carrying current at a potential exceeding 500 volts. To keep the high potential current out of the meter, hoth a series and a shunt trans- former are used, even for currents not exceeding 100 amperes. Fig, 24. Connections for Polyphase Circuits ; Current not exceeding 100 Amperes, Potential not exceeding 500 Volts. The illustration above shows the method of connecting two meters to a three-wire polyphase circuit, in which the current traversing each of the out- side wires does not exceed 100 amperes, while the potential between either of the outside conductors and the middle conductor does not exceed 500 volts. This connection is correct for a three-wire, two-phase system, and also for a three-wire three-phase system. 624 ELECTRICITY METERS. Fig. 25. Connections for Polyphase Circuits ; Current exceeding 100 Amperes, Potential not exceeding 500 Volts. The illustration herewith shows the method of connecting two of these meters to a three-wire polyphase circuit, where the current in each of the outside wires exceeds 100 amperes, while the potential between each of the outside wires and the middle wire does not exceed 500 volts. Series trans- formers are used to reduce the current to the meter. This arrangement is correct for either a three-wire two-phase or a three-wire three-phase system. Fig. 26. Connections for Polyphase Circuit ; Potential exceeding 500 Volts. The illustration shows the method of connecting two meters to a poly- phase three-wire system carrying currents at a potential exceeding 500 volts. It will be noted that both series transformers and shunt transformers are used. This connection is correct for either a three-wire two-phase or a three- wire three-phase system, WESTLN"GHOTJSE INTEGRATING WATTMETERS. 625 WESTIHrCJHOUSU IjITEGRATIHG wattmeter§, Two-Wire, Single-Phase. — The two-wire single-phase meter is rated for the average load of the installation, this being permissible on account of its ability to safely carry a load fifty per cent in excess of its rated capacity. It registers in International Watts the true energy deliv- ered to the circuit, and it is said to be correct for all power factors. The counter reads directly in watts or kilowatt hours. Series transformers are used on all circuits carrying more than 80 amperes, and for voltages above 500 volts shunt transformers are also used. These meters are connected to two-wire, single-phase circuits, as shown in Figs. 21, 22 and 23. TIiree-VTire, §>iiig-le-I*liase. — This meter is made to register the energy delivered by a three- wire circuit, through the medium of a specially designed series transformer, having two primary coils and one secondary coil. One of these primary coils is connected in series with one of the outside ires of the three-wire circuit, and the other primary coil is connected in [series with the other outside wire of the three-wire circuit. The secondary icoil, in which the current is proportional to the sum of the currents in the two primary coils, is connected to the wattmeter. The shunt circuit of the wattmeter is connected between the neutral and one of the outside wires. The current capacity, marked on the counter of the three-wire Westing- house wattmeter, represents the current in each of the outside wires of the • three-wire circuit. The voltage marked on the counter is that between one of the outside wires and the neutral wire. Fig. 27. Diagram of Connections of Westinghouse Three-Wire, Single- Phase Integrating Wattmeter. The total current capacity of a three-wire wattmeter is, therefore, twice that marked on the counter, which represents the capacity of one side only. The counter records, hoAvever, the total energy supplied to both sides of the three-wire installation ; and the watt hours recorded on the counter in one hour, when the meter is running at full load, will be twice the product of the current and the voltage marked on the face of the counter. 626 ELECTRICITY METERS. Two- or Three-I*hase Meters. The "Westinghouse polyphase meter records on a single dial the total energy delivered in all the phases of a two or three-phase circuit under all conditions of balance and of power factor. The current capacity marked on the counter of the polyphase wattmeter is the current in each wire of the circuit ; the voltage is that across a phase. No constant or factor is used. Instructions for Checking- and Testing- Westing-house Integrating Wattmeters. Registration. — These meters as shipped are ready for use, and are accurate -within the limits specified on the tag attached to them. The disk revolves 50 times per minute at full load; the direction of rota- tion being from left to right. The unit of power is the international watt, and all Avattmeters register directly in watts or kilowatt hours without the use of constants. Methods of Checking. — One of the two methods mentioned below are recommended, circumstances dictating which of the two is the better. First method is to compare the instrument to be checked with a standard indicating wattmeter, and timing the disk. Second method is by comparing with a standard integrating wattmeter. JFirst Method — Two-Wire, Single-I*hase Wattmeter. — Connect the instrument to be compared in circuit with a standard indicat- ing wattmeter, as shown in the following diagram. STANDARD INDICATING WATTMETER Fig. 28. Load the circuit until the desired reading is obtained on the indicating wattmeter, and keep it at a constant value while the integrating wattmeter is being read. Time the revolutions of the disk with a stop-watch, com- mencing to count when the spot on the disk has made one revolution (after the watch has started), and counting the revolutions for at least a minute. To arrive at the number of watts registered by the wattmeter, use the following formula : "Watts = -ThK. In this formula, R= complete number of revolutions of the disk in time T. T — time in seconds of revolutions B. K= constant. For wattmeters that are used without transformers, X = volts multiplied by amperes (as marked on the counter), multiplied by 1.2. For wattmeters that are used Avith series transformers (but checked without them), K = volts, as marked on the counter, multiplied by 6. For Avattmeters that are used with both shunt and series transformers (but checked Avithout them), A'=G00. In this way a wattmeter can be compared with a standard, and by varying the number of Avatts can be checked through its entire range. WESTINGHOTTSE INTEGRATING WATTMETERS. 627 All wattmeters for circuits exceeding 80 amperes are wound for 5 am- peres, and are made to register the energy delivered by the main circuit by means of series transformers. The primary coils of these transformers, which are of heavy capacity, are connected in the main circuits, while the secondary coil, in which the current is proportional to the current in the primary windings, is connected to the wattmeter. These wattmeters can be tested without the series transformers, but should be connected as in Fig. 31 above, and the test made in the manner indicated. The full load is, however, the product of the voltage marked as the counter multiplied by 5, and not by the current indicated on the counter. A', in this case, = volts, as marked' in the counter, x 6. All wattmeters of voltages exceeding 400 volts are provided with 100-volt shunt-coils and 5 ampere series-coils, and are connected to the main circuit through shunt and series transformers of the proper ratio. In checking, connect without the series or shunt transformers to 100-volt circuit, as shown in Fig. 28, and proceed as indicated above, remembering that full load is 500 watts, and that in the formula X= 600. Three-Wire, Single-Phase. —These wattmeters are all 5-ampere, single-phase instruments, and the method of connecting them for the first method of test is shown in Fig. 29. WESTINGHOUSE WATTMETER *J ICATING WATTMETERS Fig. 29. A. Comiect two standard indicating wattmeters, one into each side of the three-wire circuit, being careful to have the connections of these stand- ard wattmeters made on the supply side of the integrating wattmeter, as shown, so that it will not measure the energy used by them. Load the cir- cuit until the desired readings are obtained on the indicating wattmeters, and keep at a constant value while the integrating wattmeter is being read. Time the number of revolutions of the disk as before. To arrive at the number of watts registered by the wattmeter, use the following formula : Watts =r ~K. R = number of complete revolutions in time T. T =r time in seconds required for revolutions R. K= constant (volts times amperes, as marked on the counter, multiplied by 2.4). The reading of the integrating wattmeter should equal the sum of the readings of the two standard indicating wattmeters. B. A simpler method is to check the wattmeter without the series trans- former. As previously mentioned, all these wattmeters are 5-ampere, 100- volt, single-phase, two-wire instruments. For purposes of test it is neces- sary only to connect them, as shown in Fig. 31, into a single-phase, two-wire circuit, with a standard indicating wattmeter, and proceed in the same manner as for two-wire wattmeters of this capacity. Polyphase Wattmeter. — To compare a polyphase wattmeter with the standard, check each side separately on a single-phase circuit. Where transformers are not used in connection with the wattmeters, the full-load rating for each circuit of the wattmeter is the number of watts obtained by multiplying the current by the voltage marked upon the dial of the watt- meter. 628 ELECTRICITY METERS. If a series transformer is used with the Avattmeter, full load in each cir- cuit is the number of Avatts obtained by multiplying the voltage marked upon the dial by 5, as all Avattmeters used Avith series transformers are wound for 5 amperes. In testing, connect the polyphase Avattmeter as shoAvn in Fig. 30. Both shunt circuits of the integrating Avattmeter are connected. The main cur- rent, however, is passed through only one series coil at a time, by connect- ing " C " to "A" or to " B." "When one circuit of the wattmeter is fully loaded the rotating element makes 25 revolutions per minute, and 50 revolu- tions AvLen both phases are fully loaded. Fig. 30. Load the circuit until the desired reading is obtained on the indicating wattmeter, and keep it at a constant value AVhile the integrating wattmeter is being read. Time the revolutions of the aluminium disk for at least one minute. To arrive at the number of watts registered by the Avattmeter, use the fol- loAving formula : Watts —-?pK- Where R =r complete number of revolutions of the disk in time T. T= time in seconds of revolutions 1L K = constant. (For Avattmeters Avhich are used Avith both series and shunt transformers, but checked Avithout them, K = 1200.) "Always be sure to have both shunts connected when testing. Second Method: With Standard Xnteg-rating- Wattmeter. Single-Phase Wattmeters. — When using integrating Avattmeters STANDARD WESTINGHOUSE WESTINGHOUSE INTEGRATING INTEGRATING WATTMETER, WATTMETER Fig. 31. as standards, use one of same capacity and voltage as those under test. Load the circuit into Avhich the Avattmeter is connected. If the disk of the instrument under test runs in synchronism with the standard Avattmeter it is in correct calibration. Repeat for several different loads. Another method is to alloAv the instrument under test to run Avith the standard for several hours under full load. A comparison of the amount registered urn WESTINGHOUSE INTEGRATING WATTMETERS. 629 will show the difference between the two, or the error of the instrument tested. When but a single wattmeter is to be checked against the standard, it should be connected as shown in Fig. 31. When more than one wattmeter is to be checked against the standard, they should be connected as indicated in Fig. 32. Referring to Fig. 32 : If a short run is to be made, but one meter should be run with the standard at a time, otherwise the meter near the line con- nection will measure the energy taken by the shunts of those near the standard. If, however, the test is to be made by allowing the wattmeters to STANDARD WESTINGHOUSE AVESTINGHOUSE INTEGRATING INTEGRATING WATTMETERS WATTMETER run with the standard for several hours they can all be run together, as the amount of energy used by the wattmeters themselves will be so small a per- centage of the total readings that it will not be noticeable. l*olyi»Iiase Wattmeters. —Polyphase wattmeters should be checked against single-phase standards. The standard used, however, should be of twice the current capacity marked on the counters of the polyphase watt- meters. Connect as shown in Fig. 33. The wire at " A" is connected first to the upper phase of the meter and then to the lower phase, proceeding in the same manner as with single- phase meters, noting, however, that the full-load speed of the disk will be 25 r.p.m., as only one phase Avill be on at a time. Be sure to always have both shunts connected when making a test. In meters Avhich do not use series transformers there is only one shunt termi- nal (the other wire of the shunt being connected to the right-hand series terminal inside the meter). Fig. 33* Fig. 34. Fig. 31 shoAVs the method of connecting three-wire, single-phase Westing- house Avattmeters to three-Avire circuits. All three-Avire, single-phase Westinghouse Avattmeters, for circuits ex- ceeding 400 amperes per side, are connected in this manner. Fig. 35 shows the method of connecting polyphase Westinghouse watt- meters to two-phase circuits. 630 ELECTRICITY METERS. All polyphase Westinghouse wattmeters for two-phase circuits of 400 volts or less, and of 80 amperes or less, are connected in this manner. The following illustration shows the method of connecting polyphase Westinghouse wattmeters to three-phase circuits. All polyphase Westinghouse wattmeters for three-wire, three-phase circuits of 400 volts or less, and of 80 amperes or less, are connected in this manner. Fig. 36. The following illustration, Fig. 37, shows the method of connecting poly- phase Westinghouse wattmeters to two-phase circuits. All polyphase Westinghouse wattmeters for two-phase circuits of 400 volts or less, and greater than 80 amperes capacity, are connected with series transformers in this manner. Fig. 38 shows the method of connecting polyphase Westinghouse watt- meters to three-phase circuits. All polyphase Westinghouse wattmeters for three-phase circuits of 400 volts or less, and of greater than 80 amperes capacity, are connected Avith series transformers in this manner. WESTINGHOUSE INTEGRATING WATTMETERS. 631 Fig. 39 shows the method of connecting polyphase "Westinghouse watt- neters to two-phase circuits. All polyphase Westinghouse wattmeters for two-phase circuits of all jurrent capacities, and for more than 400 volts, are connected with shunt ind series transformers in this manner. Fig. 40 shows the method of connecting polyphase Westinghouse watt- meters to three-phase circuits. All polyphase Westinghouse wattmeters for three phase circuits of all current capacities, and for more than 400 volts, are connected with shunt and series transformers in this manner. 632 ELECTRICITY METERS. To Tell the Exact Current blowing- at Any Time in a Scliallenuerg-er JfEeter. Note the number of revolutions made by tbe small " tell-tale" index on the top of the movement, in a number of seconds equal to the constant o; the meter. The number of revolutions noted will correspond to the number of amperes passing through the meter. For example : the 20 ampere metei constant is 63.3 ; if the index makes ten revolutions in 63.3 seconds, 10 amperes are passing through the meter. In order to avoid errors in reading it is customary to take the number of revolutions during a longer time, say 120 seconds ; then as a formula, we have : Number of revolutions x meter constant _ Number of seconds §8© §S© OOKSO xtfrnpcre-houra. «*«?■*»—*-« "WS® "(§©0 Fig. 41. Dials showing Sample Readings. Fig. 42. Difficult Meter Readings. THE SCHEEFER WATT-METER. 633 THE iCHEErfER watt-hleter. This meter, made by the Diamond Meter Co., Peoria, 111., is another of the induction type, used for alternating currents, and has some special features. The two following cuts illustrate its latest development. Fig. 43. Round Pattern, Type D. Scheeffer Watt-Meter Closed. Fig. 44. Round Pattern, Type D. Scheeffer Watt-meter Open. A very ingenious device is used for sensitive adjustment, and the follow- ing cut and description taken from the Company's catalogue is sufficiently clear to indicate its use. FlG. 45. Meter Core. Showing Shields for Sensitive Adjustment. There are two knurled posts, A and B, secured to the meter core by screw clamps as shown in the cut. These posts carry iron shields that" can he made to embrace more or less of the disk by turning the posts. " When the iron piece or shield embraces the disk it exerts an influence indixctively on the disks so as to give it a torque, and will catise it to revolve slightly. The left-hand piece (looking at the meter in front) will cause a torque towards the right, and the right hand piece toward the left. If the two pieces equally embrace the disk they will balance each other, and no movement will result. By throwing one out the other will prevail, and cause 634 ELECTRICITY METERS. it to revolve. Thus the two pieces can he adj usted towards each other so that the meter is always balanced and just on the point of turning, and is highly sensitive to extremely small loads. Great care must be taken so the balance is perfect, as otherwise the meter will be overcompensated, and will slowly run on pressure, and record when no load is on. When this adjustment is made, a good way to establish a balance is to keep tapping the meter when adjustment is made, as this will give a better adjustment for the meter, as a meter will often not run on pressure wben quiet, but run slowly when sub- jected to vibrations. A very good way to calibrate a meter is to adjust the full load, and then adjust tbe knurled brass posts, so that by tapping a bal- ance of the meter is effected so as not to run on pressure. This condition will leave the meter highly sensitive and correct, as it is not necessary that the lower loads be calibrated by a Watt-meter. When the posts have been properly adjusted, they must then be fastened securely by screwing the clamp which holds them tight, so that they will not be distured." In testing or calibrating " Seheefl'er " meters, use a stop-watch for timing and the following formulae for determinations. MEIER CAICULATIOHTi. w _ B x 3,600 X C B — revolutions. W — watts. C — constant on meter dial. S — second. Wx S B- 3,600 X C B x 3,600 X C METER PRICE CHART. The General Electric Company furnishes a large price-chart for facilitat- ing the making of bills from meter readings. The above cut is a reduced facsimile of the chart. The figures at the bottom are kilowatt-hours ; those at the left are the amounts of bills in dollars and cents. The diagonals are different rates per kilowatt-hour. Selecting the diagonal having the rate at which charges are to be made, a point is found on it directly over the num- ber of kilowatt hours shown by the meter ; in the column at the left, on a horizontal line from the same point, will be found the amount of bill. For exar. pie, take 50 kilowatt hours at 10 cents per kilowatt hour, the amount of bill shown at the left is $5.00. 313.00 - 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r V V p y 1 | 1 1 1 II 1 II II M -IdA- ^ y • PRICE CHART i / AP/ y ' Ay y y ^ti^k ,' y PER 1000 W <:9 ' / <& / — 0 /, b / ;•; • :*> / v> ^ / ' ' y Y / / • ■ / y T, 7 / > y Jr <9 - 7 z / y ' : V '/'■// y v. / /// V _// / /■ ///, '// >A^ 1.00 - : i^^ MA ES = AMOUNT OF BILLS IN DOLLARS AND CENTS AGSC SSAE^THOUSANDSOF WATT-HOURS 1 1 1 1 0 i 8 12 0 20 I 2? .;> a ■ -1 8 5 ;u 6 1 6 a 7. Fig. 49. Meter Price Chart. WRIGHT DISCOUNT METER. 635 WRIGHT MiCOUST METER. This instrument is for use in connection with a watt hour meter for de- terming the maximum use of current during any given period; or may be used without the watt-hour meter in connection with any electrical device for which it is desired to know the maximum use of current, either direct or alternating. It is slow acting so as to take no account of momentary spurts, such as starting an elevator or street car, and is rated to record as follows : If the maximum load lasts 5 minutes, 80 % will register ; If the maximum load lasts 10 minutes, 95 % will register ; If the maximum load lasts 30 minutes, 100 Direction. Fig. 47. Wright Discount Meter. The heat due to the current passing in the circuit expands the air in the bulb, which forces the liquid down in the left column and up in the right. Should the quantity of heat be such as to force some of the liquid high enough, it will fall over into the central tube, where it must stay until the instru- ment is readjusted. The scale back of the central tube is calibrated in am- peres on the left and in watts on the right. After reading and recording the indication for any period of time, the liquid is returned to the outer tubes by simply tipping up the tubes, etc., which are hinged at the top connections for the purpose. The readings of the demand meter or discount meter, either of which names are used, together with those of the watt-hour recording meter, furnish a basis for a more rational system of charging for electricity than has been customary. This subject is being taken up by many of the larger electricity supply companies. The instrument is handy to use in circuit with a transformer to show how the maximum demand compares with the transformer capacity ; also on feeders and mains to show how heavily they may be loaded. TELEGRAPHY. In this chapter only the instruments used in telegraphy will be noticed and these, Avith their connections, in theoretical diagrams only. For th various details, whose presentation would defeat the purpose of clearness in this compilation, readers are referred to various works on telegraphy Lines, batteries, etc., are each treated in other chapters. AMERICAN, or CUOKER CIRCUIT METHOD. The following diagram shows the connections of the Morse system of single telegraphy, as used in the United States. The terminal stations only are shown, and in one case the local circuit is omitted. Several interme- LINE TO TERMINAL ft 1 ± Fig. 1. diate stations (in practice 25 is not unusual) may be cut in on one circuit ; all the instruments working in unison, in response to one key only. In Fig. 1 at either end is a key which, when open, allows the now un- attracted armatures to be withdrawn by the retractile spring, S. Closing the key restores the current to the relays, attracts the armatures to the. front stop ; the local circuit through the relay points is closed, and the signal is heard on the sounder. The attracting force of spring, S, is less than that of the relay cores as energized by the current from the battery used for a given circuit. It can, by "pulling up " on the spring, be made greater ; in which case the given current is ineffective to close the relays, and if the tension of spring, S, is maintained, battery must be added to close the relays. It is possible, therefore, by means of spring, S, to make a comparatively weak current ineffective to close the relay points. The significance of this will appear later in connection with the quadruplex. EUROPEAN, or OPES CIRCUIT METHOD. The following diagram shows the connections of one terminal station with the line connecting to the next. The ground plates may be dispensed with if a return wire from the next station is used, thus forming a metallic cir- cuit. This method of connecting Morse apparatus is used mostly in Europe, and has two advantages over the American method . a. The battery is not in circuit except when signals are being sent. b. When the key is closed and the current admitted to line, the coils of the relay are cut but of the circuit, thus lessening the hindrance to the flow of current. 636 lift TELEGRAPHY. 637 NE TO NEXT STATION KEY REPEATERS. In practical telegraphy, the high resistance of the line wire between the terminal stations, and imperfect insulation permitting leakage in damp weather, make it inexpedient to attempt to transmit signals over circuits whose lengths have well-defined limits. But a circuit may be extended, and messages exchanged over longer distances by making the receiving instrument at the distant terminal of one circuit do the work of a transmit- ting key in the next. The apparatus used for this purpose is called a re- peater,*and is usually automatic, in a sense Avhich will appear later on. From among the scores of repeaters, selection must be made of repre- sentative types, — the two in most general use. MilBiken Repeater. The following diagram illustrates the theory of the Milliken repeater, which is in general use in the United States and Canada. The essential feature of every form of automatic repeater is some device by which the circuit into Avhich the sender is repeating not only opens when he opens, but closes when he c' 638 TELEGRAPHY. In the diagram is represented the apparatus of a repeating station which appear the instruments and three distinct circuits in duplicate, viz. the east and west main line ; east and west local (dotted) ; east and west extra local (dash and dot). Starting' with both "east" and " w est" keys closed and the line at rest, battery//, whose circuit (dash and dot) is com- plete through transmitter, T/, energizes extra magnet. E', attracts the pen- dent armature, P', leaving the upright armature tree, the pendent armature, . P, being similarly held by battery, b. In his key, relay, E, opens, then transmitter, passes the west line, which opens, and wc transmitter, T' ; but at the moment Iran circuit (dash and dot) opens, releasing pel by its soring against the upright armature W, and transmitter, T', and therefore th its tongue and post. When the distant 1 begins with the west relay instead of east. the distant east opens X, through whose tongue and post uld open relay, W, and therefore smitter, T, opens, the extra local dent armature, P, which is drawn holding closed the points of relay. ■ east line, which passes through vest breaks and sends, the action and follows the same course, Weiny-Pliillips Repeater. A theoretical diagram of the Weiny-Phillips repeater is given herewith. It is in general use by one of the principal telegraph companies, and is introduced here because it involves the principle of differentiation in mag- net coils, which plays so important a part in duplex telegraphy. As in the Milliken, there are three distinct circuits in duplicate ; and in the diaj the parts performing like functions in the twro types of repeater larky lettered. The connections and functions of the main line (solid I Milli- tical with those of the 1 pendent armature of the latter, •aight iron core and its -windings, performing the same tunc ' circuits and of local (dotted) circuit ken. But instead of the extra magnets we have a tubular iron shell inclosing s the combination of shell and straight i as the usual horse-shoe core. The turns of wire around the core of the extra magnet are equally divided, and the current traverses the two halves in opposite directions. Such a core is said to be differentially wound, be- cause the core is energized by the difference in strength of the currents in the coils; but when the coils are equal in resistance, the equal currents, passing in opposite directions around the core, neutralize each other. If one of the coils is opened, the core at once becomes a magnet capable of holding the armature at the moment when, the repeater in operation, the " east "" station opens his key, opening relay, E ; then transmitter, T ; then opening the " west" wire, which would open relay, AV, transmitter, T', and therefore the east wire ; but the opening of transmitter, T', is prevented by the energizing at the critical moment of core W one coil of which is opened DUPLEX TELEGRAPHY. 639 when transmitter, T, opens. When the distant west breaks and sends, the action begins with the west relay instead of the east, and follows the same course. DO>IEX VELEGHAPHY. That method of telegraphy by which messages can be sent and received over one wire at the same time is called duplex ; and the system in general use, known as the polar duplex, is illustrated in the accompanying diagram. In single telegraphy all the relays in the circuit, including the home one, respond to the movements of the key ; the duplex system implies a heme relay and sounder unresponsive, but a distant relay responsive to the move- ments of the home key ; and this result is effected by a differential arrange- ment of magnet coils, of which the extra magnet coils in the Weiny-Phillips repeater furnished an example. A current dividing between two coils and their connecting wires of equal resistance will divide equally, and passing round the cores, will produce no magnetic effect in them. This condition WEST C it-- EAST THEORETICAL DIAGRAM OF POLAR DUPLEX balancing switch omitted Fig. 5. is established when tbe resistance of the wire marked — > < — in the diagram is balanced by the resistance of a set of adjustable coils in a rheostat marked R. This is called the ohmic balance (from ohm, the unit of resistance) ; and the static balance is effected by neutralizing the static discharge on long lines by means of an adjustable condenser, C, and retardation coil, r, shunt- ing the rheostat as shown. In the single line relay the movement of the armature is effected by the help of a retractile spring in combination with alternating conditions of current and no current on the line. In the polar relay the spring is dispensed Avith, and the backward movement of the arm- ature is effected, not by a spring, but by means of a current in a direction opposite to that wbich determined the forward movement. This reversal of the direction of the current is effected by means of a pole-changer, PC, whose lever, T, connected with the main and artificial lines, makes contact, by means of a local circuit and key, K, with the zinc ( — ) and copper (-J-) terminal of a battery alternately. The usage in practice is zinc to the line when the key is closed ; copper, when open. The law for the production of magnetic poles by a current is this: When a core is looked at "end on" a current passing round it in the direction of the hands of a clock produces south-seeking magnetism, S ; in the opposite direction, north-seeking mag- netism, marked N. A springless armature, permanently magnetized and 640 TELEGRAPHY. pivoted, as shown in the drawing, will, if its free end is placed between S and N magnetic poles, be moved in obedience to the well-known law that like poles repel, while unlike poles attract each other. The" east" and" Avest " terminal is each a duplicate of the other in every respect ; and a description of the operation at one terminal will answer for both. Under the conditions shown, the keys are open ; and the batteries, which have the same E.M.F. oppose their copper (+) poles to each other, so that no current flows in t Lie main line. But in the artificial line the current flows round the core in such direction as, according to the rule just given, to produce N and S polarities as marked, opening the sounder circuits at both terminals. If, by means of key, K', the pole-changer, PC, of " east "' station is closed, the connections ot battery, IV, are changed ; it is said to be reversed ; and it now adds its E.M.F. to that <>1 battery B, the current flowing in a direction from " west" to " east " ; i.e., from' copper to zinc. But the current in the main line is to that in the artificial as 2 to 1 ; and if the relative strength of the resultant magnetic poles is represented by small type for that produced by the current in the artificial line, and by large type for the main, the magnetic conditions can be graphically shown, as they are produced on each side of the permanently magnetized armatures marked (X) and (X'). In relay, PR7, it is Sn (S/) sX, causing it to remain open ; in relay PR it has changed to Xs (X) nS — just the reverse of that shown in the diagram — the relay therefore closes, and the sounder also. If key, K, of the west station is closed at the same time, the batteries are again placed in opposition, but with zinc ( — ) poles to the line, instead of, as in the first instance, copper (-)-) poles. The result is no current on the main line ; but the current in the artificial lines, flowing in the direction from the ground (whose potential is 0) to the zinc (— ) of the batteries, the magnetic condi- tion at " east" station is represented by n (X') s, which closes relay, PR'; and at " west " station by n (X) s, which closes relay PR. The conditions necessary to duplex work, viz., that the movement of key, K7, should have no effect on relay, PR', but should operate the distant relay, PR, are thus fulfilled, and the transmission of messages in opposite directions at the same time is made practicable. In the case of the Wheatstone Automatic duplex this exchange goes on at high rate of speed, the maximum rate being 250 words a minute. Duplex Repeater. In wires worked in the duplex or quadruplex system, the static capacity of the wire, which plays little if any part in the operation of circuits by the single method, places a limit on the length of the continuous circuit. But the distance between working stations can be greatly extended by the use of repeaters in which, by an arrangement perfectly simple, the pole-changer of a second circuit is controlled by the relay points of the first. The long- est regular circuit in the United States is that worked between Xew York and San Francisco, with six repeaters. ftUADRHPLEX. The quadruplex system of telegraphy allows of two messages being sent in either direction, over the same wire, and at the same time. In theory it is an arrangement of two duplexes, so different in principle as to permit of their combination for the purpose designated. If the accompanying dia- gram of the quadruplex is examined, there will be noticed in it the pole- changer, polar relay, and all the apparatus of the polar duplex. The polar relay at the " east" station (not shown) will respond to signals sent by the pole-changer, PC, at the " west " in the manner described in the paragraph on the Polar Duplex, so long as the working minimum of current is main- tained. This working minimum can be doubled, trebled, or quadrupled without appreciable difference to the polar relays. In the paragraph on Single Telegraphy, the operation of the single relay, fitted with a retractile spring, Avas effected by opening and closing the key ; or, in other words, by alternating periods of "no current" and "current "on the wire. It was further stated, in anticipation of its introduction at this point, that the spring could be so adjusted that a weak current, though flowing all the time through the coils, would not close it. To effect the closing an increase THE STEARNS DUPLEX. 641 of battery, and therefore of current strength is necessary, so that the relay, instead of, as in the first instance, responding to alternating periods of " no current " and " current" could be operated by alternating periods of " weak current" and "strong." In the diagram, transmitter T, when its key is open, admits to the line a current sufficient to operate the polar side ; and THE QUADRUPLEX (one terminal) Fig. 6. at the " east" station (not shown) there is a differentially wound relay, M', the duplicate of relay M in the diagram, the tension of A\Those spring makes it unresponsive. But when all the battery is on, a condition which obtains when the key closes transmitter, T, the distant relay, M/, is closed. In short, there is in the quadruplex a pair of polar relays which respond to changes in the direction, not in the strength of the current ; and a pair of neutral relays, which respond to changes in the strength, not to the direction of the current. The diagram shows the apparatus in its simplest form ; there are a number of details in connection with its operation, the complete connec- tions for which are rather too complicated for this book. On page 199 of Mavers's American Telegraphy will be found a diagram embodying the full scheme of connections ; "ami Thorn and Jones' Telegraphic Connections con- tains diagrams and detailed descriptions of the systems in general use. THE STEARH'S BlPtEX. The operation of differential relays like M in the diagram of the quadru- plex, by alternations of "no battery" and "battery," is the principle of the Stearns duplex which, as the first condenser-using, and therefore static- eliminating duplex in the world, has a certain historic interest. In Febru- ary, 1868, there were in use by the Franklin Telegraph Company a duplex, set New York to Philadelphia, and another to Boston ; and in August, 1871, STEARNS DUPLEX (ONE TERMINAL) 642 TELEGRAPHY. by the Western Union Telegraph Company, a duplex, New York to Albany — all without condensers. In March, 1872, the Stearns Duplex, with con- denser, went into operation between New York and Chicago, but it has been superseded by the polar system. Reverting to the diagram, the pole-changer with its adjuncts, and the polar relay of the quadruples, are omitted; one pole of the battery is grounded, and the lever of transmitter, T, is grounded through a resistance equal to that of battery, B. This grounds the line through tongue, T, and leaves the battery open at the post, P. The " east" station (not shown) is a duplicate of the " west," and the control of relay, D, by the distant trans- mitter, T', may be traced as follows. Suppose distant transmitter, T', sends copper to the line when closed, the current dividing equally between the main and artificial lines in distant relay, D', has no effect upon it ; but at the Avest station there is no current in the artificial line in relay, D, so that the current in the main line closes it. Open the key, K/, and the line is grounded through the lever of transmitter, T'; battery P/is open, and there being no current on the wire, relay, D, is open in response to the opening of distant key, K'. Let transmitter, T, now be closed, and trace the control of relay, D, by the distant key, K'. The current, which now flows from the ground through the lever of open transmitter, T', to the zinc pole of battery, B, is neutralized in relay, D, by an equal current flowing from the ground through its artificial line in the opposite direction around its cores, so that relay, D, remains open. Now close distant transmitter, T', and the current in the artificial line (i.e., through the rheostat, R) of relay D is over- powered as to its effects by a current on the main line of twice its strength, and relay D is closed. It 'is thus shown to be controlled by the distant key, K7, irrespective of the position of home key, K, and the conditions necessary to duplex telegraphy are met. TELEGRAPH CODES. THorse, used in the United States and Canada. Continental, used in Europe and elsewhere. l*hiliij»M, used in the United States for " press " work. Dash — 2 dots. Long dash = 4 dots. Space between elements of a letter = 1 dot. Space between letters of a word = 2 dots. Interval in spaced letters = 2 dots. Space between words = 3 dots. JDetters. Morse. Continental. E — — I M N P TELEGRAPH CODES. 643 Morse. T U V w X Y Z Numerals. Morse. Continent c 2 6 0 Punctuations, etc. Morse. Continental. : Colon : — Colon dash ; Semi-colon ? Interrogation ! Exclamation Fraction line — — Dash - Hyphen ' Apostrophe — £ Pound Sterling / Shilling mark $ Dollar mark d pence Capitalized letter Colon followed ) by quotation :" J c cents . Decimal point If Paragraph Italics or underline ( ) Parentheses [ ] Brackets ""Quotation ) marks. J Quotation within j a quotation | Phillips. . Period : Colon — — : — Colon dash ; Semi-colon , Comma ' ? Interrogation 644 TELEGRAPHY. ! Exclamation Fraction line — Dash - Hyphen ' Apostrophe € Pound Sterling / Shilling mark $ Dollar mark , R3, R4, connected as shown. Those at either end must be equal to each other, but the two ends need not be the same. These resistances must be greater than that of the line in order that the currents from T3 and T4 may pass along the line rather than around the coils. The condensers C may be placed in shunt to the coils in order not to retard the current, so that 1\ and T2 may work better. . . ... __ - R STATION » Hfjr-j Fig. 38. Series System with Microphones and Magnetos. INTERIOR TELEPHONE SYSTEMS. 665. arrangement, as well as in the one shown by Fig. 38, the microphones, receivers, and microphone batteries are directly in series with the line, no induction coils being used. Instead of vibrating bells and batteries for ringing, we may use a polar- ized bell, C, and a generator, G, as shown in Fig. 3S. In such an arrange- ment the talking current must pass through all the polarized bells except those at the stations where the receivers are removed from the hooks. Fig. 39. Bridging System with Microphones and Magnetos. A better arrangement is to use high-impedance bells bridged across the two-line wires, as shown in Fig. 39. The generator, as explained in connec- tion with Fig. 36, is normally on open circuit. Three bridging methods are shown. At Station 1 some of the current from the battery, M.B., can flow through the bell when the receiver is off the hook, but this will do no harm ; in fact, it may be beneficial, for it allows a larger direct steady current to flow through the microphone. The fluctuations in the current produced by the microphone cannot pass through the bell-magnet coils, but will pass through the line circuit on account of the lower impedance of the latter. At Station 3 the bell is cut out when the hook switch is raised, and at Station 2 both the generator and bell circuits are cut off by raising the hook. An extra contact, d, is required at these two stations, but on the other hand, there are two bells less across the circuit to form shunts or leaks for the current when two parties are conversing. On the whole, the arrangement at Station 3 is the best of the three. Fig. 40 represents a series party system (corresponding with that which was shown at Station 1 in Fig. 37) in which a battery, £, and vibrating bell, V, are used for signaling, and an induction coil, 1, is added to the speaking apparatus. The primary of the induction coil is in series with the micro- phone transmitter, T, and its battery, MB., and the secondary is in series with the telephone receiver and the line. The connections at Stations 1 and 2 are identical ; when the receiver hook, H, is down the talking instruments are entirely cut out, and when it FlG. 40. Series Party System, with Induction Coils and Signaling Batteries. is up the signaling key, battery and bell are thrown out of circuit and the main circuit passes through only the telephone receiver and the secondary of the induction coil. At Station 3 the connections are different ; when the receiver hook is down the telephone receiver and secondary of the induc- tion coil are merely short-circuited, while the transmitter, its battery, and 666 TELEPHONY. the primary of the induction coil are open-circuited. When the hook is up, the talking instruments are connected up for service and the signaling part of the apparatus is short-circuited. Fig. 41 corresponds with Fig. 40, except that magneto-generators, G, and magneto bells, C, have been substituted in the place of the signaling battery and vibrating bells shown in Fig. 40. The station connections correspond also, the receiver hook, H, at Stations 1 and 2 being arranged to throw in and out of circuit the talking apparatus and the signaling apparatus, while the hook at Station 3 merely short- circuits the signaling apparatus or the receiver circuit, according to its position. This arrangement is the preferable one of the two, for the reason that faulty switch contacts at the receiver hook will not open the circuit, so that there will always be a continuous line through which one may signal. Fig. 41. Series Party System Using Induction Coils and Signaling Magnetos. A simple system installed where there was considerable noise, dirt, and vibration, is represented diagrammatically by Fig. 42. Here, there are three line wires, a, b, and c, the line c forming a common return for both the signalling and the talking circuits, a and b, on which the apparatus is ar- ranged in series. In this system the talking line is never open-circuited, the telephone hook, JJ, serving to merely short-circuit the receiver and the secondary of the induction coil when down, and to remove the short-circuit and close the local circuit of the transmitter and induction coil primary when up. It is obvious that the middle line wire, c, gives a free path to the talking current, instead of its being forced through the signaling bells. Such an arrangement facilitates the separation of the signaling and talking ap- paratus, so that the call bells can be located where they can be easily heard, while the transmitter and receiver may be put in a sound-proof closet. The disagreeable noises due to induction from lighting or power circuits may be overcome by using a twisted three-conductor cable between stations. Such an installation is greatly superior to the series system shown by Figs. 40 and 41. H'M- HiM- "t^^ * W-W Fig. 42. Three-wire Series Party System. Fig. 43 shows a series system in which one battery is used both for signal- ling and for talking. In this system the connections are alike at all stations ; Avhen the receiver hook, H, is down and the signaling key, /, is up, there are included in the line circuit only the vibrating bells. Depressing the signal- INTERIOR TELEPHONE SYSTEMS. 667 ing key I, puts the battery in the line and causes all the bells to ring. It is preferable to have the batteries so connected up that if two or more signal- ing keys should be depressed at once the batteries will agree in polarity. When the receiver hook is up the battery is connected in series with the transmitter and the primary of the induction coil, wbile the signaling key and bells are thrown out of circuit and the telephone receiver and secondary winding of the induction coil are included in the line, as shown at Station 3. Fig. 43. Series Party System using only Battery at each Station for both Talking and Signaling. In this, as in previous series systems, with the exception of Fig. 42, the talking current must flow through the signaling bells at idle stations. The advantage of the system is obviously that it eliminates half the batteries, only the one battery being used at each station for both signaling and talk- ing. As in all series systems where vibrating bells are used, the vibrators, should be short-circuited on all bells except one. The best method for connecting a large number of telephones on a single system where only two line wires may be used is to bridge them, as shown in Fig. 44. The dots A and A', represent the binding-posts of each complete outfit. The bells are permanently bridged between the two line wires at Stations 1, 2, and 4, irrespective of the position of the receiver hooks. The magneto generator is also bridged across the two line wires in an independ- ent circuit, which is normally kept open either by a push-button, k, or by an automatic device on the magneto spindle. Fig. 44. Bridging Party-Line System ; Three Arrangements of Station Instruments. At Station 3 the magneto generator is bridged permanently across[the line as in Stations 1, 2, and 3, but the bell is connected across only when the re- ceiver hook is down, being thrown out when the hook is up. At Station 5 the bell and generator are bridged across the line wires when the receiver hook is down, and are cut out entirely when it is up. At all of the stations a third bridging circuit includes the receiver and the secondary winding of the induction coil in series, this circuit being open when the receiver hook is down, and closed when it is up. The hook also closes the local transmitter circuit in the usual way Avhen it is up, and opens it when it is down. The connections shown at Stations 3 and 5 possess the advantage of cutting out their signaling bells entirely when the receiver hooks are up, instead of leaving the bells shunted across the line continuously, as is the case at Stations 1, 2, and 3. 668 TELEPHONY, intercommunicating systems. An intercommunicating system may be denned as a system having thre or more telephones connected to the same system of wiring in such a maniie that one may from any station call up and converse with any other station without requiring any central-station switchboard whatever. Intercofl municating systems require one wire from each station to every other statio and at least one more wire running through all the stations. "Where vibr, ing bells and one common ringing battery are employed, at least two more wires than there are stations are necessary. At each station there must b a switch of some kind whereby the telephone at each station may be con nected to any one of the wires belonging to the other stations. Intercom municating systems are very practical and satisfactory up to fifteen or even twenty stations ; beyond that, the large number of wires running through all stations makes the cost of the system increase rapidly, especially when the stations are some distance apart. For a large number of stations well scattered, a simple central-station switchboard system is preferable. Fig. 45 shows a very common but not a good method of interconnecting a number of telephones, where each station is equipped with ordinary series bells and magneto generators. Theoretically any number of telephones may be connected on such a system, but practical consideration would place the limit at about twenty. In this figure there are four stations ; at Kos. 1, 2, and 4 the telephone connections are drawn in full, while at No. 3 is shown the telephone outfit as it usually appears. There are four individual line wires, numbered 1, 2, 3, and 4, and a common return wire. Thus there is one more wire than there are stations, and all these wires run through all the stations, each wire being tapped at each station and not cut. At each station there is one ordinary telephone instrument consisting of the usual I talking apparatus, magneto-generators and polarized bells. Below each I telephone there is an intercommunicating switch, the buttons of which are i connected to the respective line wires, and the common return wire. When not in use the switch at each station should remain on the home button. liii It" Fig. 45. Intercommunicating System, with Magneto Signaling Gener- ators and Polarized Bells. "With all the levers in this position, a person at any station can call up any other station by moving the switch lever to the button connected with the individual line of the station desired, and turning the generator handle ; only the bells at the home station and at the station called up will ring. The ringing and talking currents pass through only the instruments at the stations in communication. After finishing the conversation, the switch lever at the home station must be returned to its home position, otherwise the system will be crippled. INTERCOMMUNICATING SYSTEMS. 669 In Fig. 46 is shown a method of wiring the intercommunicating switch hat avoids the principal objection mentioned in connection with Fig. 18; hat is, the failure to return the switch to the Lome position does not leave he station so that it cannot be called up. Only four stations are shown, »ut the system can be extended to include as large a number as may be lesirable. The usual telephone sets, consisting of a microphone trans- mitter, induction coil, receiver, hook switch, two cells of battery, a series nagne'to-generator and polarized bell, are included in the outfits indicated >y Tlt T2, etc. The inside connections of these telephones are the same as ihown in the preceding figure. Fig. 46. In Fig. 46 one binding-post of each telephone is connected to the common return wire, and the other binding-post is connected to both the lever arm, s, and the individual line wire belonging to that particular station. The home button in this last system is the first on the left and is not con- nected to anything ; it is really a dummy button, but it should be there by all means, for the lever, s, of the switch should always be returned to it when the original calling party leaves the telephone. If all switch arms, s, are on the home buttons it will be found that the circuits of all instru- ments are open and no bell will ring, no matter what generator is turned. If Station 2 desires to call Station 1 it will be necessary to first move the switch arm, s, at Station 2 to button 1. Fig. 47 is a system similar to that shown in Fig. 46, but arranged for vi- brating bells and one common calling battery, CB, in place of magneto- FiG. 47. Common Signaling-Battery System. 670 TELEPHONY. generators and polarized bells. A battery is used at each station for oper- ating the transmitter. This is probably the best arrangement of batteries for such a system where vibrating bells are used. This system requires one more wire than that shown in Figs. 45 and 46 where magneto-calling ap- paratus is employed; thus there are two more wires throughout than there are stations. The calling battery, CB, must be connected to the two wires shown, but it may be located at any convenient place. In this arrangement only the bell at the station called will ring, the bell at the calling station remaining silent. If the bells are not arranged in this manner, the vibra- tors on the two bells that happens to be connected in series when making a call might interfere more or less Avith good ringing. Furthermore, it would not do to short-circuit any of the vibrators, because there is no telling which two stations may be connected together in making a call. -j^T Fig. 48. Common Signaling-Battery System. Trouble is experienced with intercommunicating systems similar to that of Fig. 47 by reason of the user carelessly leaving the selective switch S, off the home button after using the telephone. Fig. 48 shows a method of wir- ing such a system which obviates to a considerable extent this trouble. Here, the vibrating bell is permanently connected to the home button, and the pivot of the switch, S, is connected to the arm of the push-switch, Iv. Any station can still be called up, no matter on what button its switch, S, may be left. The same system of wiring employed in Fig. 48 is applied to the system shown in Fig. 49, in which magneto-generators, G. and polarized bells, C, are used in place of the battery and vibrating bells. There is no need of having a push button or automatic shunt on the generator, although it will do no harm. The generator is normally on open circuit because one of its terminals is connected to the under contact of the push switch, Iv. In order to call up a station, the switch, S, is placed on the button belonging to the station desired, the push switch, K, depressed, and the generator handle turned. Since no common battery is employed for ringing, this system requires one less wire through all the stations than the preceding arrange- ment. INTERCOMMUNICATING SYSTEMS. 671 In Fig. 50 is shown an arrangement in which one conveniently located common battery, C B, supplies current for ringing and also for all trans- mitters. No matter where the lever of the selective switch is left, the bell can still be rung, but conversation cannot be carried on until the switch at the station called is returned to the home button. This system includes a piece of apparatus at each station that has not been required in any of the systems previously described, to-wit : the impedance coil E. Where a common battery supplies all the local microphone circuits with current in systems of this kind, there is very apt to be cross talk between two pairs of telephones that may be in use at the same time, in which case the battery will be supplying current to four microphones. BATTERY WIRE Fig. 50. Common Battery System with Impedance Coils. The cross talk is due to the variation in the fall of potential along the battery and common return wires. The cross talk may be greatly reduced by using batteries of very low in- ternal resistance, such as storage cells, and making the common return and battery wires extra large, that is, small in resistance, so that the vari- able fall of potential through the battery and in these two wires may be small. However, it is impractical to make the resistance of these two wires low enough, especially where they are of considerable length, to eliminate all cross talk. Another way to reduce the trouble from cross talk is to insert an impe- dance coil in each microphone circuit, as shown in Fig. 50. This makes the impedance of each microphone circuit large compared to that of the two lines and battery, and in order to get the same current as before in each microphone the e. m. f . of the battery must be increased. These im- pedance coils reduce the efficiency of the system, but the reduction in cross talk compensates for this loss to a great extent. -#- ~ir Fig. 51. Radial System ; Selective at One Station Only. 672 TELEPHONY. It sometimes occurs that a system is required to be so arranged that one station can call up any one of the others, but the others can call up and converse with the first station only. Fig. 51 is a diagram of such a system; Station No. 1 or No. 2 can call up station C by merely depressing the push switch Kl or K2, but they cannot call up or converse with each other. Station C by means of the switch, S, and push, K, can call up either Station No. 1 or No. 2. There are only two wires that must run through all the stations. There is one Avire, however, from Station C to each one of the other stations. These wires, Call Wire No 1 and Call Wire No. 2, are used only when Station C calls up one of the other stations. One wire could be made to answer if there was no objection to having all but the home bell ring when Station C makes a call. In this case a certain num- ber of rings would be necessary for each station except C, and the one common call wire would be connected to the signaling key at a, Station C, and there would be no need of the switch, S. As arranged in the diagram, the push switch, K, is normally open. When Station C desires to call Station No. 2, for instance, the switch, S, must be turned to button 2 and the push switch, K, depressed. The one common battery, B, furnishes current for all ringing and talking. At each station there is an ordinary receiver, microphone transmitter, and vibrating bell. There is only one bell in circuit when a call is made so that each bell must have a vibrator. It makes no difference upon what button the switch, S, is left. In the systems so far described there is nothing to prevent the intercom- municating switch from being left off the home button when the conversa- tion is finished and the receivers hung up. Fig. 52. Ness Automatic Switch. An example of a device obviating this trouble is the Ness automatic switch, illustrated by Fig. 52, arranged so that the replacing of the re- ceiver upon the hook causes the switch to fly back to its home position. In the engraving S is the lever of the selective switch, adapted to be ro- tated over the various contact buttons, 1, 2, 3, etc. It is mounted upon a shaft, A, passing through the front board of the box and carrying a ratchet- wheel, E, inside the box. This ratchet-wheel is held in any position to Avhich it may be rotated by a pawl, F, and thus prevents the lever S, from turning backward. Upon the short arm of the hook lever, H, is pivoted a dog, G, adapted, when the receiver is removed from the hook, to engage a notch in the pawl, F; when the receiver is replaced, the dog, G, is pulled upwards and lifts the pawl out of the engagement with the ratchet-wheel, allowing a spiral spring around the shaft, A, to return the switch lever, S, to the home button. After raising the pawl out of the notch on the ratchet- wheel the dog slips out of the notch on the pawl, thus allowing the latter to return into contact with the ratchet-wheel in order to be ready for the next use of the telephone. In order, however, that the pawl may not engage the ratchet-wheel before the lever, S, has fully returned to its normal position, INTERCOMMUNICATING SYSTEMS. 673 a second dog, J, is provided which is pressed by a spring so as to occupy a position under the pin,p, carried on the pawl, F, thus holding it out of engagement with the ratchet-Avheel until the rotation of the lever is com- pleted. At this point a pin on the farther side of the ratchet-wheel pushes the dog, Jy out of engagement Avith the pin, p, and allows the pawl, F, to drop into contact with the ratchet-wheel. Fig. 53. Common Signaling Battery System ; Individual Talking Batteries. In Fig. 53 are shown the circuits of a four-station system using one com- mon battery, CB, for ringing up the various stations, each station having an ordinary vibrating bell, C. The circuits of Stations 1 and 4 are shown in full, while those of the intermediate stations, being exactly the same, are partially omitted. It will be noticed that the switch lever, S, at each station is connected with the line wire bearing the same number as that station, by means of the Avire, d. Each line wire is also connected at each of the stations not bearing its own number with a button on the switch of System having Common Talking and Signaling Battery. 674 TELEPHONY. that station which does hear the same number in the manner pre- viously described, by means of tbe wire, e. In this common-battery call system two additional wires are run, one being termed the " call wire " and the other the " common talking wire." The call wire and the talking wire are connected through tbe calling battery CB, as shown. It is evident that the number of wires passing through all the stations will be two more than the number of stations, irrespective of that number. If Station 4 desires to call up Station 1, for example, No. 4 will turn his switch lever until it rests upon button 1, then a slight pressure upon the switch knob causes the switch lever, S, to touch the contact strip, D, com- pleting a circuit from the battery, CB, to contact strip, D, lever, S, and button, 1, at Station 4; line wire, 1, wire, d, switch, H, and bell, C, at Station 1, and back to the battery through the common talking wire. "When both subscribers remove their receivers from the hooks, the circuits are completed over line wire 1 with the common talking wire as a return. At tbe close of the conversation the receiver is simply hung upon the hook, and the automatic mechanical device returns the lever to the home po- sition. Fig. 54 shows the application of the Ness automatic switch to an inter- commuicating system, using one common and centrally located battery for supplying both the ringing and talking current. The section, TB, of the battery supplies all the microphone transmitter circuits, and the whole battery, KB, supplies tbe current for ringing the ordinary vibrating bells that are used in this system. In this arrangement it is evident that the number of wires passing through all tbe stations will in any size of system be three in excess of the number of stations. ELECTRO-CHEMISTRY. - ELECTRO- METALLURGY. ELECTRO-CHEMIiTRY. Electrolysis. The separation of a chemical compound into its constituents by means of an electric current. Faraday gave the nomenclature relating to electroly- sis, tie called the compound to be decomposed the Electiolyte; and the pro- cess Electrolysis. The plates or poles of the battery he called Electrodes. The plate where the greatest potential exists he called the Anode, and the other pole the Cathode. The products of decomposition he called Ions. Lord liayleigh found that a current of one ampere Avill deposit 0.017253 grain, or 0.001118 gramme, of silver per second on one of the plates of a sil- ver voltameter, the liquid employed being a solution of silver nitrate con- taining from 15 per cent to 20 per cent of the salt. The weight of hydrogen similarly set free by a current of one ampere is .00001014 gramme per second. Knowing the amount of hydrogen thus set free, and the chemical equiva- lents of the constituents of other substances, Ave can calculate what weight of their elements will be set free or deposited in a given time by a given current. Thus the current that liberates 1 gramme of hydrogen will liberate 7.94 grammes of oxygen, or 107.11 grammes of silver, these numbers being the chemical equivalents for oxygen and silver respectively. To find the weight of metal deposited by a given current in a given time, find the weight of hydrogen liberated by the given current in the given time, and multiply by the chemical equivalent of the metal. Thus: Weight of silver deposited in 10 seconds by a current of 10 amperes = weight of hydrogen liberated per second X number seconds X current strength x 107.11 = .00001044 X 10 X 10 X 107.11 = .1118 gramme. Weight of copper deposited in 1 hour by a current of 10 amperes = .00001044 X 3600 X 10 X 31.55 = 11.86 grammes. Since 1 ampere per second liberates .00001044 gramme of hydrogen, strength of current in amperes _ weight in grammes of H. liberated per second .00001044 weight of element liberated per second ~~ .00001044 X chemical equivalent of element Resistances of Dilute Sulphuric Acid. (Jamin and Bouty.) Ohms per c.c. at Ohms pei Cu. In. at Density. ofc Sh 0&H o . oP^ °fc ' °fc o? d^ Do Co d^ o'2m Do Co © CO o ©* S8 3£ ©CO oo-tf ss cn£ 1 1 1.37 1.04 .845 .737 .540 .409 .333 .290 1.2 1.33 .926 .666 .4S6 .524 .364 .262 .191 1.25 1.31 .896 .624 .434 .516 .353 .246 .171 1.3 1.36 .940 .662 .472 .535 .370 .260 .186 1.4 1.69 1.30 1.05 .896 .666 .512 .413 .353 1.5 2.74 2.13 1.72 1.52 1.16 .838 .677 .598 1.6 4.82 3.62 2.75 2.21 1.90 143 1.08 .870 1.7 9.41 6.25 4.23 3 07 3.71 2.46 1.67 1.21 675 676 ELECTRO-CHEMISTRY. ELECTRO- METALLURGY. CD a g •^ cog P-S- - ' KjJ -: . y,- b:.i s, ~ - 2. ^ - - ^ EN •-< - -.r- WJJ ;~ 02& o 33 F OT3 co Aluminiumf . . Antimony . . . Bromine .... Calcium .... Carbon .... Chlorine .... Copper (cupric) Copper (cupreus) . Gold Hydrogen . . . Iodine Iron (ferric)t . . Iron (ferrous) . . Lead ..... Magnesium . . . Manganese . . . Mercury (mercuric) Mercury (mercurous Nitrogen .... Oxygen .... Platinum (platinie) Platinum (platinous Potassium . . . Silver ..... Sodium .... Tin (stannic) . . Tin (stannous) . . 2 P P ~< gs orq «• «■ i— aq r/Q w crc; C CO CO £ £ C — ^ >-. & ~ 2 53 CS P-C S 1 26.9 119.5 79.34 39.8 11.9 35.18 63.1 63.1 195.7 1.000 125.89 55.6* 55.6 205.36 24.1 54.6 198.5 198.5 58.25 13.93 15.88 193.4 193.4 38.82 107.11 22.88 118.1 118.1 64.9 os P CO E. 8.965 39.83 79.34 19.90 2.975 35.18 31.55 63.10 65.23 1.000 125.89 18.53 27.80 102.68 12.05 27.30 99.25 198.50 29.125 4.64 7.94 48.35 96.70 38.82 107.11 22.88 29.525 59.05 32.45 Is. i T ? | Electro-Chemi- cal Equiva- lents. Grammes per Coulomb. 0000936 0004157 0008281 0002077 0000310 0003672 0003293 0006586 0006809 00001044 0013140 0001934 0002902 0010718 0001258 0002850 0010360 • 0020719 0003040 0000484 0000829 0005047 0010094 0004052 0011180 0002388 0003082 0006164 0003387 0.3370 1.4965 2.9812 0.7477 0.1116 1.3219 1.1855 2.3710 2.4512 0.0376 4.7304 0.6962 1.0447 3.8585 0.4529 1.0260 3.7296 7.4588 1.0944 0.1742 0.2984 1.8169 3.6338 1.4587 4.0248 0.8597 1.1095 2.2190 1.2193 ►1 2.9674 0.6682 0.3354 1.3374 8.9606 0.7565 0.8435 0,4218 0.4080 26.5957 0.2114 1.4364 0.9576 0.2592 2.2080 0.9747 0.2681 0.1340 0.9137 5.7405 3.3512 0.5504 0.2752 0.6855 0.2485 1.1632 0.9013 0.4506 0.8201 |S,3 g|1 co ft co ►0 O p N S^ o- 000743 003299 006572 001648 000246 002914 002614 005228 005404 000083 010429 001535 002302 008506 000998 002262 008222 016444 002413 000384 000658 004006 008012 003216 008873 001895 002446 004892 002688 1346.0 303.1 152.1 606.6 4064.5 343.1 382.6 191.3 185.1 12063.6 95.9 051.5 434.4 117.6 1001.5 442.1 121.6 60.8 414.4 2603.8 1520.1 249.7 124.8 310.9 112.7 527.6 408.8 204.4 37'\0 Ph~ CO ELECTRO-CHEMISTRY. 677 Resistances of Sulphate of Copper at 10° C1. or 50° tP. (Ewing and MacGregor.) Ohms per Ohms per Density. c.c. Cu. In. c.c. Cu. In. 1.0167 164.4 64.8 1.1386 35.0 13.8 1.0216 134.8 53.1 1.1432 34.1 13.4 1.0318 98.7 38.8 1.1679 31.7 12.5 1.0622 59.0 23.2 1.1829 30.6 12.0 1.0858 47.3 18.6 1.2051 | 29.3 11.5 1.1174 38.1 15.0 Saturated ) Resistances of Sulphate of Zinc at 10° C or 500 V. Ohms per Ohms per Density. Density. c.c. Cu. In. c.c. Cu. In. 1.0140 182.9 72.0 1.2709 28.5 11.2 1.0187 140.5 55.3 1.2891 28.3 11.1 1.0278 111.1 43.7 1.2895 28.5 11.2 1.0540 63.8 25.1 1.2987 28.7 11.3 1.0760 50.8 20.0 1.3288 29.2 11.5 1.1019 42.1 16.6 1.3530 31.0 12.2 1.1582 33.7 13.3 1.4053 32.1 12.6 1.1845 32.1 12.6 1.4174 33.4 13.2 1.2186 30.3 11.9 1.4220 | 33.7 13.3 1.2562 29.2 11.5 Saturated j Specific resistance of fused sodium chloride (common salt) at various temperatures. Temperature Cent. 720° 740° 750° 770° 780° Ohms per cu. cm. .348 .310 .294 .265 :247 Application of Electro-Chemistry. The various forms of primary and secondary batteries may be regarded as applications of electro-chemistry, but they are treated as special subjects in other parts of this book. Other important practical applications are the processes for producing chemicals by electrolysis or by electrical heating. Among the materials thus produced in large quantities are caustic soda, carbonate of soda, chlorine, bleaching powder, chlorate of potash, calcium carbide, phosphorus, cyanide of potassium, etc. The production of caustic soda may be effected by electrolysing a solution of common salt tlie reaction being NaClx H20=NaOHx Hx CI the products being caustic soda {NaOH) which remains in solution, hydrogen and chlorine that pass off as gases the latter being collected and used for mak- ing bleaching powder. There is a tendency to form a mixed product of caustic soda and salt and a certain amount of hypochlorite of soda. These difficulties are avoided by separating the caustic soda from the rest of the solution either by a porous diaphragm or by drawing it off as fast as produced. In the Castner process, mercury is used as the cathode and absorbs the metallic sodium deposited upon it. In another chamber the sodium decomposes water and forms caustic soda. 678 ELECTRO-CHEMISTRY. ELECTRO-METALLURGY. Calcium Carbide is produced by heating a mixture of burnt lime and pulverized coke or anthracite coal in an electric furnace, the reaction being: CaO-\-3C=CaC2+CO The carbonic oxide (CO) passes off as a gas and the calcium carbide after cooling is a solid grayish mass which is broken up for use. A rotary form of furnace is used at the large works of the Carbide Company at Niagara Falls, the material being fed in at one side and the calcium carbide being taken out at the other. ELECTRO-METALLURGY. Electro-metallurgy may be defined as that branch of science which re- lates to the electrical reduction or treatment of metals. The subject may be divided into three important and quite distinct branches, as follows: 1. Electrolytic Metallurgy, which consists in reducing or separat- ing metals by the decomposing effect which occurs when an electric current is passed through their compounds while in the liquid state. These com- pounds may be rendered liquid either by dissolving or fusing them; hence there are: (a.) Wet methods with solutions. (b.) Dry methods with fused materials. Electrolytic metallurgy is applied to the following purposes: (c.) Electrotyping, which is the art of reproducing the exact form of type, engravings, medals or other articles by electrodepositing metal on the article itself or on a mould obtained from it. (d.) Electroplating, Avhich is the art of coating articles with an adherent layer of metal by ele'ctrodeposition. (e.) Electrolytic reduction of metals, which is the art of obtaining metals from their ores or compounds by electrically decomposing such ore or compound in the state of solution or fusion. (/.) Electrolytic refining of metals, which is the art of eliminating im- purities by electrodepositing the metal itself, the foreign substances being left in the anode or liquid, or vice versa. 2. Electrical smelting*, which consists in reducing metallic oxides by carbon at a high temperature produced by the passage of an electric current. 3. Electrical working- of metals, which consists in treating metals mechanically with the aid of heat generated by electric currents. Various mechanical processes which are facilitated by softening or fusing the metal may be effected in this way, the principal ones being: welding, forging, rolling, casting. Electrotyping.— To reproduce an engraving, typographical composition, or other object, a mould of gutta percha, wax, piaster or fusible alloy is made from the object. If it is not a conductor it is coated with graphite to start the action, connection being made to it by a wire or clamp put around it. It is used as the cathode in a bath consisting of a saturated solution of copper sulphate acidulated with sulphuric acid. The anode is a plate of copper. The ordinary thickness of deposit is .01 to .02 inch. The " shell" thus formed is separated from the mould and backed by a filling of type metal. Electroplating an article with an adherent coating of metal requires the article to be thoroughly cleaned mechanically and chemically. Cleaning. — Solutions for cleaning Gold, Silver, Copper, Brass and Zinc are prepared as follows: Water. Nitric Acid. Sulphu- ric. Hydro- chloric. For copper and brass Silver 100 100 100 100 100 50 10 3 100 10 8 12 2 Zinc Iron, wrought Iron, cast 2 3 ELECTRO-METALLURGY. 679 Lead, Tin, Pewter, are cleaned in a solution of caustic soda. Objects to be plated witb gold or silver must be carefully and thoroughly freed from acids before transfer to the solutions. Objects cleaned in soda or those cleaned in acid for transfer to acid coppering solutions may be rinsed in clean water, after which they should be transferred immediately to the depositing solution. Baths for plating:.— The reader is referred to the various books on electroplating for particulars, as but few, and those the most used solutions can be referred to here. Solutions should be adapted to the particular object to be plated, and must have little if any action upon it. Cyanide of gold and silver act chemi- cally upon copper to a slight extent and the objects should be connected to the electrical circuit before being immersed. Solutions are best made chemically, but can be made by passing a current through a plate of the required metal into the solvent. Copper. —A good solution for plating objects with copper is made by dissolving in a gallon of water 10 ounces potassium cyanide, 5 ounces copper carbonate, and 2 ounces potassium carbonate. The rate of deposit should be varied to suit the nature and form of the surface of the object, large smooth surfaces taking the greatest rate of deposit. Electrotype plates must be worked at a slow rate, owing to the rough and irregular surface. Non-metallic Surfaces may be plated by first providing a conducting sur- face of the best black lead or finely ground gas coke. Care is required in starting objects of this sort, to obtain an even distribution of the metal, and hollow places may be temporarily connected by the use of fine copper wire. Copper on iron or on any metal that is attacked by copper sulphate is effected by an alkaline solution. One which can be worked cold is made up of \ ounce of copper sulphate to a pint of water. Dissolve the copper sulphate in a half pint of water, add ammonia until all the first formed precipitate re-dissolves, forming a deep blue solution, then add cyanide of potassium until the blue color disappears. A heavy current is required with this solution, enough to give off gas from the surface. This solution will deposit at a high rate but ordinarily leaves a rough and crystalline surface, and will not do good work on steel. A cyanide solution is the most used, takes well on steelpr brass, as well as on iron, and permits of many variations. For each gallon of water use : Copper carbonate 5 ozs. Carbonate of potash 2 ozs. Potassium cyanide, chem. pure 10 ozs. Dissolve about nine-tenths of the potassium cyanide in a portion of the water then add nearly all the copper carbonate, Avhich has also been dis- solved in a part of the water: dissolve the carbonate of potash in water and add slowly to the above solution stirring slowly until thoroughly mixed. Test the solution with a small object, adding copper or cyanide until the deposit is uniform and strong. For coppering before nickel plating, the coating of copper must be made thick enough to stand hard buffing, and for this reason the coppering solution must be rich in cyanide and have just enough copper to give a free deposit. Use electrolytically deposited copper for anodes, as it gives off copper more freely. Regulate the current for the work in the tanks, and it should be rather weak for working this solution. Brass Solutions of any color may be made by adding carbonate of zinc in various quantities to the copper solution. The zinc should be dissolved in water with two pai'ts, by weight, of potassium cyanide, and the mixture should then be added to the copper bath. A piece of work in the tank at the time will indicate the change in color of the deposit. Two parts copper to one zinc gives a yellow brass color. For the color of light brass add a little carbonate of ammonia to the brass solution. To darken the color add copper carbonate. Varying the amount of current will also change the color, a strong current depositing a greater amount of zinc, thus pro- ducing a lighter color. Silver. — The standard solution for silver plating is chloride of silver dissolved in potassium cyanide. This solution consists of 3 ounces silver chloride with 9 to 12 ounces of 98 percent potassium cyanide per gallon of water. Rub the silver chloride to a thin paste with water, dissolve 9 680 ELECTRO-CHEMISTRY. ELECTRO-METALLURGY. ounces potassium cyanide in a gallon of water and add the paste, stirring until dissolved. Add more cyanide until the solution works freely. The bath should be cleaned by filtering. Great care should be taken to keep the proper proportions between current, silver and cyanide. A weak cui'- rent requires more free cyanide than a strong one, and too much cyanide prevents the work plating readily, and gives it a yellowish or brownish color. If there is not enough cyanide in the solution the resistance to the current is increased and the plating becomes irregular. The most suitable current for silver plating seems to De about one ampere for each sixty (60) inches of surface coated. Gold. — Cyanide of gold and potassium cyanide make the best solution for plating with gold. The solution is prepared in the same manner as the silver solution just described, using chloride of gold in place of chloride of silver. The electrical resistance of the bath ic controlled by the quantity of cyanide, the more cyanide the less the resistance, cut an excess of cyanide produces a pale color. Hot baths for hot gilding require from 11 to 20 grains of gold per quart of solution and a considerable excess of cyanide. Baths for cold gilding and for plating should have not less than 60 grains per quart and may have as much as 320 grains, this quantity being used with a dynamo current for quick dipping. Wickel. —The solution now almost universally used for nickel plating is made up from the double sulphate of nickel and ammonia, with the addition of a little boracic acid under certain conditions. The double salt is dissolved by boiling, using 12 to 14 ounces of the salts to a gallon of water, the bath is then diluted with water until a hydrometer shows a density of 6.5° to 7° Baume. Oast anodes «.re to be preferred as they give up the metal to the solution more freely. Anodes should be long enough to reach to the bottom of the work and should have a surface greater than that of the objects being plated. Current strength should be moderate, for if excessive the work is apt to be rough, soft or crystalline, voltage may vary from 3.5 to 6 volts and the most suitable current is from .4 to .8 ampere per 15 square inches surface of the object. Zinc is the only metal requiring more current than this, and takes about double the amount named. A nickel bath should be slightly acid in order that the work may have a suitable color. An excess of alkali darkens the work and an excess of acid causes " peeling." Iron. — A hard Avhite film of iron can he deposited from the double chloride of iron and ammonia, which can be prepared by the current process. It is somewhat used for coating copper plates to make them wear a long time, the covering being renewed occasionally. The Electro-motive Forces suited to the different metals are : — Copper in sulphate, Volt, ^-l* " cyanide, . . 4* - 6- Silver in " 1- -2- Gold in " -5-3- Nickel in sulphate, "5-1" The Resistance will depend on the nature of the surface. "Work is best effected with about equal surface of anode and objects, and the coating will be more even, the greater the distance between them, especially where there are projecting points or rough surfaces. Copper and silver should never show any sign of hydrogen being given off at the objects; gold may show a few bubbles if deep color is wanted. Nickel is always accompanied with evolution of hydrogen, but the bath should not be allowed to froth. The Rate of Deposit is proportional to current, as described under the head of " Electrolysis," in the proportions given in the table of electro- chemical equivalents except in the case of gold, the equivalent of which in combination with cyanogen is 195.7, but subject to modifications dependent upon the hydrogen action just described; there is also a partial solution of the metal, so that there is always a deduction to be made from the theoret- ical value. Thus : — Gold gives about 80 to 90 per cent. Nickel " 80 to 95 Silver " 90 to 95 Copper "98 " ELECTROLYTIC REFINING OP COPPER. 681 An ampere of current maintained for one hour, which serves as a unit of quantity called the " ampere hour," represents Gramme 0376 Grain 58 Ounce Troy 00121 Ounce Avoir. . . .00132 which multiplied by the chemical equivalent will furnish the weight of any substance deposited. Separation of jfletals. Aluminum. — There are several successful processes in use. HaWs process is operated on a large scale at Niagara Falls. The cell is an iron vessel lined with carbon, which forms the cathode, and contains molten cryolite (sodium and aluminum double fluoride), into which is fed the alumina, Al20-6, ; this is electrolysed, the oxygen passes oft as C02 at the anode, which is a carbon cylinder. The aluminum having a higher specitic gravity than the fluoride, settles at the bottom of the bath, from which it is tapped or ladled off. The temperature of the bath is 1,600° to 1,800° Fahr., while from 7 to 8 volts are required, and a current of 5,000 amperes is used, producing 1 pound of metal per 10 K. W. - hours. About 1 pound of carbon electrode is consumed per 1 pound of aluminum produced. The Cow les process is chiefly for producing alloys of aluminum and sili- con with copper and iron. Corundum (aluminum oxide) or bauxite is mixed with iron tilings or granulated copper, and is smelted in a furnace as fol- lows : — The furnace pit is built of fire brick with holes in the ends for admitting the carbon electrodes ; the furnace is lined internally with limed charcoal, the lime keeping apart the carbon particles, which would other- wise connect and make a short circuit. The carbon electrodes are brought together and the charge of corundum, &c, is put in, the furnace is then covered, and the current is gradually started. The electrodes are then gradually separated, and the current is increased and maintained for about an hour, when the reduced metal is drawn from the bottom of the furnace. \Vith the cupro-aluminum process the current is easily maintained steady, but with the ferro-aluminum process the conductivity of the charge varies greatly during the process, and regulation of current is very difficult. Electrolytic Refining- of Copper. The most important application of electrolytic metallurgy is the refining of copper which is carried on at many places in this country and abroad on a very large scale. The crude copper obtained from the smelting furnaces is cast or rolled in the form of plates which are used as anodes in electro- lytic cells. Theelectrolyte is a solution of copper sulphate acidulated with sulphuric acid to increase its conductivity. The cathodes are usually thin sheets of pure copper upon which the refined copper is electrodeposited, the impurities are left behind in the anodes or solution, or as a scum or sediment. In some cases the plates are arranged in series and in others in parallel. The former has the advantage of requiring electrical contracts to be made to the first and last plates only, whereas the parallel plan re- quires connection to each plate; but in the series arrangement there is a considerable leakage of current amounting to about 15 or 20 per cent. The pressure required is from .2 to .4 volt per cell with a current density of 10 to 15 amperes per square foot. It requires in practice 400 to 475 ampere-hours per pound of copper, the theoretical amount being 382.6 ampere-hours. About 8 or 9 pounds of copper are produced per kiloAvatt-hour at about .3 volt which is the ordinary value. The cost of the process is about .7 cent per pound of copper. A great advantage of the electrolytic method of refin- ing copper is the fact that the silver and gold contained in the copper is left behind in the sediment, from which it is extracted afterward usually by electrolysis. The silver and gold thus recovered constitute an important item in the output of an electrolytic refinery. The Elmore process consists in depositing the copper on a revolving iron mandrel which forms the cathode ; an agate burnisher travels along the mandrel and presses the crystals of metal into a fibrous form which is said to account for the superior strength of the metal deposited by this process. The copper is removed from the mandrel by expansion, for which purpose 682 ELECTRO-CHEMISTRY. ELECTRO-METALLURGY. steam is used. Specimens tested by Prof . Kennedy have broken at 27 to 41 tons per square inch with an extension of 5 to 1\ per cent. Tbe tubes may be cut into sheets or strips for drawing into wire. The conductivity is very high, being sometimes 2 or 3 per cent above Matthiessen's standard. Silver is refined from copper bullion by taking anodes of the bullion \ inch thick and 14 inches square, and cathodes of sheet silver slightly oiled. The electrolyte consists of water with 1 per cent of nitric acid. When the current is started the copper and silver form nitrates of copper and silver and free nitric acid from which the silver is deposited, leaving the copper in solution. Trays are placed under the cathode for catching the deposited silver, and if there is any copper deposited owing to the solution contain- ing too little silver or a superabundance of copper, the copper falls into the trays and is re-dissolved. In the Moebius process the deposit is continually removed from tbe cath- ode by means of a mechanical arrangement of brushes, and falls into the trays above mentioned. ELECTRIC HEATING, COOKING AND WELDING. HEAl UMTS AND EftUIVALEXTi. The unit of heat in mechanics is the " calorie" or " lesser calorie," which is the heat necessary to raise one cubic centimeter of water from 4° to 5° Centigrade in one second. The British Heat Unit, known as the " British Thermal Unit," or " B.T.U.," is the quantity of heat necessary to raise one pound of water from 60° to 61° Fahrenheit, and is equal to 778 foot pounds, or 1055 Joules. The Joule is the heat generated by a watt in a second. Joule's Law shows that the heat generated in a conductor is directly proportional to : Its resistance, the square of the current strength, and the time during which the current flows, or, H— c2m. According to Ohm's law, C= E -f- B, hence, C*Et = J Gilt = EC't = ~ And calling Q the quantity of electricity flowing, then and H= EQ or the heat = E.M.F. X Quantity, in which E.M.F. is the difference of potential between the end of the conductor. The table on the following page clearly shows the equivalent values of the electrical and mechanical units. VAHIOUi METHODS OF UTILIZING THE HEAT GENERATED BY THE EIECTSIC CERHEHT. I. Metallic Conductors (Uninterrupted Circuit). 1. Exposed coils of wire or strips. (a) Entirely surrounded by air. (b) Wound around insulating material. 2. Wire or strips of metal imbedded in enamel. (a) In the form of coils. { Leonard, Carpenter, Crompton, and (b) In flat layers. j others. 3. Wire or strips of metal imbedded in asbestos. (a) In the form of coils. (6) In flat layers. 4. Wire imbedded in various insulating compounds. (a) Crystallized acetate of sodium, etc. Tommasi. 5. A Film of metal. (a) Rare metal fired on enamel. ) Prntnptllw1, (6) Rare metal fired on mica. f Riometheus. (c) Silver deposited on glass. Reed. 6. Sticks of metal. (a) Crystallized silicon in tubes of glass. Le Roy. (b) Metallic powder mixed with clay and compressed- Parville\ 684 ELECTRIC HEATING, COOKING, AND WELDING. £ .1 A ,_; 1 O o3 P rt • • 2 © So N ft ^=H S JS ft § 1 1 '8S1 do «i;bos 3 £ 22l© {ft o 0 O 3 ^ ^ ^ a; • -? 3 t ■"! 3 ^ "- £ - ~ ■~ a ~ t lo ci « ^38§ •+1 ^ N ^J ^mM^'J^ 1-1 th c-icq^io^rt t"i ':: ~ - "'S'j p w| © . . 5 El i = ll al^ll 0 £+-l 0 IT © 2 -H i 5 ft 0 p ft G • ■'^'g S3 © 5 aa^S » p . fe © ^ ft o SgS. t,o > 0 O m 0 . b o~ p,-S 3 ft "S g <* £ ^^§ a ^S % ^ H • © "3 ft" •"£ 'r2 *"- 3 " £ .S^ 2 % %£ & x, 2 .- co c 2: -f ~r 1- 1- 2 -* .£ t> l . CO tQ CI ■■# L . . <>J t 5 -' ©ft S{£ aft-* -; 23 4 <■- +" 12 7"; - - 2 fcicco ^ 0 ^2888 P '2 ft!» '? iJ0H S^ sj ^P©i^i P CO co^-=3 a" co rt. p II P © -111 ha II t^ ©_a ffl ^ ^ H 5 "2 ft ,- CO ■ u (§ ^ ^ «H ©-P 5 o a" © P, © 'S 0 s N ^ P ^ . ^ © . Oh^ 2 5 a 2 -3 ° 2 • f1 -a 3 7" 2 ;3J2 -"■'- 1 so2> 9 g £ . £ g pj © ® "§ © .s^,«rtaH, ci !> © "3 2 2-9 2 &§££ ^0 +5 T • «* © >fe rife ^"Sj? "S lo *o |i-*r^|i2«|s3 &2?'2!-£J22;2f^2 s^ -h-h coct^'^ m1-1 ? --+:,- :, ■ _ ~ x !- ■'- ~ 0:0 co _^ ££ -^._- -t -;1 •" rHH Cl" -# t- MIO °? ^ CO HHNMMM-: CO (M L~ i-Tcf <^ ^<^^' ^ ; TIME IN MINUTES Fig. 1. Efficiency of Heating- Apparatus. In the foregoing references it Avill be seen that the efficiency of electric cooking apparatus varies from about 63 per cent to 90 per cent (for ovens), depending upon a number of variable conditions, such as time, size, quantity to be heated, temperature rise, etc. According to Mr. Crompton, the efficiency of an ordinary cooking-stove using solid fuel is only about 2 per cent, 12 per cent being wasted in obtain- ELECTRIC CAR HEATING. 689 ing a glowing fire, 70 per cent going up the chimney, and 16 per cent being radiated into the room. In a gas-stove, considering that the number of heat units obtainable from the gas at a certain price is but small compared with solid fuel, the venti- lating current required for the operation alone consumes at least 80 per cent of the heat units obtained by burning the gas. In the case of an electrical oven, more than 90 per cent of the heat energy can be utilized ; and thus, although possibly 5 to 6 per cent only of the heat energy of tbe fuel is present in the electrical energy, 90 per cent of this, or 4£ per cent of the whole energy, actually goes into the food, and thus the electrical oven is practically twice as economical as any other oven, whether heated by solid fuel or by gas. ELECTRIC RAUIATORi, Unless electricity is produced at a very low cost it is not commercially practicable to heat residences or large buildings. While this is true, the electric heater still has a field of application, in heating small offices, bathrooms, snuggeries, cold corners of rooms, street railway waiting rooms, the summer villa on cool evenings, and in mild climates a still wider range. It has the peculiar advantage of being instantly available, and the amount of heat is regulated at will. The heaters are perfectly clean, do not vitiate the atmosphere, and are portable. According to Houston and Kennelly, one joule of work expended in producing heat will raise the temperature of a cubic foot of air about is ° F- The amount of power required for electrically heating a room depends greatly upon the amount of glass surface in the room, as well as upon the draughts and admission of cold air. In order to make a comparison between heating an ordinary city house by means of coal burnt in a furnace and by electricity furnished by a cen- tral station, let it be assumed that 100 lbs. of coal are consumed per day in the furnace. Assuming the furnace to have an efficiency of 50 per cent, 50 lbs. of coal are utilized throughout the building in the form of heat. Reducing this to actual horse-power we have 700,000 700,000 X 778 = 544,600,000 ft.-lbs. 544,600,000 _ 33,000 16,503 : 16,503 H.-P. minutes. - = 275 H.-P. hours. Assuming that a H.-P. hour is furnished at 5 cents the cost would be 275 X .05 = $13.75. ELECTRIC CAM HEiTIAG. At the Montreal meeting of the American Street Railway Association in 1895, Mr. J. F. McElroy read an exhaustive paper on the subject of car- heating, from which the following abstracts are taken : In practice it is found that 20,000 B. T. U. are necessary to heat an 18 to 20 foot car in zero weather. When the outside temperature is 12J- ° F. only 16,000 B. T. U. are required, etc., which shows the necessity of hav- ing electric heaters adjustable. The amount of heat necessary in a car to maintain a given inside tem- perature, depends on : 1. The amount of artificial beat which is given to it. 2. The number of passengers carried. The average person is capable of giving out an amount of heat in 24 hours which is equal to 191 B. T. TJ. 690 ELECTRIC HEATING, COOKING, AND WELDING. Cost of Car Heating-. The following table was compiled by Mr. McElroy from the reply re- ceived from the Albany Railway Company : Average fuel cost on Albany Railway, per amp. hour = .241 cent. Average total cost for fuel, labor, oils, waste, and packings per amp. hour = .423 cent. Cost of fuel per hour for heating a car with electric beaters with coal at $2.00 per 2000 lbs. Position of Switch. Amperes equal. 2.88 6.88 8.09 cts. cts. cts. .58 1.40 1.62 .54 1.30 1.51 .52 1.27 1.47 .48 1.17 1.36 Simple high speed condensing . . Simple low speed condensing . . Compound high speed condensing Compound low speed condensing 2.41 2.24 2.20 2.03 Average Cost Per Dav for Stores. 33 lbs. of coal at $4.55 per ton $.075 Repairs 005 Dumping and removing coal and ashes, coaling up and kindling fire, including cost of kindling, and part of cleaning car 100 Removing stoves for summer, installing for win- ter, repairing head linings, repainting, etc., average per day 0125 Total $.1925 ELECTKIC WELDIKG. 691 EIECTKIC IROXS EOIft DOMESTIC AEf» ODUII- TBIAI PURPOSES. Comparing the hand-irons heated by gas with those heated electrically, it is claimed that if gas can be purchased at 81.25 per 1000 cu. It., and the cost of electricity is about 1 cent per H. P., the two systems are about on a par, as far as cost only is concerned. According to the American Electric Heating Corporation, the power con- sumption for the various types of irons is as follows : — Watts 4 lbs. Troy Polishing, diamond face 330 Q lbs. Small Seaming (can be connected to lamp socket) . . . 200 4 lbs. Gentleman's Small Hat Iron 200 5£ lbs. Light Domestic 500 5A- lbs. Light Domestic, round nose ; . . 500 7 "lbs. Domestic 600 9 lbs. Heavy Laundry 680 9 lbs. Hatters' 550 9 lbs. Corset 500 15 lbs. Hatters' Factory 550 5£ lbs. Morocco Bottom 500 Morocco Bottom, round nose 500 XIECTRIC WELDHfG JLND FOR^OG. The current employed in electric welding may be either continuous or alternating. By the use of alternating currents, a slightly more uniform heating of the contact surfaces is obtained, because alternating currents tend to develop a greater heat at the surface of a large mass than at the central portions. Thomson Electric Welding* Process. The principle involved in the system of electric welding, invented by Prof. Elihu Thomson, is that of causing currents of electricity to pass through the abutting ends of the pieces of metal which are to be welded, thereby generating heat at the point of contact, which also becomes the point of greatest resistance, while at the same time mechanical pressure is applied to force the parts together. As the current heats the metal at the junction to the welding temperature, the pressure follows up the softening surface until a complete union or weld is effected ; and, as the heat is first developed in the interior of the parts to be welded, the interior of the joint is as efficiently united as the visible exterior. Horse-Power Used in Electric "Welding. The power required for the different sizes varies nearly as the cross sec- tional area of the material at the joint where the weld is to be made. Within certain limits, the greater the power, the shorter the time ; and vice versa. The following tables are based upon actual experience in various works, and from very careful electrical and mechanical tests made by reliable experts. The time given is that required for the application of the current only. Round Iron or Steel. Diameter. Area. H.-P. Applied to Dynamo. Time in Seconds. iin. .05 2.0 10 IS: .10 4.2 15 .22 6.5 20 fin. .30 9.0 25 Jin. .45 13.3 30 692 ELECTRIC HEATING, COOKING, AND WELDING. Extra, Heavy Iron JPipe. Inside Area. H.-P. applied Time in Diameter. to Dynamo. Seconds. i in. .30 8.9 33 f in .40 10.5 40 1 in .60 16.4 47 li in .79 22.0 53 1J in 1.10 32.3 70 2 in 1.G5 42.0 84 2h in 2.25 63.7 93 3 in 3.00 96.2 106 General Table. Iron and Steel. Copper. Area in Time in H.-P. applied Area in Time in H.-P. applied sq. m. Seconds. to Dynamos. sq. in. Seconds. Dynamos. 0.5 33 14.4 .125 8 10.0 1.0 45 28.0 .25 11 23.4 1.5 55 39.4 .375 13 31.8 2.0 65 48.6 .5 16 42.0 2.5 70 57.0 .625 18 51.9 3.0 78 65.4 .75 21 61.2 3.5 85 73.7 .875 22 72.9 4.0 90 83.8 1.0 23 82.1 Axle Welding-. V round axle requires 25 Horse-power for 45 seconds. \" square " \\" round " 30 " 35 \\" square " 2// round " 40 75 2" square " 90 The slightly increased time and power required for welding the square axle is not only due to the extra metal in it, but in part to the care which it is best to use to secure a perfect alignment. \\" x §" 2" xf 2" xf Tire "Welding-. tire requires 11 Horse-power for 15 seconds. 23 " 23 " 62 ictual The time above given for welding is of course that required for the a application of the current only, and does not include that consumed ny placing the axles or tires in the machine, the removal of the upset, and other finishing processes. From the data thus submitted, the cost of welding can be readily figured for any locality where the price of fuel and cost of labor are known. HYDRO— ELECTROTHERMIC SYSTEMS. 693 HYDBO-E&ECTIIOTHERMIC SYSTEMS. XSolio and Eagrange System. In this system an electrolytic bath is employed, into which an electric current of considerable E.M.F. is led, passing from the positive pole which forms the boundaries of the bath and presents a large surface to the elec- trolyte and thence to the negative pole, consisting of the metal or other material to be treated, and which is of relatively small dimensions. Through the electrolytic action hydrogen is rapidly evolved at the nega- tive pole and forms a gaseous envelope around the pole ; as the gas is a very poor conductor of electricity, a large resistance is thus introduced in the circuit, entirely surrounding the object to be treated. The current in passing through this resistance develops thermal energy, and this is com- municated to the metal or other object which forms the negative pole. This system has been extensively used in England, and is described in The Electrical World, Dec. 7, 1895. ISurton Electric Eorge. In a patent granted to George D. Burton on an electrolytic forge, the portion to be heated is placed in a bath consisting of a solution of sal soda, or water, carbonate of soda, and borax. The tank is preferably made of porcelain or fire-clay. The anode plate has a contact surface with the liquid much greater than the area of contact of the article to be heated. This plate is composed of lead, copper, carbon, or other suitable conducting material. Zerener System. In this system an arc is used in combination with a magnet which deflects the arc, making a flame similar to that of a blow-pipe, but having the tem- perature of the arc. The apparatus contains a self-regulating device which is driven by a small electric motor ; for welding iron a current of 40 to 50 amperes at 40 volts will suffice for strips of metal three mm. thick. Hernardos System. In this system the article to be operated upon is made to constitute one pole of the electric circuit, while a carbon pencil attached to a portable insulated holder, and held by the workman, constitutes the other pole, the electx-ic arc — which is the heating agent of the process — being struck between the two poles thus formed. This system has been used extensively in England for the repair of machinery. The Barrbeat-Strange Patent Barrel Syndicate use this system for the welding of the seams of sheet- steel barrels. Voltex Process for "Welding- and Brazing- Consists in the use of an electric arc formed between two special carbon rods inclined to each other at an angle of about 90°. The whole apparatus can generally be held in one hand. With gas and coke, gas costing only 70 cents per 1000 cubic feet, it is claimed the complete cost of brazing and filling up a bicycle frame is $1.43, while with the Voltex process, at 6 cents per kilowatt hour, it is only 46 cents. Stassano Process of Electric Smelting Consists of heating, in an arc furnace, briquettes composed of iron ore, carbon, and lime made into a paste with tar. The smelting process occurs in a blast furnace, the iron being reduced, and the siliceous matter of the ore slagged off. Annealing- of Armor JPlate. The spot to be treated is brought to a temperature of about 1000 ° F. The current used is equivalent to 40,000 amperes per square inch, a density which is only possible by the use of cooling by water circulation. The operation generally takes seven minutes. 694 ELECTRIC HEATING, COOKING, AND WELDING. Electric Mail Welding-. The " Electric " joint, applied by the Lorain Steel Co., is made by welding plates {on both sides of the web of the rail. The plates shown in Fig. 4 are 1 inch by 3 inches, by 18 inches, and have three bosses, three welds OIAGRAM OF CONNECTIONS OF RAIL WELDER (^T t • trolley c . b • circuit breahir r.r-rheostats m • icotqr b - booster RT' ROTARY TRANSFORMER W.T WELDING TRANSFORMER 3 Vf; SWITCH SKETCH OF BAR. USED IN WELDING o <3 ... *&■, 8' 1 ¥ , X | («0i) * ' " Web Plates Figs. 3 and 4. — The Lorain Steel Company Method of Electric Welding. being made at each joint. Great pressure up to 35 tons is maintained on the joint whilst making and cooling. The welding current runs as high as 25,000 amperes. The connections are shown in Fig. 3. FT§E DATA. In a lecture on " The Eating and Behavior of Fuse Wires," before the A. I. E. E., in October, 1895, Messrs. Stine, Gaytes, and Freeman arrived at the following conclusions : 1. Covered fuses are more sensitive than open ones. 2. Fuse wire should be rated for its carrying capacity for the ordi- nary lengths employed. 2 (a). When fusing a circuit, the distance between the terminals should be considered. FUSE DATA. 695 On important circuits, fuses should be frequently renewed. The inertia of a fuse for high currents must be considered when protecting special devices. Fuses should be operated under normal conditions to ensure cer- tainty of results. Fuses up to five amperes should be at least 1| inch long, one- half inch to be added for each increment of five amperes capacity. Round fuse wire should not be employed in excess of 30 amperes capacity. For higher currents flat ribbons exceeding four inches in length should be employed. (For additional data on Fuses see p. h60.) SOME NOTES ON THE OPERATION OP ELECTRIC MINING PLANTS. From Pamphlet by General Electric Company. Mr. F. J. Piatt of the Scranton Electric Construction Company, Scranton, Pa., gives some figures on electric haulage. They are from plants which have been in operation for one year or longer. The expenses given are the actual figures for labor, oil, repairs, etc. In figuring the cost of mule-power, the cost per mule has been taken at 50 cents per working-day, which includes feed, attendance, medicine, shoe- ing, harness, and the item of mortality. Depreciation on the electric plant is figured at 5%, and is given per working-day. The first plant on whicli Mr. Piatt presents figures is the Green Ridge Colliery, installed in March, 1895, for Mr. O. S. Johnson, in the city of Scranton. The Green Ridge Colliery. The Green Ridge Colliery plant consists of one 100 H.P. automatic, high- speed engine, and one 75 H.P. dynamo, with switchboard and station equipment, all of which are installed in a frame building 30 feet by 45 feet. From the dynamo a feeder wire is run down the slope 1,000 feet to the main gangway, where a 6g- ton electric locomotive is in operation over about 1\ miles of trolley road. This locomotive gathers trips from three different points in the mines, and delivers them to the foot of the outside slope. The main gangway, whiuh is very crooked, is about 3,100 feet long, and branching from it are two other roads, one of which is 1,000 feet and the other 2,100 feet in length. For the past year this locomotive has made a daily average of twenty trips, each trip consisting of eight cars, which is very much below its capacity. The grades on the main roads are about 1% in favor of the loaded and ag mist the empty cars. On the 1,000 foot branch the locomotive has about 500 feet of 3% and 500 feet of \% grade against the empty cars. On the 2,100 foot branch the grades are very uneven, and most of them are against the loaded cars. The grades of this road, against the loaded cars, consist ap- proximately of 150 feet of 7% grade, 500 feet of 2% grade, 350 feet of 5£% grade, and 450 feet of 3i-% grade. This 6£ ton locomotive has been hauling trips of four cars up these grades ever since it was installed, and on some days has hauled trips of five cars. The roof of the mine is very low, being about five feet in the highest places ; and as this height was obtained by blowing the roof over the center of the road, the height on the main road will not average much over four feet. This is one difficulty which would have been met had a steam loco- motive been introduced instead of an electric locomotive. Cost of Haulage at the Green Ridg-e Colliery. After very carefully going over all the expenses connected Avith this plant, the following results were obtained : The plant cost .$7,625.18. Depreciation at 5% per year would amount to $381.25, or taking 200 working-days per year the depreciation per working- day would be $1.90. Cost of operation per day is as follows : Station Engineer $1.75 Motorman 1.75 Helper 1.60 Repairs 76 Depreciation 1.90 Oil and waste 20 Total $7.96 OPERATION OF ELECTRIC MINING PLANTS. 697 The coal hauled per day by the electric locomotive is 288 tons, at a cost per ton, as shown above, of 2.76 cents. To haul this coal by mule-power would require Seventeen mules at 50 cents each $8.50 Three drivers at $1.45 each 4.35 Three drivers at $1.25 each 3.75 Four boys at $1.00 each 4.00 Total $20.60 This shows a cost for haulage by mule-power of 7.15 cents per ton, and a saving by electric haulage of 4.39 cents per ton. On the 2S8 tons hauled per day the saving is $12.64, and for a year of 200 working-days it amounts to $2,528.00. This locomotive has averaged 30 miles per day, making a total of about 12,700 miles since it was installed. The expense of repairs taken on the basis of mileage is a trifle over two cents per mile. This statement shows the actual results at this particular plant, and what is being saved per day. The number of mules saved in the above case, is the number that it would require to haul an amount equal to the output of the locomotive on any one day ; but it is doubtful if seventeen mules would be able to do this work continually, as they would interfere with each other on the main roads, and would not deliver the coal as regularly as does the locomotive. Among others referred to are the two electric haulage plants at the mines of the New York and Scranton Coal Company, at Peckvilie, Pa. The figures given are based on the expenses, of the year 1896. The Hew York and Scranton Coal Company. One of the mines operated by the New York and Scranton Coal Company is known as The Sturges Shaft. The plant consists of a 160 H.P. engine and. generator and a 6£ ton locomotive, operating over 4,500 feet of trolley road. The cost of the plant was $6,103.00. The depreciation per year at 5% would amount to $305.15, or for 200»wor king-days, $1.52 per day. Cost of operation per day is as follows : Motorman $1.75 Helper 1.25 Electrician .78 Repairs 1.03 Depreciation 1.52 Oil 24 Total $6.57 The coal hauled per day is 250 tons, at a cost per ton, as shown above, of 2.62 cents. To haul this coal by mule-power would require Fourteen mules at 50 cents each $7.00 Seven boys at $1.35 each 9.45 Total $16.45 This shows a cost for haulage by mule-power of 6.58 cents per ton, and a saving by electric haulage of 3.96 cents per ton. On the 250 tons hauled per day the saving is $9.90, and for a year of 200 working-days it amounts to $1,980.00. 698 ELECTRICITY IN MINES. The locomotive runs about 32 miles per day, and up to tbis time has covered about 7,800 miles, with a cost for repairs of 2.7 cents per mile. The otlier haulage plant operated by the New York and Scranton Coal Company is located at the tunnel opening. The cost of the plant was $7,039.00. The depreciation per year at 5% would amount to $351.95, or for 200 working-days $1.75 per day. Cost of operation per day is as follows : Motorman $1.75 Helper 1.25 Electrician .78 Repairs .65 Depi-eciation 1.75 Oil 24 Total $6.42 The coal hauled per day is 600 tons, at a cost per ton as shown above, of 1.07 cents. To haul this coal by mule-power would require Twelve mules at 50 cents each $6.00 Six boys at $1.35 each 8.10 Total $14.10 This shows a cost for haulage by mule-power of 2.35 cents, and a saving by electric haulage of 1.28 cents per ton. On the 600 tons hauled per day the saving is $7.68, and for a year of 200 working-days it amounts to $1,536.00. ■» The Hillside Coal and Iron Company. The Hillside Coal and Iron Company was one of the first companies to install electric haulage. At Forest City, Pa., they have two openings operated by electric haulage from one power-house. The power-house con- tains about 150 Kw. direct connected generators and one 62 Kw. belt driven machine. At what is known as the "No. 2 Shaft" they have one twenty- ton, eight-wheel locomotive, one twelve-ton single motor locomotive, and one six-ton locomotive. At the Forest City Slope there is a twelve-ton single motor locomotive. In addition to this, they have two electric pumps. The plant here has been in operation since 1891, although the power-house has been increased and rebuilt since the original plant was installed. Mr. W. A. May, Superintendent, very kindly furnished the following figures, which are on exactly the same basis as the figures in Mr. Piatt's paper. Cost of operation per day is as follows : No. 2 Shaft. Forest City Slope. Engineer of power-house . . . $1.20 $0.60 Motormen 4.23 2.11 Helpers (Brakemen) 3.20 1.60 Electrician 1.67 .83 Repairs to motors 5.95 4.09 Depreciation, 5% ..... . 5.20 2.60 Oil and waste .22 .14 Total $21.67 $11.97 Coal hauled per day — tons . . 989 541 Cost per ton $.0219 $.0221 This plant has never been operated with mules, but the mine foreman has gone over the matter very carefully, and has made up the following estimate of the number of mules it would require to do the work. He finds that it would take fifty-three mules in the shaft and twenty-four in the slope. Again using Mr. Piatt's figures, we get the folloAving cost per day for haul- age by mule-power in No. 2 Shaft. OPERATION OF ELECTRIC MINING PLANTS. 699 Fifty-three mules at 50 cents each $26.50 Twenty-four drivers at $1.48 each 35.52 Twenty-four team leaders at $1.04 each .... 24.96 Total $86.98 This shows a cost for haulage by mule-power of 8.79 cents per ton and a saving by electric haulage of 6.60 cents per ton. On the 989 tons hauled per day the saving is $65.27, and for a year of 200 working-days it amounts to $13,054.00. In the Forest City Slope the cost per day for haulage by mule-power is as follows : Twenty-four mules at 50 cents each $12.00 Ten drivers at $1.48 each 14.80 Ten team leaders at $1.04 each 10.40 Two runners at $1.59 each 3.18 Total • • • $40-38 This shows a cost for haulage by mule-power of 7.47 cents per ton, and a saving by electric haulage of 5.26 cents per ton. On the 541 tons hauled per day the saving is $28.46, and for a year of 200 working-days it amounts to $5,692.00. Mr. May remarks that in their particular case this estimate is not entirely correct, as the expenses of the engineer, motormen, helpers, etc., are steady expenses, their time on idle days being occupied with more or less running around and making repairs about the mines. They have therefore made an additional set of figures, using the actual number of days that the mines were running, with the actual cost. The No. 2 Shaft ran 141J days, and the Forest City Slope 138| days. Under these circumstances the cost of oper- ation per day is as follows : No. 2 Shaft. Forest City Slope. Engineer of power-house . . . $2.84 $1.45 Motormen Helpers (Brakemen) .... Electrician Repairs to motors Repairs to line Repairs to generators . . . Fireman . Depreciation, 5% Oil and waste for motors . . Oil and waste for generators Interest on plant at 3% . . Total Coal hauled per day— tons . Cost per ton Then, again, taking their own figures on the cost of keeping Avhat mules they have, they obtained the following cost per working-day for haulage in No. 2 Shaft : The depreciation on 53 mules, at $1.67 each per month, is $88.51, and for 12 working-days per month the depreciation per day is $7.38. Depreciation on 53 mules $7.38 Feed for 53 mules (at 25 cents each per day per month) 33.12 Shoeing and harness 1.59 Care of mules 3.97 Forty-eight drivers and team-leaders 60.48 Total ,,,",. , , o $106.54 9.31 4.76 3.61 2.63 3.68 1.87 8.42 5.89 .46 .03 .61 .30 2.50 1.26 8.17 4.16 .35 .21 .74 .37 4.41 2.25 $45.10 $25.18 989 541 $.0456 $.0465 700 ELECTRICITY IN MINES. This shows a cost for haulage by mule-power of 10.77 cents per ton, and a saving by electric haulage of 6.21 cents per ton. On the 989 tons hauled per day the saving is $61.42, and for a year of 141^ days it amounts to $8,615.75. In the Forest City Slope the depreciation ligured as above on 24 mules is $3.34, and the detailed cost of haulage by mule-power is as follows : Depreciation on 24 mules $3.34 Feed for 24 mules (at 25 cents each per day per month) 15.00 Shoeing and harness .72 Care of mules 1.80 Twenty-two drivers, leaders, and runners . . . 28.38 Total $49.24 This shows a cost for haulage of mule-power of 9.10 cents per ton, and a saving by electric haulage of 4.45 cents per ton. On the 541 tons hauled per day the saving is $24.07, and for a year of 138| days it amounts to $3,339.71. To the cost of the msle-power might yet be added interest at 3% on the value of the mules and harness, but as it has not heretofore been included, it has been left out here. From the foregoing it will be seen that in either case there is a consider- able saving in favor of electric haulage, and that this saving will increase as the number of idle days increases and with the increase in tonnage in the colliery. LIGHTNING CONDUCTORS. Views concerning the proper function and value of lightning rods, con- ductors, arresters and all protective devices have undergone considerable modification during the past ten years. There may he said to be four periods in the history of the development of the lightning protector. The first embraces the discovery of the identity of lightning with the disruptive discharge of electrical machines and Franklin's clear conception of the dual function of the rod as a conductor and the point as a discharger. The second begins with the experimental researches of Faraday and the minia- ture house some twelve feet high, which he built and lived in while testing the effects of external discharges. Maxwell's suggestion to the British Association, in 187(5, embodies a plan based upon Faraday's experiments, for protecting a building from the effects of lightning by surrounding it with a cage of rods or stout wires. The third period begins with the experiments of Hertz upon the propagation of electro-magnetic waves, and finds its most brilliant expositor in Dr. Oliver J. Lodge, of University College, Liverpool, whose experiments made plain the important part which the momentum of an electric current plays, especially in discharges like those of the lightning flash, and all discharges that are of very high potential and oscilla- tory in character. The fourth period is that of the present time, when individual Hashes are studied ; and protection entirely adequate for the particular exposure is devised, based upon some knowledge of the electrical energy of the flash, and the impedance offered by appropriate choke coils or other devices. For example, under actual working conditions, with ordinary commercial voltages, effective protection to electrical machinery connected to external conductors may be had with a few choke coils in series with intervening arresters. A good idea of the growth of our knowledge of the nature and behavior of the lightning flash may be obtained from the following publications : Franklin's letters. Experimental Researches. . . . Faraday. Report of the Lightning Rod Conference, 1882. Lodge's "Lightning Conductors and Lightning Guards," 1892. "Lightning and the Electricity of the Air." . . . McAdie and Henry, FIG. 1 EFFECT OF THE ACTION OF LIGHTNING UPON A ROD. That a lightning rod is called upon to carry safely to earth the discharge from a cloud was made plain by Franklin, and the effect of the passage of the current very prettily shown in the melting of the rod and the point (aigrette). Here indeed was a clew to the measurement of the energy of the lightning flash. W. Kohlrausch in 1890 estimated that a normal lightning discharge would melt a copper conductor 5 mm square, with a mean resistance of 0.01 ohm in from .03 to .001 second. Koppe in 1895 from measurements of two nails 4 mm in diameter fused by lightning, determined the current to be about 200 amperes and the voltage about 20,000 volts. The energy of the flash, if the time be considered as 0.1 second, would be about 70,000 horse power, or about 52,240 kilowatts. Statistics show plainly that buildings with conductors when struck by lightning suffer comparatively little damage compared with those not pro- vided with conductors. The same rod, however, cannot be expected to serve equally well for every flash of lightning. There is> great need of a classification of discharges based less upon the appearance of the flash than upon, its electrical energy. Dr. Oliver J. Lodge has made a beginning with 701 702 LIGHTNING CONDUCTORS. liis study of steady strain and impulsive rush discharges. " The energy of an ordinary flash," says Lodge, " can he accounted for by the discharge of a very small portion of a charged cloud, for an area of ten yards square at the height of a mile would give a discharge of over 2,000 foot-tons energy." We must get clearly in our minds then the idea that the cloud, the air, and the earth constitute together a large air condenser, and that when the strain in the dielectric exceeds a tension of \ gramme weight per square centimeter, there will be a discharge probably of an oscillatory character. And as the electric strain varies, the character of the discharge Avill vary. Remember too that the air is constantly varying in density, humidity and purity. We should therefore expect to find, and in fact do, every type of discharge from the feeble brush to the sudden and terrific break. Recent experiments indicate that after the breaking-down of the air and the pas- sage of the first spark or flash, subsequent discharges are more easily ac- complished ; and this is why a very brilliant flash of lightning is often followed almost immediately by a number of similar flashes of diminishing brightness. The heated or incandescent air we call lightning, and these lines of fracture of the dielectric can be photographed ; but the electrical waves or oscillations in the ether are extremely rapid, and are beyond the limits of the most rapid shutter and most rapid plate. Dr. Lodge has calculated the rapidity of these oscillations to be several hundred thousand per second. Lodge has also demonstrated experimentally that the secondary or induced electrical surgings in any metallic train cannot be disregarded ; and, as in the case of the Hotel de Ville at Brussels Avhich was most elaborately pi'otected by a network, these surgings may spark at nodal points, and ignite inflammable material close by. While therefore it cannot be said that any known system of rods, wires, or points affords complete and absolute protection, it can be said with con- fidence that we now understand why " spitting-off " and " side " discharges occur ; and furthermore, to quote the words of Lord Kelvin, that " there is a very comfortable degree of security . . . when lightning conductors are made according to the present and orthodox rules," Selection and. Installation of Rods. — The old belief that a copper rod an inch in diameter could carry safely any flash of lightning is perhaps true, but we now know that the core of such a rod would have little to do in carrying such a current as a lightning flash, or, for that matter, any high frequency currents. Therefore, since it is a matter of surface area rather than of cubic contents, and a problem of inductance rather than of simple conductivity, tape or cable made of twisted small wires can be used to advantage and at a diminished expense. All barns and exposed buildinr/s should hare lif/htninr/ rods with the neces- sary points and earth connections. Ordinary dwelling-houses in city blocks well built up have less need for lightning conductors. Scattered or isolated houses in the country, and especially if on hillsides, should have rods. All protective trains, including terminals, rods, and earth connections, should be tested occasionally by an experienced electrician, and the total resist- ance of every hundred feet of conductor should not greatly exceed one ohm. Use a good iron or copper conductor. If copper, the conductor should weigh about six ounces per linear foot ; if iron, the weight should be about two pounds per foot. A sheet of copper, a sheet of iron, a tin roof, if with- out breaks, and fully connected by well soldered joints, can be utilized to advantage. b FIG. 2 AND 3 APPROVED CONDUCTORS AND FASTENINGS. PERSONAL SAFETY DURING THUNDER-STORMS. 703 In a recently published* set of Rules for the Protection of Buildings from Lightning, issued by the Electro-Technical Society of Berlin, Dr. Slaby gives the results of the work of various committees for the past sixteen years studying this question. The lightning conductor is divided into three parts, — the terminal points or collectors, the rod or conductor proper attached to the building, and the earth plates or ground. All projecting metallic sur- faces should be connected with the conductors, whicb, if made of iron, should have a cross section of not less than 50 mm square (1.9 sq. inches) ; copper, about half of these dimensions, zinc about one and a half, and lead about three times these dimensions. All fastenings must be secure and lasting. The best ground which can be had is none too good for the light- ning conductor. For many flashes an ordinary ground will suffice, but there Avill come occasional flashes when even the small resistance of ^ ohm may count. Bury the earth plates in damp earth or running water. The plates should be of metal at least three feet square. " If the conductor at any part of the course goes near water or gas mains, it is best to connect it to them. Wherever one metal ramification ap- proaches another, connect them metallically. The neighborhood of small bore fusible gas pipes, and indoor gas pipes in general, should be avoided." — Db. Lodge. FIG, 4 CONDUCTORS AND FASTENINGS.. (FROM ANDERSON, AND LIGHTNING ROD CONFERENCE.) The top of the rod and all projecting terminal points should be plated, or otherwise protected from corrosion and rust. Independent grounds are preferable to water and gas mains. Clusters of points or groups of two or three along the ridge rod are good. Chain or linked conductors should not be used. It is not true that the area protected by any one rod has a radius equal to twice the height of the conductor. Buildings are sometimes, for reasons which we understand, damaged within this area. All connections should be of clean well-scraped surfaces properly soldered. A few wrappings of wire around a dirty water or gas pipe does not make a good ground. It is not necessary to insulate the conductor from the building. BIIIECTIOAS FOR PEKSO^AL iAfETI D1IRIHG IHVIKDEn STORMS Do not stand under trees or near wire fences ; neither in the doorways of barns, close to cattle, near chimneys or fireplaces. Lightning does not, as a rule, kill. If you are near a person who has been struck do not give him up. Ztschrift, 1901, May 29, ei 704 LIGHTNING CONDUCTORS. as beyond recovery, even if seemingly dead. Stimulate respiration and circulation as best you can. Keep the body warm ; rub the limbs energet- ically, give water, wine, or warm coffee. Send for a physician. TESTS Of L1&HTX1AG RODS. To make the test, first determine the resistance of the lead wire lx and call it lv Then connect E{ and E2 as shown in the diagram, call the result lix ; then connect Ex and EM call the result Ji2 ; connect E2 and E3 and call the result JRS. TESTS OF LIGHTNING RODS. THIS LEAD MUST BE SOLDFRED TO THE PIPE OR OTHER EARTH SO AS E NO RESISTANCE AT THIS JOINT. FIG, 6 DIAGRAM OF CONNECTIONS FOR TEST OF LIGHTNING RODS. Now, Rx = lx-\-Ex-\-E2 and Es=E1—l1 — E1 Jti = l1-\-Ex-\-E« and j0, = i?2 — I, — JE. XSi = E2-t-Es ' solving, we have All lightning rods should be tested for continuity and for resistance of ground plate each year, and the total resistance of the whole conductor and ground plate should never exceed an ohm. DETERMINATION OF WAVE FORM OF CUR- RENT AND ELECTRO MOTIVE FORCE. ^^ There are numerous methods of determining wave form, those used in laboratory experiments commonly making use of the ballistic galvanometer. Of the simple methods used in shop practice, R. D. Mershon, of the West- inghouse Electric and Manufacturing Co., has applied the telephone to an old ballistic method in such a manner as to make it quite accurate and readily applied. Jtlershon's Method.— The following cut shows the connections. A telephone receiver, shunted with a condenser, is connected in the line from the source of current, the wave form of which it is wished to determine. A contact-maker is placed in the other leg, and an external source of steady current, as from a storage battery, is opposed to the alternating current, as shown. The pressure of the external current is then varied until there is no sound in the telephone, when the pressures are equal and can be read a.c. terminals from the voltmeter. The contact- maker being revolved by successive steps, points may be determined for an entire cycle. Duncan's Method. — Where it is desirable to make simultaneous de- terminations it will ordinarily require several contact-makers, as Avell as full sets of instruments. Dr. Louis Dun- can has devised a method by which one contact-maker in connection with a dynamometer for each curve will ena- ble all readings to be taken at once. The following cut shows the connec- tions. The fixed coils of all the dy- namometers are connected to their respective circuits, and the fine wire Fig. movable coils of about 1,000 ohms each, are connected in series with a contact- maker and small storage battery. The contact-maker is made to revolve in synchronism with the alternating current source. Now, if alternating cur- rents from the different sources are passed through the fixed coils, and at intervals of the same frequency current from the battery is passed through the movable coils, the deflection or impulse will be in proportion to the instantaneous value of the currents flowing in the fixed coils, and the deflections of the movable coils will take permanent position indicating that value, if the contact-maker and sources of alternating current are revolved in unison. The dynamometers are calibrated first by passing continuous currents of known value through the fixed coils, while the regular in- terrupted current from the battery is being massed through the movable coils. Ryan's Method. — Prof. Harris J. Ryan, of Cornell University, designed a special elec- trometer for use in connection with a very fine series of transformer tests. This instrument Avill be found described and illustrated in the chapter on description of instruments. The method of using it is shown in the cut below, in which the contact- maker shown is made to revolve in synchronism with the source of alter- 705 MershOn's method of de- termining Wave Form. Fig. 2. Duncan's method of determining curves of several circuits at the same time. 706 WAVE FORM. TRANSFORMER nating current. The terminals, d dt, of the indicating instruments can be connected to anyone of the three sets of terminals, a ax b 6X c clm The terminals, a ar, are for reading the instantaneous; value of the pri- mary impressed E.M.F. ; b bu the same value of the current flowing through the small non-inductive re- sistance, R ; and c ct the same value of the secondary impressed E.M.F. ; the secondary current being read from the ammeter shown. Of course if the contact-maker be cut out, then all the above values will be Vmean2. UK aafifyee, ™ WAVE METER. The instrument illustrated and de- scribed in the following pages has been in use in the laboratory of the General Electric Company at Schenectady, since early in 1896, and is, I think, the simplest form of apparatus yet sug- gested for determining wave forms in alternating currents. The General Electric Company very kindly furnished the following de- scription, and the diagrams and illustrations accompanying it. Fig. 3. Prof. Ryan's method of ob- taining curves of wave form for studying transformers. Fig. 4. This device consists of a synchronous motor intended to run in synchro- nism with the machine under test. On the shaft of the motor is placed a contact device similar to the contact device usually placed directly on the shaft of the generator. By the use of a synchronous motor, the device be- comes much more flexible, and enables the Avave to be taken on any part of any alternating current circuit by merely attaching a pair of lead wires, thus doing away with all mechanical attachments to the generator. Since the advent of alternators with a considerable number of poles, the old method of mechanical connection has been found to be unsuitable on account of the great degree of accuracy required in dividing a cycle into the requisite number of degrees, owing to the fact that a complete cycle of 360° forms such a small part of the arc of the armature. The operation of the machine in detail is as follows : — The field requires about 1.35 amperes I). C, and the armature about 4 amperes for starting. The machine should then be started by means of the crank (marked A in Fig. 4) until it has been brought up to the frequency of the A. C. circuit, which a\ ill be indicated by tachometer (mai-ked H). At 60 cycles the speed is 900 R. P. M. As soon as it is in synchronism (which can be easily told by the running of the machine) the lever (marked B) on the crank standard should be pressed, which releases the gear mechanism and allows the motor to run free. After the machine is running, current in the armature should be reduced to 3 amperes. WAVE METER. The following precautions are necessary in order to procure satisfactory working of the apparatus : — 1. The resistance in all the circuits must be unvarying ; the contact, therefore, must be perfect. 2. The E.M.F. of the A. C. and D. C. circuits must be steady and unchan- ging. Complying with No. 1 and No. 2 secures steady currents in all the circuits. 3 Above all, the speed of the source must be kept constant ; and if this is not possible, readings must be taken only at a certain speed, that speed being preferred to which the generator most frequently returns. 4. Avoid any leads other than shown on the diagram coming in contact with the terminals of the D. C. voltmeter. It will be noticed that a con- nection between the large and small segments will cause alternating cur- rent to flow through the direct current voltmeter. 5. The tension on the contact spring " F" must be stiff enough to insure a good contact. If the brush does not make an even contact on the contact- disk, it can be remedied by placing a piece of emery cloth on the contact- disk and revolving the brush over the rough side of the emery cloth by hand. G. The carbon brushes must make as perfect contact on collector rings as possible. 7. In taking a wave, it is recommended that the voltmeter reading should vary from a minimum of zero to a maximum of nearly a full scale deflection. Fig. 5. This absolute zero can be obtained by loosening the set screw (marked C) on the end of index lever " D." The contact disk, " G," can then be rotated on the shaft until the voltmeter reading is at zero, with index pointer set on zero degrees. In case the maximum deflection is too low, it can be in- creased by either inserting more capacity in the circuit or by using a higher voltage on the condenser circuit ; this would be accomplished by using a small step-up transformer or compensator at the point marked T in Curve Sheet IN o. 8. The transformer voltage should not exceed 150 volts at this point. 8. In case the voltage is too low to give a readable deflection on the volt- meter, a DArsonval galvanometer can be used in place of the voltmeter. 9. The oil-cups (marked E) should be kept full of oil, as a thorough lubri- cation is found necessary to procure perfect results. 10. If the machine sparks at contact disk, that is, if spark causes arcing from one segment to the following one, it will be necessary to rub the sur- face of the disk with fine sandpaper. The external wiring connections of the machine are shown on Curve Sheet A attached. The connections of the contact device are also shown. This consists of a contact-disk with 4 large and 4 small segments. The 4 large segments are connected to the inner copper ring on the side of the contact- disk. By means of a spring contact and leads the latter is connected to the terminal V. Similarly the smaller segments are connected through the outer ring and spring contact and leads to the terminal T. 708 WAVE FORM. The revolving brush is in contact by means of brush and contact ring as seen on the end of the shaft (marked 1) to the frame, and from the latter by means of wire under the base to the terminal C. The principle on which the method is based is the following : When the revolving brush, F, leaves the small segment of tbe contact disk, U, and breaks the contact between the condenser and the E.M.F. to be measured, it leaves the condenser charged with the potential difference which oc- curred at that instant. As soon as the revolving brush touches the large segment, the condenser discharges into the voltmeter until the brush leaves it. As the speed is constant, the time of discharge is constant, and as the discharging circuit is unaltered during the test, the instantaneous E.M.F's cause proportional deflections ; the latter follow so quickly as to give steady deflections. Reading', Plotting, and Calculating-. — The movable index pointer is turned till the spring-actuated pin drops into the small hole above zero on the fixed scale, and the deflection of the voltmeter noted on a sheet of paper having two parallel columns counting the degrees from zero to 360, as indicated below : — DEFLECTION. DEFLECTION. 180 185 190 195 etc. 180 355 360 After taking the reading at zero, the pointer is moved to 5, then to 10, and so on. If after finishing this series of readings a marked difference is noted between corresponding deflections in the left and right hand columns, such points must be taken over again. The average of the two corresponding deflections is taken, and the results are then multiplied by such a constant as to make the maximum = 10. These values are plotted as Ordinates, and the corresponding degrees are abscissae. See sample test and Curve Sheet B. To find the average E.M.F. , divide the area in terms of squares of the paper used, by 10 times the actual length of one cycle in terms of one side of the same squares, as the maximum is plotted to a scale of 10 instead of one. On Curve Sheet B the length of the half -cycle = 9 units, and therefore the area must be divided by 90. The effective E.M.F., or Vmean square J\ the square root of the mean squares of the same instantaneous values used before. The simplest method of obtaining this is the fol- lowing : Plot the same deflections on polar co- ordinate paper similar to that used in Curve Sheet C, and find the area of the resulting curve. The effective E.M.F. is then equal to the radius of a semi-circle whose area expressed in terms of squares of the rectilinear co-ordi- nate paper, is equal to the area enclosed by the wave plotted on . the polar co-ordinate paper after being reduced to the same dimensions by multiplying by the ratio of ^2^2 EXPERIMENTAL ALTERNATOR FIG. 6. Curve Sheet B. '(f)* To find the area a planimeter is used, or the curve is traced or copied by means of carbon paper on paper of uniform thickness, which is then weighed on a chemical balance, or in case neither of the above methods is avail- SAMPLE TEST. 709 able, the area can be found by actually counting the number of squares it contains. The form factor is the ratio of the effective to the mean E.M.F. The form factor of a sine wave is 1.11. The amplitude factor is the ratio of the max- imum to the effective E.M.F., which, as the maximum is one, is equal to the reciprocal of the effective E.M.F. The amplitude factor of a sine wave is 1.414. These values are to be used in making cal- culations for alternat- ing currents whose wave shapes have been determined by means of the wave meter in- stead of employing the usual values based on the sine curve. The accom- panying record sheets give the results obtained with an actual E.M.F. wave taken with the machine. In the sample test, columns 2 and 4 give the read- ings obtained for the different angular deflections. Column 5 is the average of the readings obtained. These values are then multiplied by a constant, which in this case is .1127, to give a maximum of 10. The resultant values plotted in rectilinear and polar co-ordinates are shown on curve sheets B and C. ^ — /—+— i— i_^ 1 III — /\ W X r~-4-^.fe — 1^\ a A/ /\- Fig. Curve Sheet C. SAIKPIE TEST. (Nov. 21, 1897.) JS.M.IT. Wave of Experimental Alternator. Ko. 1. No. 2. No. 3. No. 4. No. 5. No. 6. Degrees. 0 —4.5 180 —4.5 —4.5 — .507 175 5 +2.5 185 +2.5 +2.5 + .28 0 10 190 7. +7- .79 5 15 11. 195 11. +11. 1.24 10 20 21. 200 19.5 +19.75 2.23 15 25 29.5 205 29.5 29.5 3.32 20 30 30. 210 29.5 29.75 3.35 25 35 29.5 215 29.5 29.5 3.32 30 40 36. 220 36. 36. 4.06 35 45 51. 225 50. 50.5 5.695 40 50 71. 230 72. 71.5 8.05 45 55 72.5 235 71.5 72. 8.12 50 60 66. 240 66. 66. 7.44 55 65 70. 245 71. 70.5 7.95 60 70 85. 250 85. 85. 9.58 65 75 89. 255 88.5 88.75 10. 70 80 75.5 260 73.5 74.5 8.4 75 85 68.5 265 67.5 68. 7.66 80 90 61.5 270 60.5 61. 6.87 85 95 59.5 275 60. 59.75 6.73 90 100 72. 280 72.5 72.25 8.15 95 105 81. 285 81. 81. 9.13 100 110 87. 290 87.5 87.25 9.82 105 710 WAVE FORM. SAMPLE TEST — (Continued). Mo. 1. No. 2. No. 3. No. 4. No. 5. No. 6. Degrees. J 15 84. 295 83. 83.5 9.40 110 120 72.5 300 73. 72.75 8.2 115 125 67.5 305 68. 67.75 7.63 120 130 77. 310 77.5 78.25 8.81 125 135 82.5 315 82.5 82.5 9.3 130 140 60. 320 59.5 59.25 6.68 135 145 41. 325 41.5 41.25 4.65 140 150 32. 330 32.5 32.25 3.64 145 155 30.5 335 30.5 30.5 3.44 150 160 37.5 340 37. 37.25 4.2 155 165 30. 345 30.5 30.25 3.31 160 170 16. 350 15.5 15.25 1.775 165 175 8.5 355 9.0 8.75 .98 170 The different constants of this wave are given below in "Method of De- termining Constants of E.M.F. Curve." This also gives the constants for a sine wave for comparison. SPECEAE DATA ©]¥ THE MOTOR IILISTRATED. Resistance of field = 10.87 ohm. Resistance armature and brushes = 2.055 ohm. Armature alone = .560 ohm. Armature winding — 14 turns of No. 28 D. C. C. copper wire doubled in each slot. Field frame consists of Txff H.P. U. I. Fan Motor — 125 cycles, 104 volts. METHOD OE »ETEI»]fII]¥I]¥« COISTAHiTS OE E.M.E. CURVE. Area Rect. Co-ord. Curve " B " = 51.32. Mean E.M.F. = ^^5 = -571. Polar Area = 4,062, which must be multiplied by I ' ) to be com- parable to the area in rectilinear co-ordinates. 11.33 is the maximum ordi- nate of the rectilinear co-ordinate in centimeters, and 8.95 is the maximum ordinate of the polar co-ordinate curve ; therefore the corrected polar area = 40.62 X 1.6 = 64.992. Now ^nr* = 64.992, therefore r = .643, which is the effective E.M.F. The form factor being therefore The amplitude factor : effective mean maximum .643 _ ^571" 1.127. For comparison the constants of a sine wave are also given in the recapit- ulation below. Mean E.M.F. Rect. Co-ord. Curve B 51.32 | _— Polar " " C 40.62 { ■i3il Sine Wave 637 Effect. E.M.F. .643 .707 Form factor. 1.127 1.110 Amp. factor. 1.554 1.414 CERTAIN USES OP ELECTRICITY IN THE UNITED STATES ARMY. Electricity enters into nearly every branch of the military art, being used for the operation of searchlights, turret-turning, manipulation of coast-de- fense guns, ammunition hoists, range and position tinders; for firing sub- marine mines ; field and fortress telephones and telegraphs ; firing devices for guns, ground mines ; in tide gauges ; submarine boats and dirigible tor- pedoes ; while electrically operated chronographs are employed in the solu- tion of ballistic problems. SEARCHLIGHTS. Searchlights are used both as offensive and defensive auxiliaries ; defen- sive when used by shore fortifications to light channels or by a vessel to discover the approach of torpedo boats ; offensive when used as " blinding- lights " to smother the light of an approa ching vessel and confuse her pilot. The accompanying illustrations show the searchlight manufactured by Schuckert & Go. of Nurnberg, Germany. The lamp is placed on top of the two lowest longitudinal rods of the cas- ing, and is held in place by four lugs, two on each side. The carbon holders reach upward through a slit in the casing, and there is a small wheel in rear for moving the light parallel to the axis of the reflector, for the purpose of focusing it. The trunnions of the casing are fastened to two longitudinal rods on each side, parallel to the axis of the cylinder, and can be moved for- ward or back so that the casing and what is carried with it will have no pre- ponderance. The trunnions are supported in trunnion beds in the ends of supports which project upwards from the racer. The elevating arc is attached to another longitudinal rod beneath the cylindrical casing and is capable of adjustment on this rod. Engaging in this arc is a small gear attached to a horizontal shaft passing through the right trunnion support and carrying a small hand wheel. This small hand wheel is for the purpose of elevating or depressing the light rapidly. The light may be elevated or depressed slowly by means of a small hand wheel attached to another horizontal shaft in front of the one just described. This shaft near its center carries a worm, engaging in a worm wheel on a vertical shaft, to which is also attached a bevel gear. This gear engages in another, which is attached to the quick- motion shaft, but is free to turn about it until it is connected with the elevating gear wheel by means of a friction clamp. The relation between the worm and worm wheel is such that a slow motion is obtained. The racer rests upon live rollers and is joined by a pintle to the base ring. Attached to the base ring is a toothed circular rack, into which on the outside a gear wheel attached to a vertical shaft engages. This vertical shaft projects upward through the racer and carries a worm wheel, which engages in a worm carried on a horizontal shaft having a hand wheel. The worm wheel is entirely independent of its vertical shaft, except when con- nected with it by means of a friction clamp. When so connected, by turn- ing the hand wheel the light is traversed by a slow motion. To traverse the light rapidly, the friction clamp is released and the light turned by hand, taking hold of the trunnion supports. One of the ends of the slow motion elevating and traversing shafts is connected with a small electric- motor, which is encased in a box on top of the racer. By means of these motors the motion of the searchlight can be controlled from a distant point. A switch is provided with contacts so arranged that the current can be passed into the armatures of the motors in either direction, so as to obtain any movement the operator may desire. The current needed for the move- ment is obtained from the lines supplying the current used in the light itself. The current is brought to the motors by means of contact points, bearing on circular contact pieces attached to the racer. The reflector is a parabolic mirror embedded in asbestos in a cast-iron frame, and is held in place by a number of brass springs. The frame of the reflector is fastened to the overhanging rear ring of the casing with studs and nuts, the overhanging part of the ring protecting the reflector from 711 712 CERTAIN USES OF ELECTRICITY IN U.S. ARMS". Fig. 1. Schuckert Searchlight as used in U. S. army, moisture. In order to enable the operator to observe the position of the carbons and the form of the crater while the apparatus is in use small optical projectors are arranged at the side and on top of the casing, which enables images of the arc as seen from above and from the side to be observed. When the light is properly focused the positive carbon reaches a line on the glass on top of the casing. There are two screws on the positive carbon holder which enable the end of this carbon to be moved vertically or horizontally to bring it to a proper adjustment. In consequence of the ascending heat the carbons have a tendency to be consumed on top ; and to avoid this there is placed just back of the arc and concentric with the positive carbon a centering segment of iron, attached to the casing, which, becoming magnetic, so attracts the current as to equalize Searchlights. 713 the upward burning of the carbons. In taking the light out of the casing this centering segment must be unfastened, and swung to the side on its hinge. SAFETY FUSE Fig. 2. Diagram showing Searchlight Connections. An example of the method of calculating the intensity of the light sent out by the mirror follows : — Diameter of parabolic mirror, 59.05 inches. Diameter of positive carbon, 1.5 incbes. Diameter of negative carbon, 1 inch. Power consumed, 150 amperes x 59 volts. Maximum intensity of rays impinging upon tbe mirror, 57,000 candle- power. 714 CERTAIN USES OF ELECTRICITY IN U.S. ARMY. •iu in -ui 0001 W uopimiumni J° PPki L11C* Tt< -* L 333 Ci CO -* t- 00 LO •aoisjadsiQ; X — co o :- ~l X O i- lO IQ •<* •* ■* lO i-< CO -* OOO OOO OOO «N« NIMO) Ci CO ent out by mirror, 45,600x4,253 = 194,000,000 can- dle-power. The focal distance of the mirror is 25.5 * inches. The dispersion angle of the concen- trated beam is 2° 2'. The diameter of the illuminated area at a distance of 1,111 yards is 84 yards. The resistance Rm on the switchboard at the light is in series with the main current for the purpose of regulating the amperage at the lamp. The volt- meter at the lamp should indicate about 60 volts. The connection of the dis- tance governor with the two motors for elevating and traversing is also shown. The largest searchlight so far built is the one that was on exhibition at the Paris Exposition of 1900 in the section " Navigation de Commerce et Armees de Terre et de Mer," which is 6 feet 6 inches in diameter, and gives a beam of 316.000,000 candles. The table on preceding page gives data in regard to searchlights of various sizes. CHROIOGRlPH§. In the experimental work of testing guns, etc., it becomes necessary to ascer- tain the velocity of projectiles both while passing through the bore of the gun and during flight. Chronographs of various sorts are used for this purpose. In order to ascertain the velocity of a projectile during flight, two screens or targets are set up in the course of the projectile, generally 100 feet apart. These screens ordinarily consist of a frame of wood carrying a number of small parallel copper wires. The break- ing of the wires in the successive frames by the projectile causes the interruption of the current through the instrument, and thus registers the time of flight between the screens. Probably the best-known instrument of this class is the one invented by Cap- tain Le Boulenge of the Belgian artil- lery, which was afterwards modified by Captain Breger. Bouleng-e Chronograph. This instrument depends for its accu- racy upon the law of falling bodies or the acceleration due to gravity, namely 32 feet per second. It consists of a vertical column (Fig. 3 ) to which are affixed two electro- magnets ; the right-hand one, A , is actuated by the current of the first frame 716 CERTAIN USES OF ELECTRICITY IN U.S. ARMY. and supports an armature called the chronometer ; the left-hand magnet, jB, is actuated by the current of the second frame, and supports an ai ma- ture, D, called the registrar. The chronometer, C, is a long, cylindrical brass tube terminating at its upper extremity in a piece of soft iron, and bearing at its lower extremity a steel bob. It is surrounded by a zinc or copper cylinder called the recorder. The rupture of the first target causes the demagnetization of the magnet A, releasing the rod C. The registrar is of the same weight as the chronome- ter, and is a tube with soft iron and* bob. The cores of the electro-magnets and the soft iron of the armatures terminate in cones slightly rounded at their vertices in order that the armatures when suspended can take a verti- cal position. When the registrar is set free by the rupture of the second target it strikes a horizontal plate («), which turns upon its axis (c) and releases the spring (d). The spring is furnished with a square knife (, a steel spring, is fastened at one end to a, the free end being limited in its motion by, two small pins set into a. The electro-magnet B, Avhich holds the balance-wheel at the starting- CHRONOGRAPHS. 719 point, is operated by the current passing through the first screen, and is mounted on the face piate so that the core is radial with reference to the balance-wheel. Tbe core of the magnet projects beyond the coil and acts upon the small armature c, mounted on the rim of the balance-wheel. The electro-magnet C, with its mechanism operated by the current pass- ing through the second screen, stops the balance-wheel the instant the current is broken. This magnet is somewhat larger than the other, and is Fig. 7. Interior Schimdt Chronograph. placed tangentially with reference to the balance-wheel. It acts upon the two armatures d, d', placed opposite the extremities of the core. These armatures are fastened to the ends of the two levers K, K', which are mounted on the axis e, e', parallel to the axis of the balance-wheel and Fig. 8. Construction of Needle. similarly supported. The other ends of the levers are joined by the coiled spring L with its adjusting-screw. Set in the levers near this end are four pins, /,/,/',/', that ordinarily, due to the tension of the spring, bear against 720 CERTAIN USES OF ELECTRICITY IN U.S. ARMY. the rim of the balance-wheel, holding it fast. "When the current passes through this magnet, the armatures on the levers are attracted by the core, the spring is elongated, and the pressure of the pins upon the balance- wheel is released. When the current is broken the armatures are released, and the tension of the spring closes the pins upon the wheel. To insure effective action the pins are accurately set and the rim of the wheel is The face of the chronograph is a graduated dial concentric with the balance-wheel axis. When the wheel is held at its starting-point the needle points at the zero of the graduation. The scale in black indicates the time in thousandths and two-ten-thousandths of a second. Another scale, in red, gives the velocity directly in meters per second when the screens are placed 50 meters apart. . . . . The dial is covered with glass inclosed m the circular metal frame E. A pin, g, fixed in the glass, is used to set the. needle at zero by turning the frame, to which is also fastened the lens h, to facilitate reading. This lens is provided with two pointers so placed that the reading is always taken in the vertical plane. The button F is for the purpose of reestablishing the current through the magnet C after it has once been broken. Pressing the button closes the circuit ; the magnet then attracts the armatures d, d', fixed to the ends of the levers X, K'. This motion of "he levers brings the small spring I, mounted on K', in contact with the projection k, thus forming a circuit through which the current continues to flow after the pressure on F has been released. This contact is broken by the motion of the lever when the current is interrupted by the shot. This arrangement prevents the current from passing through the magnet and releasing the balance-wheel before the circuit is closed by pressing the button F, even though the broken screen is repaired, and gives the operator time to take the reading and prepare for the next shot. This is especially important when targets that close the circuit automatically are used. The rheostats for regulating the currents are placed above the dial, their resistance coils being inside the case. One binding-post of each rheostat is provided with a circuit closer for passing the currents through the resis- tance coils or directly into the rheostats. The Squire-Creliore Photo-Chronograph. This instrument was designed to overcome the minute errors inherent in other forms of chronographs, such as the inertia of the amature, the time required to magnetize iron, or in instruments employing a sparking de- vice, the fact that successive sparks do not proceed from the same point by identically the same path. The agents employed in this instrument are light and electricity. Briefly stated, a ray of light from an electric arc is reflected upon a revolving photographic plate. The interposition of a tuning-fork gives on the plate a curve which is used as a scale of time. In the path of the beam of white light is placed a Nicol prism in order to obtain a beam of plane polarized light. This prism is made of two crystals of Iceland spar, which are cemented together by Canada balsam in such a way as to obtain only a single beam of polarized light. The crystal is a doubly refracting medium ; that is, a light beam entering it is in general divided into two separate beams which are polarized and have different directions. One of these beams in the Nicol prism is disposed of by total reflection from the surface where the Canada balsam is located, and the other emerges a completely polarized beam ready for use. A second Nicol prism exactly like the first is iiow placed in the path of the polarized beam. This second prism is called the " analyzer," and is set so that its plane is just perpendicular to that of the first prism, called the "polarizer," so that all the light vibrations not sorted out by the one prism will be by the second. In this position, the planes being just perpen- dicular to each other, the prisms are said to be " crossed," and an observer looking through the analyzer finds the light totally extinguished as though a shutter interrupted the beam. By turning the analyzer ever so little from the crossed position, light passes through it, and its intensity increases until the planes of the prisms are parallel, when it again diminishes ; and if one of the prisms is rotated MANIPULATION OF COAST-DEFENSE GUNS. 721 there will be darkness twice every revolution. In order to accomplish this same end without actually rotating the analyzer a transparent medium which can rotate the plane of polarization of the light subject to the con- trol of an electric current is placed between the two prisms. The medium used is carbon bisulphide contained in a glass tube. To produce a mag- netic field in the carbon bisulphide a coil of wire through which passes an electric current, is wound around the glass tube. When the current ceases the carbon bisulphide instantly loses its rotatory power, and the ray of light is free to pass tiirough the prisms. Breaks in the current are made in the same way as in other ballistic chronographs. For a complete description of this instrument, with an account of experiments, see The Polarizing Photo-Chronograph, John Wiley & Sons, New York. MiAHfUPUI-ATIOlir Of COASI-DEFEVSE CJUWS. Until recently all gun carriages installed in the coast fortifications of the United States were designed for the use of hand power iu their manipula- tion. Tests, however, having demonstrated the adaptability of electrical power for this purpose, such guns are now being equipped with electric motors. The following data is taken from recent tests of the equipment of a 10- inch disappearing carriage. The equipment installed consists of : One 3 h.p. motor connected directly by spur gearing to the crank shaft of the traversing mechanism. One 5 h.p. motor for operating both the elevating mechanism and the retraction gear. A hand brake applied to a drum on main crank shaft of traversing gear. Control switches, wiring, etc. The iron-clad motors and switch boxes are water and dust tight. The mechanical hand brake is used to overcome the tendency of the carriage to settle back when stopped quickly at a particular point, due to the great weight and inertia. The weight of the gun is 67,000 pounds, and moving parts of carriage, approximately 170,000 pounds, a total of 237,000 pounds. TRAVERSING MOTOR. - S&esults. 130 volts. At full speed, Jg-^g-tortgt 1.1.8 effective H.P. H19 volts. A t h a 1 f sinppd J 23 amPeres to start. At halt speed. ^ 22 it running. ^2.9 effective H.P. (120 volts. «,„„„. „„„„ A J 23 amperes to start. Slowest speed. ^ 20 am^eres rurming. (.2.4 effective H.P. Time required to traverse through entire field of fire, 106° 30/ twenty-five seconds [of time]. ELEYATING AND RETRACTING MOTOR.— f 128 volts full speed. In depressing through extreme J 13 amperes full speed, range, + 15° to — 5°. 1 1.8 effective H.P. (^Time, 22 seconds. ( 122 volts full speed. In elevating gun through ex- J 20 amperes, full speed, treme range. | 1.8 effective H.P. ^Time, 22 seconds. ill CERTAIN USES OF ELECTRICITY IN U.S. ARMY. RETKACTION.- { 120 volts full speed. ■'. 20 amperes full speed. ( Time, 2 min. 2 sec. To bring gun from firing loading position. A more complete description of this apparatus may be found in the Electrical World and Engineer, January 19, 1901. ELECTRIC FUSES. It is often necessary to fire at a distance from the gun, as in experiments, and for this purpose electric fuses are used. The fuse consists of a ^-inch length of tine wire of platinum-iridium alloy, called the bridge, surrounded by a little gun-cotton or powder ; next to this c\/c FiCx. 11. Firing Key. A, copper case. B, hollow wood cap. CC, wires, .035 inch. D, bridge, .0025 inch. F, priming. H, fulminate of mercury, 10 to 24 grains. 1, paper discs held by drop of collodion. K, plug of beechwood. A, copper case. B, plug (beechwood). C, insulated wires. D, bridge. F, gun-cotton priming. H, rifle powder. I, cotton string. K, tin foil cap. Figs. 9 and 10. Electric Fuses. is placed, when required for detonating, a few grains of fulminate of mer- cury. The whole is usually fixed inside a copper case. The bridge being inserted in an electrical circuit is heated by the current which ignites the gun-cotton and fires the fuse. Fig. 9 shows a gun-fuse. Fig. 10 is a mine-fuse, which is similar in con- struction, and is used in firing high explosives, or where it is desired to DEFENSIVE MINES. 723 ignite several charges simultaneously, as in a group of submarine mines. Fig. 11 shows the tiring-key, in which T is a turnbuckle of ebonite which prevents accidental closing of the circuit. DEF1LYM1YK M L\EM. A mine is a charge of explosive contained in a case which is moored be- neath the surface of the land or water. The mines laid and operated in and around seacoast fortifications are for the most part defensive in their char- acter, fixed in position, and hidden. A defensive mine is either self-acting, — a mine which, once placed, ceases to be under control, and is tired by means within itself, mechanical or elec- trical,—or controlled, a mine fitted with electrical apparatus, which ena- bles a distant operator to ascertain its condition, and to fire it at any time ; it may also be tired automatically. A controlled mine may be fired in four different ways : (a) by contact with the mine only ; (b) at will of the operator only ; (c) by contact and will, both of which are necessary ; (4=/r> ^^ SPRING BOARD 3C Fig. 12. Electrical Land Mine. The sketch shows a self-acting electrical land mine, and is self-explana- tory. By using three lead wires the mine may be fired by the enemy's con- tact with it, or by the operator at the station. Circuit Closer or Torpedo. NS, circular permanent magnet with attached electro-magnets N and S. A, armature whose adjusting spring near K holds it away from the mag- net, while a weak current flows in through the electro-magnet coils in a direction to assist the permanent magnet. But if a stronger current flows, the armature is attracted, and sticks to the magnet, until a reverse current is sent in. The spring then draws the armature away, and breaks the contact of the circuit closer K on W. B, a brass ball f inch diameter, held by spiral S. T, a silk thread running through the vertical axis of the ball from adjust- ing screw to the armature. When the vessel strikes the mine the brass ball being knocked sidewise pulls, by means of the sti'ing, the armature against the poles where it sticks. R, 1000-ohm resistance coil, which is cut out of the mine circuit by the contact of K on W. PC, priming-charge. F, fuse. 724 CERTAIN USES OF ELECTRICITY IN U.S. ARMY. Operating-Box on Shore. WB', watching-battery of gravity cells and brass bar. FB', tiring-battery of Sampson cells and brass bar. P', firing-plug. M'M', ordinary electro-magnet, 100 ohms. (See Relay No. 7.) A7, armature pivoted at the center. (See Relay No. 7.) S', spring holding armature back against a weak current. (Relay No. 7.) I/, shutter arm pivoted above its center of gravity. When set as in relay No. 1, shutter-arm 1/ makes electrical connection with the armature at N' ; when armature is attracted it releases L/, whose lower end strikes a bell, and makes electrical contact with the firing-bar at B'. b, terminal of mine circuit which may be plugged to WB'. a, terminal for testing-set. o, o, two reversing-keys. Xand Y are two stations, 1 to 3 miles apart, each having a key and an ob- server of the mine field. Operation. The torpedo having been planted and connected with its relay, whose shutter-arm L' is set as in relay No 1, a small steady watching-current Hows through G', WB', b, M'M', H, NT/, J', O', V, coil S, cod, N, W, K (1,000 ohms), G to G' again. The direction of the current is such as to preserve the mag- netism of the magnet. If the circuit closer is accidentally closed (indicated by a change of the resistance in the circuit) it can be opened by using the reversing-key from shore. The fuse F may be tired in four ways : — (a) By contact with the mine only. Insert firing-plug P'. When a vessel strikes a mine the brass ball B in the circuit-closer is thrown aside, closing K on W and thus short circuiting R. The watching-current, thus made stronger, flows from coil N through K, A, Z, fuse, &„ to G'. Coming from gravity cells it cannot fire the fuse, but is strong enough to operate the relay and drop L/, which throws in the firing-battery. A strong current now flows through G", FB', P', B', J', O', V, coil S, coil N, W, K, A, Z, F, G„ to G" again, and fires the fuse. (b) At will of_ operator only, who may at any time drop the shutter arm 1/ by hand and insert the firing-plug. The firing-current is strong enough, even through R in the torpedo, to close K, short-circuiting R, and to fire the fuse. (c) By contact with the mine and at operator's will. Remove firing-plug P'. The watching-current flows as above in (a). When the vessel strikes the mine 1/ drops, striking the bell, when the operator inserts P', throwing in the firing-current which fires the mine. (d) By observation from two stations ; shutter arm L/ set, and firing-plug P' in. When a hostile vessel appears over the mine from the position of X the observer closes his key. Y has like instructions. When both keys are closed the main part of the current from WB' flows through G', WB', b, M'M', H, Q', X, Y, G, to G' again, drops the shutter-arm and fires the mine. For obvious reasons the foregoing is not a description of the service cir- cuit closer, but the principle of construction and operation of the mines of all countries are much alike. MISCELIAHEOUS. Fortress Telephones and Telegraphs. Covering as it does a considerable area, the modern fortification must have its several units within instant communication, in order to insure that con- cert of action so necessary to a successful command. The fort commander must communicate his orders to the battery commanders, and they in turn transmit the necessary commands to the gun commanders ; and while much time and ingenuity has been spent in devising means of communication through the medium of printing and dial telegraphs, the telephone is to-day practically the universal method of communication from one part of a fire command to another. As ordinary commercial telephones are employed, no special description of them need be given in this section. The telephone is, however, at best, but an unsatisfactory method of communication, and will be rendered more so by the noise and confusion of battle. DEFENSIVE MINES. 725 CIRCUIT CLOSER IN TORPEDO OPERATING BOX ON SHORE Fig. 13. Diagram of torpedo circuit closer and connections. G*3 726 CERTAIN USES OF ELECTRICITY IN U.S. ARMY. field Telephones and Telegraphs. But little is to be said of field telephones and telegraphs, as they do not differ from commercial instruments except in their portability. The wire is carried on reels mounted on wheeled trucks, and may or may not be strung on poles as the occasion demands. Light reels are also provided which may be strapped to a man's back to run wires to places otherwise in- accessible. The work to be done by field telegraphers is, however, an im- portant one in keeping a commander constantly in touch with his outposts. ELECTRICITY IN THE UNITED STATES NAVY. The application of electricity in ships in the United States Navy at the tresent time (July, 1901) is as follows : — All ship's lights, searchlights, and signal lights are entirely electric. Of power appliances the turret turning, elevating and loading of big guns, and hoisting ammunition, are always done electrically ; ship's ventilation is partly steam and partly electric, Avith the practice rapidly going to complete electric ; deck winches and boat cranes are usually steam, but very successful electric ones are in use ; steering-gear is entirely steam, hydraulically or mechanically controlled, and electric appliances are in the experimental stage ; an electric system of opening and closing water-tight doors is now in progress of development ; anchor-handling gear is entirely steam. Interior communication appliances are almost entirely electric, but are in some cases paralleled with mechanical equivalents, as for example voice tubes paralleling telephones. uvx.tno ROOM. The generating plant is located in a compartment called the " Dynamo Room," which is under the protective deck and adjacent to the boiler rooms, so as to secure a direct lead of steam pipes. «EATERATI9[G-§ETS. The following are the principal requirements contained in the standard specifications for generating-sets : — Generators. Generators to be of the direct current compound-wound multipolar type, giving 80 volts at the terminals. The compounding to be such that at the designed normal speed the voltage shall at no point of the external char- acteristic curve vary more than 1.5 volts from SO volts. There shall be no sparking whatever at the brushes when the generator is in operation with a constant load, nor shall there be any detrimental sparking with a change of one-half load, the brushes not being moved. The allowable temperature rises above the air after a four-hour run at full load are : — Field and armature windings 60° F. Commutator 72° F. The temperature of windings to be calculated from their resistance rise, and of the commutator to be measured by thermometer. Generator to stand an over-load of 33 per cent for two hours without injury, and the engine to be able to produce normal voltage with this over- load. Insulation resistance to be one megohm, tested with a pressure not greater than 1000 volts. The change of voltage at the terminals of the generator as measured on a dead-beat voltmeter not to exceed 10 volts, when full load is suddenly thrown on or off. External magnetic field to be inappreciable at a distance of 15 feet. Insulating substance used not to be injured by a temperature of 200° F. Engines. Engines to work most economically at 100 pounds steam pressure if com- pound, and 80 pounds if simple, vacuum being 25 inches ; but they must be able to work with pressure 20 pounds above and below these normal pressures. _~_ 728 ELECTRICITY IN THE UNITED STATES NAVY. Cylinders to be of hard cast iron cross-heads connecting rods, shafts, pistons and valve rods all nuts bolts, etc., to be of best forged steel. The design must be such that all parts subject to wear shall be accessible for adjustment and repair, especially those parts which by reason of wear would affect the alignment of the engine. Cylinders must be fitted with relief valves, arranged to work automati- cally, in addition to the usual drain cocks. All parts must be able to bear without injury the throwing on or off of the entire load by quickly making or breaking the external circuit of the generator. The governor must control the speed automatically, the throttle being wide open, within the following limits : Variation of Load. Variation of Steam Pressure. Allowed Speed Variation. Full load to 20% load . Constant normal 1\% Constant load .... 20 lbs. above to 20 lbs. below normal 3h% Full load to no load . 20 lbs. above to 20 lbs. below normal Wo If engines have more than one cylinder, the work done in each cyclinder must be practically equal at full load and normal pressure. Cylinders and valve chests must be covered with suitable non-conducting material. Cylinders must be fitted with indicator motions. It is very desirable that engines shall be capable of continuous running, without the use of lubricants in steam spaces. The gross weight of complete sets not to exceed one-third of a pound per watt of rated capacity. Generator and engine to be mounted on a common bed-plate and direct connected. The style of sets installed on the latest ships is a tandem compound engine with a six-pole generator, manufactured by the General Electric Company. The sizes used are 32 k.w. and 50 k.w. The two cylinders are cast together, the L.P. on top, and separated by a hollow cast-iron head, which forms the stuffing-box for the L.P. piston rod. The engine is entirely inclosed, and is provided with forced oil lubrica- tion for the main bearings, crank pin, wrist pin, and cross-head guides. Rocker arms, governor and valve stems are provided with automatic grease cups. A cylinder lubricator is provided, but is only used a few minutes before shutting down, so that the cylinders will be coated with a film of oil while standing idle. United States Metallic packing is used. 32 k.w. size runs at 400 r.p.m. and the 50 k.w. size at 310 r.p.m. Tests. The 50 k.w. sets of the U.S.S. " Kearsarge" and " Kentucky" gave the following average results on tests : STEAM CONSUMPTION AT FULL LOAD. Steam pressure 100 pounds Vacuum 24 inches Steam per I.H.P. per hour 21 pounds Steam per K.W 35.2 pounds Combined efficiency of set 80 % KEGULATION. Normal speed 310 r.p.m. Steam constant 100 pounds, full load to 20% load, gives variation of 1.56 % SWITCHBOARDS. 729 Constant full load steam 120 pounds to 80 pounds gives variation of 1.5% No load with 120 pounds to full load with 80 pounds gives variation of 3.85 % Normal voltage 80 volts Throw off full load suddenly gives total fluctuation of 9.6 volts Throw on full load suddenly gives total fluctuation of 6.9 volts HEATIXG AFTER FOUR HOURS FULL LOAD. Armature core surface 21° C rise Commutator bars 28 " Shunt-field spool surface 11.4 " Outboard bearing 7. " Armature conductors, by resistance 23.4 " Field conductors, by resistance ......... 17.7 " Engine has L.P. cylinder 18 inches diameter, H.P. 10£ incbes diameter, with stroke of 8 incbes. Clearance in H.P. cylinders is 7£%, in L.P. cylinder is 7i%. Weigbt of complete set is 15,000 pounds. STEAM-pipiarG. The dynamo room is supplied by a special steam pipe which usually is so connected that it can take steam direct from any boiler or from the auxil- iary steam pipe, it passes into a steam separator from which branches lead to each of the generating-sets in the dynamo room. This separator is drained by a steam trap which sends the water back to the hot well in the main engine room. The exhaust pipe from each set joins a common exhaust which connects with the auxiliary exhaust service of the ship. If the sets are located below the level of the ship's auxiliary exhaust pipe, a separator is placed in the common exhaust pipe before it goes up and joins the ship's auxiliary exhaust. This separator is drained by a small steam pump, which is automatically started and stopped by means of a float in the body of the separator, which float starts the pump when the separator is full and stops it when empty. SWITCHBOARDS. The general problem of the design of a generator switchboard for a naval vessel is to be able to connect any number of generators to any set of bus- bars. There are usually four separate sets of busbars, one for the lighting system, one for the power system, and one for the turning-gear of each turret. The Ward-Leonard system of motor control being used for turning the turrets, it is necessary to use a separate generator for each turret. Separate equalizer buses are provided for both the lighting and power systems. Current is supplied to the different appliances by means of distribution switchboards, which have two sets of busbars, one for lighting and one for power, and are connected directly to the corresponding busbars on the main generator board. Feeders run direct from these distribution boards, each feeder being provided with a fused switch. Distribution boards are sometimes located at various parts of the ship and sometimes made con- tinuous with the main board. The diagram of generator switchboard and turret turning system on page 12w shows connections as made on the U. S. S. " Illinois," except there are four more generators connected on exactly like the four shown. Each generator has a headboard carrying a double-pole circuit breaker, and clips for a series field short circuiting shunt used for turret turning. The diagram shows generators Nos. 1 and 2 operating in parallel on the power system, No. 3 alone on the light system, and No. 4 operating the after turret turning motors. It is to be noted that the three generators on the power and light- ing systems have the right-hand blades of their triple pole field switches closed, giving self-excitation through the field rheostat, while the machine for turret turning has the middle blades closed, giving separate field excita- tion from the power bus-bars and through the field resistance attached to the controller in the turret. 730 ELECTRICITY IN THE UNITED STATES NAVY. ■ WIRING. Specifications. The principal requirements of the Navy standard specifications for light and power conductors are : — All layers of pure Para rubber must contain at least ninety-eight (98) per cent of pure Para rubber ; must be of uniform thickness, elastic, tough, and free from Haws and holes. All layers of vulcanized rubber must contain not less than forty (40) per cent nor more than fifty (50) per cent of pure Para rubber ; must be concen- tric, continuous, and free from flaws or holes ; must have a smooth surface and circular section ; and must be made to a diameter in the finished con- ductor that will be in exact conformity with the diameter as tabulated. All layers of cotton tape must be filled with a rubber insulating com- pound, the tape to be of the width best adapted to the diameter of that part o'f the conductor which is intended to bind. The tape must lay one- half (i) its width, and be so worked as to insure a smooth surface and circular section of that part of the finished conductor which is beneath it. All exterior braid must be closely woven ; and all, except silk braid, must be thoroughly saturated with an insulating waterproof compound which will neither be injuriously affected, nor have any injurious effect on the braid, at a temperature of '200° F. (dry heat), or at any stage of test, the conductor being sharply bent. "Wherever a diameter over vulcanized rubber or outside braid is tabulated or specified, it is intended to secure a neat working-fit in a standard rubber gasket of that diameter for the pur- pose of insuring water-tightness of the joint, and no departure from such tabulated or specified diameter will be permitted. All conductors to be of soft annealed pure copper wire. No single wire larger than No. 14 B. & S. G. to be used. "When greater conducting area than that of No. 14 B. & S. G. is required, the conductor shall be stranded in a series of 7, 10, 37, 61,91 or 127 wires, as may be required ; the strand consisting of one central wire, the remainder laid around it concentrically, each layer to be twisted in the opposite di- rection from the preceding, and all single wires forming the strand must be of the diameter given in the American wire gauge table as adopted by the American Institute of Electric Engineers, October, 1893. The material and manufacture of the strand must be such that the measured conductivity of each single wire forming the strand shall not be less than ninety-eight (98) per cent of that of pure copper of the same number of circular mills, the measured conductivity of the conductor as a whole to be not less than ninety-five (95) per cent of that of pure copper of the same number of circular mills. Each wire to be thoroughly and evenly tinned. All conductors shall be insulated as follows : — First. — A layer of pure Para rubber, not less than one sixty-fourth (^) of an inch in thickness taped or rolled on; if lapped, the tape to lap one- half of its width. Second. — A layer of vulcanized rubber, of exact diameter as tabulated. Third. — A layer of commercial cotton tape, lapped to about one thirty- second (Jj) of an inch in thickness. Fourth.— A close braid to be made of No. 20 2-ply cotton thread, braided with three (3) ends for all conductors under 60,000 circular mills, and of No. 16 3-ply cotton thread braided with four (4) ends for all conductors of and above 60,000 circular mills. The outside diameter over the braid to be in exact conformity with that tabulated. Tests. Two samples, each 500 feet long, will be selected by the Bureau from the coils of wire to be supplied, and must be sent by the Contractors to the New York Navy Yard for test. (a) Both samples, after 24 hours imersion in sea water, must have an insulation resistance of not less than 1,000 megohms per nautical mile. (b) Test to be at 72° F. (c) To be tested by the direct deflection method at a potential of not less than 200 volts. (d) Both samples will be tested for a conductivity of not less than 96 per cent of that of pure copper, having a cross-section of the specified num- ber of circular mills. LIGHTING— SYSTEM. 731 (e) Chemical tests will be made to determine the constituents of the different layers of the insulation. (f) Braid will be tested for water-proof qualities. (g) Physical tests will be first made for qualities of strength, toughness, dimensions, etc. (h) The physical and electrical characteristics of the insulation under change of temperature will be tested by exposing the finished conductor for several hours at a time, alternately, to a temperature of 200° F. (dry heat) and the temperature of the atmospere, during a period of three days. (i) The tests for characteristics of the insulation will then be repeated and must show no practical deterioration on the results of the former tests. Methods of Insulating- Conductors. Three methods of insulating conductors are used. 1. Conduit ; 2. Molding ; and 3. Porcelain supports 1. Conduit is the principal method, being used in almost all spaces below the protective deck, and wherever wiring is exposed to mechanical injury or the weather. Iron-armored insulated conduit is used, except in maga- zines, and within 12-feet of the standard compass, where brass is used. Conduit passing through water-tight bulkheads is made water-tight by means of stuffing-boxes and hemp-packing. Water-tightness is provided at the ends of conduit by a stuffing-box and a soft-rubber gasket, through which the conductor passes. Long lines of conduit passing through several water-tight compartments are provided with gland couplings at proper intervals, which divide the run into water-tight sections, thus preventing an injury in a flooded compartment from allowing the water to run through the conduit into another compartment. These gland couplings are also used where conduit passes vertically through decks, and all vertical leads are run in conduit. 2. Wood molding is generally used in living spaces. It consists of a backing piece fastened to the iron work of the ship, to which the molding proper is secured by screws and covered with a wooden capping-piece. Where leads installed in molding pass through water-tight bulkheads, a bulkhead stuffing-box is provided for water-tightness. 3. Porcelain supports are used in dynamo rooms and for the long feeders which are run in the wing passages where there is no danger of interference. Stuffing-tubes are used where the wires pass through bulkheads, the same as with molding. Connection Boxes. All conductors are branched by being run into standard connection boxes, which are usually provided with fuses. Where conduit is used these boxes are tapped, to have the conduit screwed into them ; where molding or porcelain is used the boxes are provided with stuffing-tubes. The box covers are made water-tight with rubber gaskets ; inside the fuses and connection strips are mounted on porcelain bases. IIGHTI5fC.§T8TEl[. "Wiring1. The maximum drop allowed on any main is 3 per cent at the farthest lamp. Mains are required to be of the same size throughout, and to be of 1000 circular mills per ampere of normal load. fixtures. ' Most incandescent lamps are installed in air-tight glass globes of different shapes, depending upon position or location. Magazines are lighted by "Magazine Light Boxes," which are water-tight metal boxes set into the magazines through one of its walls, and provided with a water-tight door opening into the adjacent compartment, so that the interior of the box is accessible without entering the magazine. The sides of the boxes have 732 ELECTRICITY IN THE UNITED STATES NAVY. j glass windows, and each box is fitted with two incandescent lamps, each lamp having its own separate fused branch to the main, so that one lamp can be used as a spare. . " Switch Receptacles " containing a snap switch and a plug socket are I; provided for attaching portable lamps. Lamps. The principal requirements of the standard Navy specifications are : — They must be of the best quality and finish, and uniform size ; the bases must fit and be interchangeable in the standard socket. All leading-in wires and anchors must be fused in the glass ; all anchors must be made of metal. The filaments must be centered in the bulb, and must not drop when the lamps are run in a horizontal position. Each lamp must be marked on the inside of the bulb Avith the date of manufacture, and must have its rated candle-power, the voltage necessary to give this candle-power, and the name of the manufacturer conspicuously labeled on the outside of the bulb. The material used for cementing the bases to the bulb must be so treated as to insure against danger of short circuiting the lamp when exposed to moisture. When porcelain is used all holes must be filled. They must be designed for 80 volts, the rated candle-powei to be given at not less than 78 nor more tnan 82 volts. No fraction of a volt beyond these limits will be permitted. The efficiency of all 16 c. p. and 32 c. p. lamps must be not less than 3 j6n, nor more than 4 watts per candle-power, and that of 150 c.p. lamps not less than 3 f$ nor more than 3 ^ watts per candle-power, the efficiency to be measured when the lamps are new. The contractors shall guarantee that all lamps supplied will have an average life of at least 600 hours, and that the rated candle-power shall not have decreased more than 20 per cent after burning for this length of time at the initial potential. Before acceptance a test lot will be selected at random from the lot of each type of lamp delivered as follows ; — From lots not exceeding 50 lamps, all lamps. From lots exceeding 50 but not exceeding 500, 50 lamps. From lots exceeding 500 lamps, 10 per cent of the lot. The test lot will be subject to the following tests: — (a) For design, dimensions, and construction. (6) For vacuum, by trembling of filament and spark. (c) For voltage and efficiency when rotating at a speed of 180 revolutions per minute. (d) For rated candle-power by standard photometer. A secondary standard lamp, standardized from the Bureau's standards, will be used in the tests. A failure of 30 per cent of the test lot to comply with foregoing specifica- tions will cause rejection of the lot represented by that test lot. Divingr-ljanterns. Diving-lanterns consist of a glass cylinder closed at each end with a metal cap, having the joint between the glass and metal packed with a soft-rubber gasket. On the inside of one of the caps is provided a standard marine lamp-socket for 100 candle-power incandescent lamp, to which is connected 100 feet of twin conductor cable, at the other end of which is connected a double pole plug for a standard marine receptacle. When first submerged a considerable amount of moisture is deposited in the inside, which is drawn out through a small hole made water-tight by a screw with a rubber gasket. Searchlig-lats. The requirements of the standard Navy specifications are : — It shall, in general, consist of a fixed pedestal or base, surmounted by a turntable carrying a drum. The base shall contain the turning mechanism and the electric connections, and be so arranged that it can be bolted securely to a deck or platform. LIGHTING-SYSTEM. 733 The turntable to be so designed tbat it can be revolved in a horizontal plane freely and indefinitely in either direction, both regularly and gradu- ally by means of a suitable gearing, and rapidly by hand, the gearing being thrown out of action. The drum to be trunnioned on two arms bolted to tbe turntable, so as to have a free movement in a vertical plane, and to contain the lamp and re- flecting mirror. The drum to be rotated on its trunnions, both regularly and gradually by means of suitable gearing, and rapidly by hand ; the gear- ing being thrown out of action. The axis of the drum to be capable of a movement of not less than 70° above and 30° below the horizontal. The drum to be thoroughly ventilated and well-balanced ; to be fitted with peep sights for observing the arc in two planes, and with hand holes to give access to the lamp. It must be so designed that eitber a Mangin or a para- bolic mirror can be used, and means for balancing it with either mirror must be provided. The mirror to be made of glass of the best quality, free from flaws and holes, and having its surface ground to exact dimensions, perfectly smooth and highly polished. Its back to be silvered in the most durable manner ; the silvering to be unaffected by heat. To be mounted in a separate metal frame lined with a non-conducting material, in such a manner as to allow for expansion due to heat and to prevent injury to it from concussion. The lamp to be of the horizontal carbon type, and designed for both hand and automatic feed. The feeding of the carbons must be effected by a posi- tive mechanical action, and not by spring or gravitation. It must burn quietly and steadily on an 80-volt circuit in series Avith a regulating rheostat, and shall be capable of burning for about six hours without removing the carbons. The front of the drum to be provided with two glass doors, one composed of strips of clear plate glass, and the other of strips of plano-concave glass lenses, so designed as to give the beam of light projected from the mirror a horizontal divergence of at least 20°. The doors to be interchangeable, and to be so arranged that either can be put in place on the drum easily and quickly. Electrically Controlled Projector. To be in all respects similar to the hand controlled, with the addition of two shunt motors, each with a train of gears ; one motor for giving the ver- tical and the other the horizontal movement of the projector. The motors and gears to be contained in the fixed base, and to be well protected from moisture and mechanical injury. A means to be provided for quickly throwing out or in the motor gears, so that the projector can be operated. ! either by hand or by motor, as desired. The motors to be operated by means of a compact, ligbt, and water-tight I controller, which can be located in any desired position away from the pro- i jector. The design of the controller to be such that the movement of a | single handle or lever, in the direction it is wished to cause the beam of I light to move, will cause the current to flow through the proper motor in the I proper direction to produce such movement. The rapidity of movement of ! the projector to be governed by the extent of the throw of the handle or j lever. A suitable device to be included whereby the movement of the pro- jector can be instantly arrested when so desired. All projectors to be finished in a dead-black color throughout, excepting the working-parts, Avhicb shall be bright. The lamps to be designed to produce the best results when taking current as follows : 18-inch, 30 to 35 amperes ; 24-inch, 50 to 60 amperes ; 30-inch, 75 to 90 amperes. The 18-incb projector shall project a beam of light of sufficient density to render plainly discernible, on a clear, dark night, a light-colored object 10 by 20 feet in size, at a distance of not less than 4,000 yards ; the 24-inch pro- jector, at a distance of not less than 5,000 yards ; and the 30-inch projector, at a distance of not less than 6,000 yards. The connections for the electrically controlled projectors as manufactured by the General Electric Company are shown in the diagram. The fields of the two training motors an in series with each other and connected across the 80-volt circuit. Both horizontal and vertical training can be simultane- ously produced. The controller-handle when released, is brought to the off 734 ELECTRICITY IN THE UNITED STATES NAVY. POWER SYSTEM. 735 position by springs and short circuits both motor armatures thus stopping all movement. The horizontal training motor drives through a worm gear, and the verti- cal motor through a revolving nut on a vertical screw shaft : all gearing can be easily thrown out for quick hand control. The highest speeds are 360° in 30 seconds horizontally, and 100° in 60 seconds vertically. The motors may also be operated at four lower speeds. The lamp has a striking magnet in series with the arc and feeding magnet in shunt with the arc. When the arc becomes too long, sufficient current is forced through the shunt feeding magnet to cause it to make its armature vibrate back and forth, and thus move the carbons together through a ratchet which turns the feed screws. The point at which the magnet Avill begin to feed is adjustable by means of a spring attached to armature. The feed screws are so proportioned that the positive and negative carbons are each fed together at the same rate that they are con- sumed, thus keeping the arc always in the focus of the mirror. Sight holes are provided through which the arc may be watched. A permanent magnet, fastened to the inside of the projector and surrounding the arc on all sides but the top, causes the arc to burn steadily near the upper edge of the carbons and in focus with the mirror. The rheostat is located near the switchboard, and after being once set for proper working does not need to be again changed. Double-pole circuit breakers are used at the switchboards for switches. SIGXAX. EIGHTS. Ardois Signals. The Ardois signals consist of four double lanterns, each containing a red and a white light, which are hung from the top of the mast, one under the other and several feet apart. By means of a special controller any number of lanterns may have either their red or white lamps lighted, thus producing combinations by which any code can be signaled. The lamps used are clear, and the color is produced by having the upper lens which forms the body of the lantern colored red ; the lower lens is clear. The controller consists of eight semi-circular plates, with pieces of hard rubber set in the inner edges where needed, and a rotating center stud with eight plunger contacts rubbing on the edges of the plates. By suita- bly placing the pieces of hard rubber for any given position of the contacts, any desired combination of lights can be produced. The operation consists in moving the arm carrying the contacts to the position desired (as shown by a pointer on an indicating dial; and closing the operating switch, when the proper lamps will light. Truck Xiig-lsfs. The truck lights are lanterns of construction similar to the Ardois lanterns, mounted, one on the top of both the fore and main masts. By means of a special controller the red or white light in either lantern can be lighted. POWJEK SYSTEM. Motors are kept entirely separate from lights by the use of different bus-bars on the generator switchboard and distribution boards. Each motor or group of motors is supplied by its own feeder running from the distribution board, where it has its own fused switch. A maximum drop of 5 per cent is allowed. Principal Requirements of Specifications for Motors. Motors to be wound for 80 volts direct current. Sizes above 4 H. P. to be multipolar ; 4 H. P. and below may be bipolar. Armatures to be of iron-clad type, and coils preferably to be separately Wound and easily removable. 736 ELECTRICITY IN THE UNITED STATES NAVY. Band wires to be of non-magnetic material, and not more than three to be used under poles. Commutator segments to be of pure copper, insulated with mica of such quality that it will wear evenly with the«copper. Carbon brushes to be used carrying not more than 30 amperes per square inch at full load. PILOT LAMP Fig. 2. Diagram of Ardois Signal Set. POWER SYSTEM. 737 No sparking to occur up to full load with no shifting of the brushes. To prevent deterioration from rust and corrosion, such parts as holts, nuts, screws, pins, and fittings of a light character, which if rusted or corroded would injure the operation, strength, ease of adjustment or taking apart, or appearance, are to be made of tobin bronze, or similar metal, and not of iron or steel. No insulating substances to be used that can be injured by a temperature of 94° 0. Test for dielectric strength to be made with a pressure of 1500 volts alternating for 60 seconds, using a transformer and generator of at least 5 k. w. capacity. Allowed temperature rises above surrounding air are : — Continuous running motors, open type, windings 35° C, commutator 40° C, after eight hour full-load run. Same as above, but closed type, 50° C, for both winding and com- mutator. Intermittent running motors have special requirements depending upon use ; but nearly all require 45° C. for all parts after one hour at constant full-load. Bearings of all motors 40° C. Lubrication of continuous running motors is by oil rings or slight feed cups, the intermittent running motors by grease pockets. Every motor to be protected by an automatic circuit-breaking device, capable of being set to 50 % above the normal full load. Turret-Turning* Gear. The motors are controlled by the Ward-Leonard system, the principle of operation of which is illustrated by the elementary diagram on the diagram of generator switchboard and turret-turning system, page 12w. The motors are shunt wound, and have the fields constantly separately excited from the bus-bars of the ship's power system. A separate generator is required which cannot be used for any other purpose when used with the turret. The generator is also separately excited from the power bus-bars; but a variable rheostat, located in the turret, is connected in the shunt- field circuit. The brushes of the motor are directly connected to the brushes of the generator, and the generator is kept running at constant speed by its driving-engine. It is now evident that by varying the rheostat in the turret, the held excitation, and consequently the voltage produced by the generator, will be varied ; and any variation in the voltage of the generator will produce a corresponding variation in the speed of the motor, which has a constant field from separate excitation. The direction of rota- tion of the motor is reversed by reversing the leads to the armature. The actual connections for the application of the above principles are shown in the main part of the diagram. Generator No. 4 is shown connected for operating the after-turret. Closing the after-turret field switch and the center blades of the generator field switch, separately excites the fields of the motors and generator from the power bus-bars. The regular field rheostat of the generator is entirely disconnected, and a rheostat located in the turret and operated by the tur- ret turning controller is used instead. Closing the positive and negative single-pole switches on the after-turret bus-bars connects the generator armature to the motor armatures, through a circuit breaker, the reversing contacts of the controller, and separate armature switches for each of the two motors, which are operated in parallel. The controller has one shaft, at the top of which are located the con- nections for the generator field rheostat, so arranged that as the controller is turned either way from the off position the rheostat is gradually cut out ; below are located the reversing contacts, which reverse the connections between the generator armature and the motor armatures ; ■ these contacts are so arranged that at the off position the motor armatures are entirely disconnected from the generator, and are short-circuited through a low resistance called the " Brake resistance." The effect of this brake resist- ance is to bring the turret to a quick stop when the controller is brought to the off position, as the motor armatures revolving in a separately excited field generate a large current, which passes through the braking resist- ance, and thus absorbs the kinetic energy of the turret, giving a quick and 738 ELECTRICITY IN THE UNITED STATES NAVY. POWER SYSTEM. 739 smooth stop. In parallel with each of the large main fingers of the re- versing contacts is a small auxiliary finger and an auxiliary resistance connected to it. This auxiliary finger makes contact a little before and breaks it a little after the main linger, and thus reduces the sparking. The controller is also provided with a magnetic blow-out for reducing sparking at contacts. When used on this system for operating a turret the generator has its series coil short circuited by a very low resistance shunt, so that it has very little effect on the field excitation, but this resistance is so proportioned that enough of the total current generated by the generator will pass through the series coil to give a quick and positive start of the turret ; be- cause if the series coil is absolutely short circuited, and only the separately excited shunt coil used, the time required for the held to build up is suffi- cient to make the starting of the turret very sluggish and irregular, and pre- vents very tine training from being obtained. On the U.S.S. " Kearsarge " and " Kentucky," two 50 H. P. motors of 400 r.p.in. are used to turn each double turret, which weighs 710 tons and is mounted on 32-rlanged conical rollers, 15|-inches diameter, running on a track 21 feet in diameter. Each motor drives through a worm and wheel, connected to a spur pinion meshing into a stationary circular rack. The motors are geared together by a cross shaft. Friction clutches are inserted in the transmission gearing to prevent sudden stops, firing the guns, or im- pact of shot, from breaking the gearing. Full speed of the turret is at the rate of one revolution per minute. The controller is provided with a me- chanical automatic stop which brings it to the off position when the turret reaches the limit of its train at either side. The following results were obtained on test of the four turrets of the two ships. The friction varied considerably for different turrets. Forward turret of the " Kearsarge " gave : — Turning at constant full speed, Input of motors 22 E.H.P. Output of motors 13 H.P. Maximum when accelerating at rate of attaining full speed in 10 seconds, Input of motors 44.5 E.H.P. Output of motors 36.3 H.P. This was the easiest running of the four turrets. The hardest running gave, Turning at constant full speed, Imput of motors 41 E.H.P. The motors are seen to be greatly over-powered for the work, this to allow for overcoming any deformation of track, rollers, etc., which might occur during action. Fineness of train obtained : — The turrets were easily started and stopped with a resulting movement of 10 seconds of arc, which equals a movement of about 2 inches at 1,000 yards range. This is a movement much smaller than the visual angle covered by the cross hair of the sighting telescope, so that the fineness of train is much greater than that of sighting. A turret was turned thro ugh its extreme train from one side to the other 48 times in one hour, with a stop being made at each beam position during each trip. The motors used were entirely inclosed and weighed 5,700 pounds. JLoading* and Training: Gear for Guns. Guns of 12-inch and over are elevated and rammed by power, smaller guns have hand gear. The elevating gear for 12-inch and 13-inch guns consists of a 2J H.P., SO-volt, 300 r.p.m. series motor, geared to a revolving screw which raises or lowers a nut crosshead from which connecting rods go to the gun. Ordinary rheostatic control is used with no braking appliance. To train a 13-inch gun at the rate of 30° per minute, an armature input of from 1.5 to 740 ELECTRICITY IN THE UNITED STATES NAVY. 3 E.H.P. is required, depending npon the condition of the load and whether elevating or depressing. The motors used are entirely inclosed and weigh 550 pounds. Rammers consist of a telescopic tube worked through spur and chain- gearing by a 5 H.P., 80-volt, 775 r.p.m. series motor. A friction slip clutch is inserted in the gearing to prevent damage when the shell seats itself in the breach. Ordinary rheostatic control is used. When ramming a shell but little power is required, as the shell slides along the breech, but as it is being forced to its seat at the end of the breech chamber a sudden rush of current of from two to three times the full-load current of the motor is produced. The motors used are similar to the elevating motors, except wound for higher speed. AUKmUNITIOlV HOISTS. Power ammunition hoists are of two kinds ; first, those in which a car or cage is hoisted up and down by a line wound on a drum on the motor counter-shaft ; and second, those in which the motor runs an endless chain provided with toes or buckets on which the ammunition is placed and con- veyed up through a trunk. Hoists for 13-inch and 13-inch Ammunition. These hoists are of the first kind. The motor frame is provided with bearings for a counter-shaft, geared by a spur-gear and pinion to the arma- ture shaft; on the counter-shaft is mounted a grooved drum for the hoisting- cable. On the armature shaft is mounted a solenoid band-brake. The cores of the solenoid are weighted and attached to the brake-setting lever so that when free their weight is sufficient to hold the loaded car from falling; when the solenoids are energized the cores are drawn up and the brake re- leased. The controller is constructed so that on the off position the solenoids are not energized and the brake is set ; but at all other points, both hoisting and lowering, the solenoids are energized and the brake released. Shunt motors are used, and the control for hoisting is ordinary rheostatic ; the resistance being put in series with the armature and gradually cut out, the field is always constantly excited as soon as the feeder-switch is closed. For lowering, the entire rheostat is thrown directly across the line, one armature lead connecting to one side of the line and the other lead gradu- ally moved (as the motor is brought to full speed) from the condition of a short-circuited armature at the off position to direct connection to the other side of the line at the full on position ; in all intermediate positions the armature is in shunt with a part of the rheostat. The object of this is to cause the armature to take current from the line and run as a motor when lowering a light load which will not overhaul, but to run as a generator and send current through the rheostat if the load is very heavy and overhauls the motor and gearing. In either case the speed will depend upon the amount of the rheostat that is in shunt across the armature. The off posi- tion of the controller short-circuits the armature, and since the fields are always excited, this gives a quick stop and also holds the load. The 13-inch hoists of the tJ.S.S. "Kearsarge" and "Kentucky" used 20 H.P. motors running at 350 r.p.m., with a gearing ratio of 6.43 from arma- ture to counter-shaft. The load was, empty car 1,846 pounds, and full charge 1,628 pounds, or a total of 3,474 pounds. The following average results were obtained when testing at hoisting full charge : — Hoisting-speed, feet per minute 180 Mechanical H.P. in load 18.96 Input of motor, E.H.P . 28.5 Total efficiency 66.6% Motors were designed to be suspended under the turret, were entirely inclosed, and weighed 3,000 pounds complete with brake. AMMUNITION HOISTS 741 Hoists for 8-inch Ammunition. Hoists for smaller ammunition are made and controlled in a similar manner as the above, except the solenoid brakes are replaced with an ordi- nary band-brake, operated by a foot or hand lever. The 8-inch hoists used a 6 H.P.,375 r.p.m. shunt motor to hoist a total load of 910 pounds at 163 feet per minute. Tests gave average results of, — Mechanical H.P. in load 4.5 Input of motor, E.H.P 7.4 Total efficiency . [ 60.8% 742 ELECTRICITY IX THE UNITED STATES NAVY. Endless Chain Auimuuitioii Hoists. These hoists run continuously, the ammunition being fed in as desired. The motor is geared to the chain sprockets by spur gearing, is shunt wound, and is started and stopped by a controlling panel, which is provided with no voltage and overload release, a held rheostat fur varying the speed of the motor, and a reversing-switch. A solenoid brake, similar to the one above described for the 13-inch hoist, is mounted on the armature shaft, and is set when the starting-arm is in the off position, but has its coils energized and is released when the arm makes the first contact in starting. At the full on position, part of the starting rheostat is in series with the brake, thus cutting down the current consumed by it. This does not affect the reliability of the brake, since the current required to hold up the cores is much less than that required to first start them, and at the start the full-line voltage is on the coils. To lower ammunition the reversing-switch is thrown down, which re- verses the connections to the motor armature, and puts in the armature circuit a safety switch. This safety switch is attached to the lever which operates the catch pawls in the hoist trunk. These pawls will allow am- munition to go up, but will catch and prevent it from going down, and are used to keep the ammunition from falling in case any part of the hoist should be shot away. When the pawl lever is thrown down it throws the pawls out of action, and allows ammunition to be lowered by reversing the motor ; it also closes the safety switch which completes the armature cir- cuit for the lowering position of the reversing-switch. This style of hoist is used for all kinds of ammunition up to and includ- ing 6-inch. Packages are so made that they weigh about 100 pounds each. Motors rated at 3| H.P., continuous running, with speed variation of 360 to 475 r.p.m. are used ; power required varies greatly with kind and style of hoist. Motors are entirely inclosed and weigh 980 pounds. BOAT CRAIES. For handling steam cutters and other boats a revolving crane having the general shape of a davit is used ; it extends down to the protective deck, and has a steady bearing at each deck passed through, and the weight is carried by a roller thrust bearing. The operating machinery is carried on a circular platform fastened to the crane. The cranes for the U.S.S. "Kearsarge" and "Kentucky" have two motions ; namely, rotating the entire crane, and raising or lowering the hook. One motor only is used for both motions, clutches and gearing being used to produce either at will. Two counter-shafts are driven by the motor, each having a worm at the end, one driving a worm wheel on the hoisting-drum and the other a worm wheel on the shaft of the rotating pinion. Each of the counter-shafts contains a friction clutch, so that it can be disconnected from its worm at will. A band-brake is provided on the rotating-worm to hold the crane from rotating. A strap brake is provided on the hoisting-drum, which consists of a wrought-iron strap, one end of Avhich is permanently fastened to the platform, wound three times around the hoisting-drum and the free end attached to a weighted lever which pulls it taut. This strap is wound around the drum in the direction it turns when lowering, so that any motion in this direction causes the friction to make the strap bind tighter and hold the drum from turning ; but rotation of the drum in the hoisting direction causes the friction to make the strap loosen up and allow the drum to continue rotating. Thus the brake automatically holds the load from over-hauling the drum when the motor is disconnected. For lowering, the brake has its free end raised by a hand lever, thus loosening it, and allow- ing the drum to turn in the lowering direction. The motor is shunt wound with field constantly excited as soon as the feeder switch is closed at the distribution board. The controller cylinder gives ordinary rheostatic control with resistance in series with the armature, but there is a commutating switch which when closed gives the same kind of control as used for lowering with the 13-inch ammunition hoist described above ; this control is used for lowering and BOAT CRANES. 743 rotating, since it gives a smoother stop, and the rheostatic control is used for hoisting. The off position of the controller short circuits the arma- ture, giving a quick and positive stop. A 40-foot steam cutter is the largest boat handled, and weighs complete 16,000 pounds. Fig. 5. Diagrams of Connections for Boat Crane Motors The weight of the complete crane is 54,000 pounds. Motor is 50 H.P,, 400 r.p.m., is entirely inclosed and water-tight, and weighs 5,890 pounds. Current is supplied through collector rings mounted on the cranes. The controller is water-tight, and the circuit breaker is 744 ELECTRICITY IN THE UNITED STATES NAVY. mounted in a water-tight iron box ; all were tested for water-tightness by playing a stream of salt water on them from the fire-hose. The following results were obtained on test : — Load of 16,000 pounds Hoisting-speed, feet per minute ... 25 Mechanical H.P. in load 13.64 H.P. Input of motor to hoist ...... 30.6 E.H.P. Total efficiency 44.5 % Rotating speed 1 r.p.m. Imput of motor to rotate 14.8 E.H.P. EMPTY HOOK. Input of motor to hoist 7.3 E.H.P. Input of motor to rotate 8.9 E.H.P. It is seen that the motor is very much overpowered for the ordinary work required, but this is done to have a large margin to be able to handle boats in rough weather when the ship is rolling. Especial strain will be pro- duced when rotating a boat in when the ship is heeled over, and also from the inertia effect of rolling. DECK WIWCHES. The electric deck winches of the U.S.S. "Kearsarge" and "Kentucky" consist of a series motor geared through a system of spur-gearing to the shaft carrying the winch heads. The control is ordinary rheostatic, with the controller suspended horizon- tally from the deck underneath the winch and operated by a vertical shaft and a pair of bevel gears. Braking is accomplished by a foot lever, operat- ing a brake-band. For ordinary working the controller is turned to the full speed and the winch allowed to run continuously, the load being con- trolled by taking several turns of the hoisting-rope around the winch head. The maximum load can be very nicely controlled in this manner. The capacity of the winches is 2,200 pounds at 300 feet per minute ; and two winches are provided with a compound gear which can be thrown in to give a speed of 50 feet per minute with a corresponding pull of 13,000 pounds. The motors are 25 H.P., with a full-load speed of 320 r.p.m., but when the winch is allowed to run without load the speed of the motor increases to about 900 r.p.m. When hoisting 2,200 pounds at 300 feet per minute, the average test results were : — Mechanical H.P. in load 20 H.P. Input of motor 34.3 E.H.P. Total efficiency 58.4% Motors are entirely inclosed and water-tight, and were tested for water- tightness by playing a stream of salt water from the fire-hose on them without any water entering. ViafTILATIOHr FAlfS. Nearly all compartments of a ship have artificial ventilation by power fans ; both exhaust and pressure systems being employed. Both steam and electric drive is used, steam being used almost entirely for forced draught in the boiler rooms, while electric predominates for all other places. Shunt motors are used, started, and stopped by a controlling panel having " no voltage"' and "overload" release. Speed variation is obtained by a field rheostat. The following table gives results of tests on different sizes and styles of fans when run at full load and speed : STEERING-GEAR. 745 Fan. 03 5 o 03 03 > £ a H H Kind -d >- -8.5 3 o !W *w Name of Firm. of — ._ J) • Work. l~ T.r= ■2«S u £w c3 g,00 i g^j 0^2 s o ° CO © H M W ^ ft fc ,4 Lane & Bodley .... E. & W.W. 58 132 2.27 J. A. Fay & Co W. W. 100 15 85 15 300 M.I Hi 3.53 Union Iron Works . . E.,M. M. 400 95 305 23 1G00 4.0ii 5.24 Frontier Iron &Brass W'ks M.E.,etc. 25 8 17 32 150 i ;.oi) 8.82 Taylor Mfg. Co E. 95 230 '1.42 Baldwin Loco. Works L. 2500 2000 500 80 4100 1.64 S.20 W. Sellers & Co. (one de- partment) H.M. 102 41 61 40 300 2.93 4.87 Pond Machine Tool Co. . M. T. 180 75 105 41 432 •'.40 4.11 Pratt & Whitney Co. . . " 120 725 6.04 Brown & Sharpe Co. . . 230 900 3.91 33000 " Horse-power = .0001904 x d X w x revolutions per minute. POWJEH lTSE» BY MlCHIi\E-T«OI§. (K-. E. Dinsmore, from the Electrical World.) 1. Shop shafting 2T35 in. x 180 ft. at 160 revs., carrying 20 pulleys from 6 in. diam. to 36 in., and running 20 idle machine belts . 1.32 H. P 2. Lodge-Davis upright back-geared drill-press with table, 28 in. swing, drilling § in. hole in cast iron, with a feed of 1 in. per minute 0.78 H. P, 3. Morse twist-drill grinder No. 2, carrying 26 in. wheels at 3200 revs ' 0.29 H. P. 4. Pease planer 30 in. x 36 in., table 6 ft., planing cast iron, cut \ in. deep, planing 6 sq. in. per minute, at 9 reversals .... 1.06 H. P. 5. Shaping-machine 22 in. stroke, cutting steel die, G in. stroke, \ in. deep, shaping at rate of 1.7 square inch per minute . . . 0.37 H. P. 6. Engine-lathe 17 in. swing, turning steel shaft 2| in. diam., cut T3g deep, feeding 7.92 in. per minute 0.43 H. P. 7. Engine lathe 21 in. swing, boring cast-iron hole 5 in. diam., cut j3s diam., feeding 0.3 in. per minute 0.23 H. P. 8. Sturtevant No. 2, monogram blower at 1800 revs, per minute, no piping 0.8 H. P. 9. Heavy planer 28 in. X 28 in. X 14 ft. bed, stroke 8 in., cutting steel, 22 reversals per minute 3.2 H. P. Horse-power in Machine-shops; friction; Men Employed. (Flather.) MISCELLANEOUS. 759 Horse-power in Machine-shops. Horse-power. cS § © The followiug-named tools will probably be required in constructing lines for city or commercial ligbting : (Davis.) Stubs' pliers, plain Climbers and straps Pulley-block and ecc. clamp Come-along and strap . . . . Splicing-clamps Linemen's tool-bag and strap . Soldering-furnace Gasoline blow-pipes Soldering coppers Pole-bole sbovels Pole-bole spoon, regular . . . Octagon digging-bars . . . . Tamping-bars Crowbar , Pick-axe Carrying-hook, beavy . . . , Cant-hook Pike-poles Pole-supporter Comb, pay-out reel and straps . Nail-hammer Linemen's broad hatchets . . Drawing-knives Hand-saw Ratchet-brace, bits Screw-drivers Wrench Bastard file Size. Cost about 8 in. $2.00 3.00 ( To 8.00 \ No. 3 2.25 (B. &S. 2.50 4.80 6.00 6.00 2 1b. .95 8 ft. 1.50 7 ft. 1.25 8 ft. 3.50 7 ft. 2.60 10 1b. .90 .75 6.00 4 ft. 2.00 16 ft. 2.40 6 ft. 12.00 20.00 lib. 1.00 6 in. 1.50 12 in. 2.10 26 in. 1.50 10 in. 3.00 8 in. .80 12 in. 1.25 12 in. .30 APPROXIMATE LliT OF SUPPHIS REQUIRED IN INSTALLING 15 CITY LAMPS AND 20 COMMERCIAL LAMPS ON A FIVE-MILE CIRCUIT, SETTING POLES 132 FEET APART. (Davis.) Size or Diameter. Price about Quantity. Electric-light poles Electric-light poles Electric-light poles Cross-arms, 4-pin . Painted oak pins . Oak pins and bolts Iron break-arms . Lag-screws and washer Glass insulators, D. G. Pole steps .... Guy stranded cable . Cross-arm brace and bolts Line wire 30 ft., 35 ft., 40 ft., 4 ft. ljin. l£in. 7 in. 7 in. £X71 f X 8*i $2.40 each 4.15 5.50 .07 .75 .04 ■n .05 .07 lb. .20 each 125.00 mi. 40 200 800 24 25 400 850 2500 500 lbs. 40 6 miles MISCELLANEOUS. 761 MATERIAL HE«V1 1 IIK1> FOR CONITOCTIIYC} IN 1AOTP§. Sleet-proof pulleys . . Street-lamp cleats, iron Arc-lamp cordage . Suspension cable . Hard-rubber tube . Soft-rubber tubing Arc cut-out . . . Porcelain insulators , Oak brackets and spikes (Davis.) £in. $0.75 eacb. 30 .25 " 15 1.25 bd. ft. 25 .02i ft. 3000 ft. 1.50 lb. 5 lbs .20 ft. 200 ft. 3.50 eacb 20 2.40 bd. 400 2.50 " 150. 'NATIONAL ELECTRICAL CODE.' RULES AND REQUIREMENTS OF THE NATIONAL BOARD OF FIRE UNDERWRITERS FOR THE INSTALLATION OF WIRINHi AND APPARATUS FOR ELECTRIC LIGHT, HEAT, AND POWER AS RECOMMENDED BY THE UNDERWRITERS' NATIONAL ELECTRIC ASSOCIATION. EDITION OF 1901. The National Electrical Code, as it is here presented, is the result of the united efforts of the various Electrical, Insurance, Architectural, and allied interests which have, through the National Conference on Standard Elec- trical Rules, composed of delegates from various National Associations, unanimously voted to recommend it to their respective Associations for approval or adoption. The following is a list of the Associations represented in the Conference, all of which have approved of the Code : American Institute of Architects. American Institute of Electrical Engineers American Society of Mechanical Engineers American Street Railway Association Factory Mutual Fire Insurance Companies National Association of Fire Engineers National Board of Fire Underwriters National Electric Light Association Underwriters' National Electric Association OE^fRAL M,A]¥ «OVERJI\C THE ARRAK^E- ittEarx of rilei. CLASS A. — Central Stations, Dynamo, Motor, and Storag-e- Battery-Rooms, Transformer Substations, etc. Rules 1 toll. CLASS B. — Outside Work, all systems and voltages. Rules 12 and 13. CLASS C — Inside Work. Rules 14 to 39. Subdivided as follows : General Mules, applying to all systems and voltages. Rules 14 to 17. Constant-Current systems. Rules 18 to 20. Constant-I*otentiai systems. All voltages. Rules 21 to 23. Voltage not over 550. Rules 24 to 31. Aroltage between 550 and 3,500. Rules 32 to 37. Voltage over 3,500. Rules 38 and 39. CLASS D. — Specification for Wires and Fitting's. Rules 40 to 63. CLASS E. — Miscellaneous. Rules 64 to 67. CLASS F. —Marine Wiring-. Rules 68 to 80. CIASS A.-§TATIOS§ AVD DYKAMO ROOMS. INCLUDES CENTRAL STATIONS, DYNAMO, MOTOR, AND STORAGE-BATTERY ROOMS, TRANSFORMER SUBSTATIONS, ETC. 1. Generators — a. Must be located in a dry place. b. Must never be placed in a room where any hazardous process is carried on, nor in places where they would be exposed to inflammable gases or flyings of combustible materials. 762 CLASS A. STATIONS AND DYNAMO IIOOMS. 763 ; c. Must be insulated on floors or base frames, which must be kept filled ;o prevent absorption of moisture, and also kept clean and dry. Where Erame insulation is impracticable, the Inspection Department having juris- diction may, in writing, permit its omission, in which case the frame must jDe permanently and effectively grounded. ! A high-potential machine which, on account of great weight or for other reasons, cannot have its frame insulated from the ground, should be sur- rounded with an insulated, platform. This may be made of wood, mounted jn insulating supports, and so arranged that a man must always stand upon lit in order to touch any part of the machine. I In case of a machine having an insulated frame, if there is trouble from static electricity due to belt friction, it should be overcome by placing near the belt a metallic comb connected with the earth, or by grounding the frame through a very high resistance of not less than 200 ohms per volt generated by the machine. i d. Every constant-potential generator must be protected from excessive current by a safety fuse, or equivalent device, of approved design in each lead wire. These devices should be placed on the machine or as near it as possible. ;, Where the needs of the service make these devices impracticable, the ' Inspection Department having jurisdiction may, in writing, modify the requirements. e. Must each be provided with a waterproof cover. I /. Must each be provided with a name-plate, giving the maker's name, the capacity in volts and amperes, and the normal speed in revolutions per minute. 2. Conductors — ] From generators to switchboards, rheostats, or other instruments, and jthence to outside lines. ! a. Must be in plain sight or readily accessible. b. Must have an approved insulating covering as called for by rules in Class "0" for similar work, except that in central stations, on exposed circuits, the wire which is used must have a heavy braided non-combustible outer covering. Bus bars may be made of bare metal. c. Must be kept so rigidly in place that they cannot come in contact. d. Must in all other respects be installed under the same precautions as required by rules in Class " C " for wires carrying a current of the same volume and potential. 3. $w*tcnt>oards — a. Must be so placed as to reduce to a minimum the danger of communi- cating fire to adjacent combustible material. Special attention is called to the fact that switchboards should not be built down to the floor, nor up to the ceiling, but a space of at least ten or twelve inches should be left between the floor and the board, and from eighteen to twenty-four inches between the ceiling and the board in order to prevent fire from communicating from the switchboard to the floor or ceiling, and also to prevent the forming of a partially concealed space very liable to be used for storage of rubbish and oily waste. b. Must be made of non-combustible material or of hardwood in skeleton form filled to prevent absorption of moisture. c. Must be accessible from all sides when the connections are on the back, but may be placed against a brick or stone wall when the wiring is entirely on the face. d. Must be kept free from moisture. e. Bus bars must be equipped in accordance with rules for placing conductors. 4. Resistance Boxes and Equalizers — {For construction rules, see No. 60.) a. Must be placed on a switchboard or, if not thereon, at a distance of a a foot from combustible material, or separated therefrom by a non-inflam- mable, non-absorptive3 insulating material. 764 NATIONAL ELECTRICAL CODE. 5. Lightning- Arresters — {For construction rules see No, 63.) a. Must be attached to each side of every overhead circuit connected with the station. It is recommended to all electric lightand power companies that arresters be connected at intervals over systems in such numbers and so located as to pi event ordinary discharges entering (over the wires) buildings connected to the lines. b. Must be located in readily accessible places away from combustible materials, and as near as practicable to the point where the wires enter the building. Station arresters should generally be placed in plain sight on the switch board. In all cases, kinks, coils, and sharp bends in the wires between the airesters and the outdoor lines must be avoided as far as possible. c. Must be connected with a thoroughly good and permanent ground con- nection by metallic strips or wires having a conductivity not less than that of a No. 6 B. & S. copper wire, which must be run as nearly in a straight line as possible from the arresters to the earth connection. Ground wires for lightning arresters must not be attached to gas-pipes within the buildings. It is often desirable to introduce a choke coil in circuit between the arresters and the dynamo. In no case should the ground wire from lightning arrester be put into iron pipes, as these Avould tend to impede the discharge. G. Care and Attendance. a. A competent man must be kept on duty where generators are operating. b. Oily waste must be kept in approved metal cans and removed daily. Approved waste cans shall be made of metal, with legs raising can three inches from the floor, and with self-closing covers. 1. Testing" of Insulation Resistance. a. All circuits, except such as are permanently grounded in accordance with Rule 13 A, must be provided with reliable ground detectors. Detectors which indicate continuously, and give an instant and permanent indication of a ground, are preferable. Ground wires from detectors must nut be attached to gas-pipes within the building. b. Where continuously indicating detectors are not feasible, the circuits should be tested at least once per day, and preferably oftener. c. Data obtained from all tests must be preserved for examination by tli Inspection Deptrtment having jurisdiction. These rules on testing to be applied at such places as may be designated by the Inspection Department having jurisdiction. H. Motors — a Must be insulated on floors or base frames, which must be kept filled to prevent absorption of moisture ; and must be kept clean and dry. Where frame insulation is impracticable the Inspection Department having juris- diction may, in writing, permit its omission, in which case the frame must be permanently and effectively grounded. A high-potential machine which, on account of great weight or for other reasons, cannot have its frame insulated, should be surrounded with an insulated platform. This may be made of wood mounted on insulating supports, and so arranged that a man must stand upon it in order to touch any part of the machine. In case of a machine having an insulated frame, if there is trouble from static electricitv due to belt friction, it should be overcome by placing near the belt a metallic comb connected to the earth, or by grounding the frame through a very high resistance of not less than 200 ohms per volt generated by the machine. b. Must be wired under the same precautions as required by rules in class " C," for wires carrying a current of the same volume and potential. The leads or branch circuits should be designed to carry a current at least iifty per cent greater than that required by the rated capacity of the motor ■BOHH^BIB^ CLASS A. STATIONS AND DYNAMO ROOMS. 765 to provide for the inevitable overloading of the motor at times without overf using the wires. c. The motor and resistance box must be protected by a cutout and con- trolled by a switch (see No. 17 a), said switch plainly indicating whether "on" or "off." Where one-fourth horse-power or less is used on low- tension circuits a single-pole switch will be accepted. The switch and rheostat must be located within sight of the motor, except in such cases where special permission to locate them elsewhere is given in writing by the Inspection Department having jurisdiction. d. Must have their rheostats or starting-boxes located as to conform to the requirements of No. 4. In connection with motors the use of circuit-breakers, automatic start- ing-boxes and automatic under-load switches is recommended, and they must be used when required. e. Must not be run in series-multiple or multiple-series, except on con- stant-potential systems, and then only by special permission of the Inspec- tion Department having jurisdiction. /. Must be covered with a waterproof cover when not in use, and, if deemed necessary by the Inspection Department having jurisdiction, must be inclosed in an approved case. From the nature of the question the decision as to what is an approved case must be left to the Inspection Department having jurisdiction to de- * "mine in each instance. /. Must, when combined with ceiling fans, be hung from insulated hooks, or else there must be an insulator interposed between the motor and its support. h. Must each be provided with a name-plate, giving the maker's name, the capacity in volts and amperes, and the normal speed in revolutions per minute. ©. Maihvaj Power Plants*. i. Must be equipped in each feed wire before it leaves the station with an approved automatic circuit-breaker (see No. 52) or other device, which will immediately cut oft the current in case of an accidental ground. This device must be mounted on a fireproof base, and in full view and reach of the attendant. 1©. Storage or Primary Hatteries. (. When current for light and power is taken from primary or secondary batteries, the same general regulations must be observed as applied to similar apparatus fed from dynamo generators developing the same differ- ence of potential. b. Storage battery rooms must be thoroughly ventilated. c. Special attention is directed to the rules for rooms where acid fumes exist (see No. 24, j and k). d. All secondary batteries must be mounted on non-absorptive, non- combustible insulators, such as glass or thoroughly vitritied and glazed porcelain. e. The use of any metal liable to corrosion must be avoided in cell con- nections of secondary batteries. 11. Transformers. {For construction rules, see No. 62.) i. In central or substations the transformers must be so placed that . smoke from the burning out of the coils or the boiling over of the oil (where oil-filled cases are used) could do no harm. CliJLSS B. — ©XJXSII9E WORK. ALL, SYSTEMS ASD VOLTAGES. 13. Wires. t,. Service wires must have an approved rubber insulating covering (see No. 41). Line wires, other than services, must have an approved weather- proof, or rubber insulating covering (Nos. 41 and 44). All the wires must have an insulation equal to that of the conductors they confine. 7C6 NATIONAL ELECTRICAL CODE, b. Must be so placed tliat moisture caunot form a cross connection be- tween tliem, not less than a foot apart, and not in contact with any sub- stance other than their insulating supports. Service blocks must be covered over their entire surface with at least two coats of waterprool paint. c Must be at least seven feet above the highest point of flat roofs, and at least one foot above the ridge of pitched roofs over which they pass or to which they are attached. . il Must be protected by dead insulated guard iron or wires from pos- sibility of contact with other conducting wires or substances to which cur- rent may leak. Special precautions of this kind must be taken where sharp angles occur, or where any wires might possibly come in contact with electric light or power wires. . e. Must be provided with petticoat insulators of glass or porcelain. For- celain knobs or cleats and rubber hooks will not be approved. f Must be so spliced or joined as to be both mechanically and electri- cs il'lv secure without solder. The joints must then be soldered, to insure , preservation, and covered with an insulation equal to that on the con- AlMomts must be soldered, even if made with some form of patent spli- cing device. This ruling applies to joints and splices in all classes of wiring covered by these rules. , , . , . , , .. a Must, where they enter buildings, have drip loops outside, and the holes through which the conductors must be bushed with non-conibustible, non-absorptive insulating tubes slanting upward toward the inside. h Telegraph, telephone, and similar wires must not be placed on the same cross-arm with electric light or power wires ; and when placed on the same pole with such wires the distance between the two inside pins of each cross-arm must not be less than twenty-six inches. i. The metallic sheaths to cables must be permanently and effectively connected to " earth." TROLLEY WIRES. j. Must not be smaller than No. 0 B. & S. copper or No. 4 B. & S. silicon bronze, and must readily stand the strain put upon them when muse. le Must have a double insulation from the ground. In wooden-pole con- struction the pole will be considered as one insulation. I Must be capable of being disconnected at the power plant, or of being divided into sections, so that, in case of fire on the railway route, .he cur- rent may be shut off from the particular section and not interfere with the work of 'tin* firemen. This rule also applies to feeders. m. Must be safely protected against accidental contact where crossed by other conductors. , , , , , , , , . Guard wires should be insulated from the ground, and should be electric- ally disconnected in sections of not more than 300 feet in length. GROUND RETURN WIRES. n. For the diminution of electrolytic corrosion of underground metal g work, ground return wires must be so arranged that the diilerence oi ' potential between the grounded dynamo terminal and any point on the return circuit will not exceed twenty -five volts. J It is suggested that the positive pole of the dynamo be connected t<> tin trolly line,' and that whenever pipes or other underground metal work an found to be electrically positive to the rails or surrounding earth, that thej be connected by conductors arranged so as to prevent as far as possible current flow from the pipes into the ground. 13. Transformers — (For construction rules, see No. 62.) a. Must not be placed inside of any building, excepting central station!- unless by special permission of the Inspection Department having juris T^Must not be attached to the outside walls of buildings, unless ser arated therefrom by substantial supports. CLASS B. OUTSIDE WORK. 767 13. A. Grounding- JLow Potential Circuits. The grounding of low potential circuits under the following regulations is only allowed when so arranged that under normal conditions there will be no flow of current through the ground wire. Direct Current 3 -Wire Systems. a. Neutral wire may be grounded, and when grounded the following rules must be complied with : — 1. Must be grounded at the Central Station on a metal plate buried in coke beneath permanent moisture level, and also through all available underground water- and gas-pipe systems. 2. In underground systems the neutral wire must also be grounded at each distributing-box through the box. 3. In overhead systems the neutral wire must be grounded every 500 feets as provided in Sections c, e, and/. The Inspection Department having jurisdiction may require grounding if they deem it necessary. Two-wire direct current systems having no accessible neutral point are not to be grounded. Alternating' Current Secondary Systems. b. The neutral point of transformers, or the neutral wire of distributing systems, may be grounded, and when grounded the following rules must be complied with : — 1. Transformers feeding 2-wire systems must be grounded at the center of the secondary coils. 2. Transformers feeding systems with a neutral wire must have the neutral wire grounded at the trausformer and at least every 250 feel beyond. Inspection Department having jurisdiction may require grounding if they deem it necessary. Ground Connections. c. The ground wire in D. C. 3-wire systems must not at Central Stations be smaller than the neutral wire and not smaller than No. 6 B. & S. else- where. d. The ground wire in A. C. systems must never be less than No. 6 B. & S., and must always have equal carrying capacity to the secondary lead of the transformer, or the combined leads where transformers are banked. e. The ground wire must be kept outside of buildings, but may be di- rectly attached to the building or pole. The wire must be carried in as nearly a straight line as possible, and kinks, coils and sharp bends must be avoided. f. The ground connections for Central Stations, transformer sub- stations, and banks of transformers must be made through metal plates buried in coke below permanent moisture level, and connections should also be made to all available underground piping systems. For individual transformers and building services the ground connection may be made as above, or may be made to water or other piping systems running into the buildings. This connection may be made by carrying the ground wire into the cellar and connecting on the street side of meters, main clocks, etc._ In connecting ground wires to piping systems, where possible the wires should be soldered into one or more brass plugs and the plugs forcibly screwed into a pipe-fitting, or where the pipes are cast iron into a hole tapped to the pipe itself. For large stations, where connecting to under- ground pipes with bell and spigot joints, it is well to connect to several lengths, as the pipe joints may be of rather high resistance. Where such plugs cannot be used the surface of the pipe may be filed or scraped bright. the wire wound around it, and a strong clamp put over the wire and firmly bolted together. Where ground plates are used a No. 16 copperplate, about 3 x 6 feet in size, with about two feet of crushed coke or charcoal about pea size both under and over it, would make aground of sufficient capacity for a mod- erate size station, and would probably answer for the ordinary sub-station 768 NATIONAL ELECTRICAL CODE. or bank of transformers. For a large Central Station considerable more area might be necessary, depending upon the other unground connections available. The ground wire should be riveted to such a plate in a number of places, and soldered for its whole length. Perhaps even better than a copperplate is a cast-iron plate with projecting forks, the idea of the fork being to distribute the connection to the ground over a fairly broad area, and to give a large surface contact. The ground wire can probably best be connected to such a cast-iron plate by brass plugs screwed into the plate to which the wire is soldered. In all cases the joint between the plate and the ground wire should be thoroughly protected against corrosion by suit- able painting with waterproof paint or some equivalent. CLASS C. — 1\SII)E WORK. ALL SYSTEMS AND VOLTAGES. « EX 12 It A JL RELE§ — ALL lYiTKIUS VX I> VOLTAGES. 14. Wires. (For special rules, See Nos. 18, 24, 32, 38, and 39.) a. Must not be of smaller size than No. 14 B. & S., except as allowed under Rules 24 t and 45 b, b. Tie wires must have an insulation equal to that of the conductors they confine. c. Must be so spliced or joined as to both mechanically and electrically secure without solder ; they must be then soldered to insure preservation, and the joint covered with an insulation jqual to that on the conductors. Standard wires must be soldered before being fastened under clamps or binding screws ; and, when they have a conductivity greater than No. 10 B. & S. copper wire, they will be soldered into lugs. All joints must be soldered, even if made with some form of patent splicing device. This ruling applies to joints and splices in all classes of wiring covered by these rules. d. Must be separated from contact with walls, floors, timbers, or parti- tions through which they may pass by non-combustible, non-absorptive insulating tubes, such as glass or porcelain. Bushings must be long enough to bush the entire length of the hole in one continuous piece, or else the hole must first be bushed by a continuous waterproof tube, which may be a conductor, such as iron pipe ; the tube then is to have a non-conducting bushing pushed in at each end so as to keep the wire absolutely out of contact with the conducting pipe. e. Must be kept free from contact with gas, water, or other metallic piping, or any other conductors or conducting material which they may cross, by some continuous and firmly fixed non-conductor, creating a sepa- ration of at least one inch. Deviations from this rule may sometimes be allowed by special permission. /. Must be so placed in wet places that an air space will be left between conductors and pipes in crossing, and the former must be run in such a way that they cannot come in contact with the pipe accidentally. Wires should be run over, rather than under, pipes upon which moisture is likely to gather or Avhich, by leaking, might cause trouble on a circuit. 15. Underground Conductors — a. Must be protected, when brought into a building, against moisture and mechanical injury, and all combustible material must be kept removed from the immediate vicinity. b. Must not be so arranged as to shunt the current through a building around any catch-box. 1©. Table Carrying- Capacity of Wires. Below is a table which must be followed in placing interior conductors, showing the allowable carrying capacity of wires and cables of ninety-eight per cent conductivity, according to the standard adopted by the American Institute of Electrical Engineers. -INSIDE WORK. 769 Table A. Table B. Table A. Table B. 6 Rubber- Weather- Rubber- Weather- Cfi Covered proof Covered proof Wires. Wires. Circular Circular Wires. Wires. ^3 See No. 41. See No. Mills. Mills. See No. 41. See No. W 42 to 44. 42 to 44. Amperes. Amperes. Amperes. Amperes. 18 3 5 1,624 200,000 200 300 16 6 8 2,583 300,000 270 400 14 12 16 4,107 400,000 330 500 12 17 23 6,530 500,000 390 590 10 24 32 10,380 600,000 450 680 8 33 46 16,510 700,000 500 760 6 46 65 26,250 800,000 550 840 5 54 77 33,100 900,000 600 920 4 65 92 41,740 1,000,000 650 1,000 3 76 110 52,630 1,100,000 690 1,080 2 90 131 66,370 1,200,000 730 1,150 1 107 156 83,690 1,300,000 770 1,220 0 127 185 105,500 1,400,000 810 1,290 00 150 220 133,100 1,500 000 850 1,360 000 177 262 167,800 1,600,000 890 1,430 0000 210 312 211,600 1,700,009 1.800,000 1,900.000 2,000,000 930 970 1,010 1,050 1,490 1,550 1,610 1,670 The lower limit is specified for rubber-covered wires to prevent gradual deterioration of the high insulations by the heat of the wires, but not from fear of igniting the insulation. The question of drop is not taken into con- sideration in the above tables. The carrying capacity of sixteen and eighteen wire is given, but no smaller than fourteen is to be used, except as allowed under Rules 2it and 45 b. Iff. Switches. Cutouts, Circuit-Breakers, etc. — (For construction rules, see Nos. 51, 52, and 53.) a. Must, whenever called for, unless otherwise provided (for exceptions, see No. 8 c and No. 22 c), be so arranged that the cutouts will protect, and the opening of the switch or circuit-breaker will disconnect, all of the wires ; that is, in a two-wire system the two wires, and in a three-wire • system the three wires, must be protected by the cutout, and disconnected by the operation of the switch or circuit-breaker. b. Must not be placed in the immediate vicinity of easily ignitible stuff or where exposed to inflammable gases or dust or to flyings of combustible material. c. Must, when exposed to dampness, either be inclosed in a waterproof box or mounted on porcelain knobs. COMSTA]¥T CURKEIT SYSTEMS. PRINCIPALLY SERIES ARC LIGHTING. IS. Wires — (See also ATos. 14, 15, and 16.) a. Must have an approved rubber insulating covering (see No. 41). b. Must be arranged to enter and leave the building through an approved double-contact service switch (see No. 51), mounted in a non-combustible case, kept free from moisture, and easy of access to police or firemen. So- called " snap switches " must not be used on high-potential circuits. 770 NATIONAL ELECTRICAL CODE. c. Must always be in plain sight, and never incased, except when required by the Inspection Department having jurisdiction. d. Must be supported on glass or porcelain insulators, which separate the wire at least one inch from the surface wired over, and must be kept rigidly at least eight incbes from each other, except within the structure of lamps, on hanger-boards, in cutout boxes, or like places, where a less distance is necessary. e. Must, on side walls, be protected from mechanical injury by a sub- stantial boxing, retaining an air space of one inch around tbe conductors, closed at the top (the wires passing through busbed holes), and extending not less tban seven feet from the floor. When crossing floor-timbers in cellars or in rooms, where they might be exposed to injury, wires must be attached by their insulating supports to the underside of a wooden strip not less than one-half an inch in thickness. lO. Arc Lampo — {For construction rules, see No. 57.) a. Must be carefully isolated from inflammable material. b. Must be provided at all times with a glass globe surrounding the arc, securely fastened upon a closed base. No broken or cracked globes to be used. c. Must be provided with a wire netting (having a mesh not exceeding one and one-fourth inches) around the globe, and an approved spark arrester (see No. 58), when readily inflammable material is in the vicinity of the lamps, to prevent escape of sparks, melted copper or carbon. It is recom- mended that plain carbons, not copper-plated, be used for lamps in such places. Arc lamps, when used in places where they are exposed to flyings of easily inflammable material, should have the carbons inclosed completely in a globe in such manner as to avoid the necessity for spark arresters. For the present, globe and spark arresters will not be required on so- called " inverted arc" lamps, but this type of lamp must not be used where exposed to flyings of easily inflammable materials. d. "Where hanger-boards (see No. 56) are not used, lamps must be hung from insulating supports other than their conductors. 30. Incandescent Stamps in Series Circuits — a. Must have the conductors installed as provided in No. 18, and each lamp must be provided with an automatic cutout. b. Must have each lamp suspended from a hanger-board by means of rigid tube. c. No electro-magnetic device for switches and no system of multiple- series or series-multiple lighting will be approved. d. Under no circumstances can they be attached to gas fixtures. COHSTAJtfT potential systems. GENERAL RULES, ALL VOLTAGES. 81. Automatic Cutouts (Fuses and Circuit-Breakers). (See No. 17, and for construction Nos. 52 and 53.) a. Must be placed on all service wires, either overhead or underground, as near as possible to the point where they enter the building and inside the walls, and arranged to cut off the entire current from the building. Where the switch required by rule No. 22 is inside the building, the cut- out required by this section must be placed so as to protect it. b. Must be placed at every point where a change is made in the size of wire [unless the cutout in the larger wire will protect the smaller (see No. 16)]. c. Must be in plain sight, or inclosed in an approved box (see No. 54) and readily accessible. They must not be placed in the canopies or shells of fixtures. CONSTANT POTENTIAL SYSTEMS. 771 d . Must be so placed that bo set of incandescent lamps, whether grouped on one fixture or several fixtures or pendants, requiring more than 660 watts, shall be dependent upon one cutout. Special permission may be given in writing by the Inspection Department having jurisdiction for departure from this rule in case of large chandeliers, stage borders, and illuminated signs. e. Must be provided with fuses, the rated capacity of which does not exceed the allowable carrying capacity of the wire ; and, when circuit- breakers are used, they must not be set more than about thirty per cent above the allowable carrying capacity of the wire, unless a fusible cutout is also installed in the circuit (see No. 16). 22. Switches — (See No. 17, and for construction No. 51.) a. Must be placed on all service wires, either overhead or underground, in a readily accessible place, as near as possible to the point where the wires enter the building, and arranged to cut off the entire current. 6. Must always be placed in dry, accessible places, and be grouped as far as possible. Knife switches must be so placed that gravity will tend to open rather than close the switch. c. Must not be single -pole, except when the circuits which they control supply not more than six 16-candle power lamps or their equivalent. d. Where flush-switches are used, whether with conduit systems or not, the switches must be inclosed in boxes constructed of or lined with fire- resisting material. No push-buttons for bells, gas-lighting circuits or the like shall be placed in the same wall-plate with switches controlling elec- tric light or power wiring. 23. Electric Heaters — a. Must, if stationary, be placed in a safe situation, isolated from inflam- mable materials, and be treated as sources of heat. b. Must each have a cutout and indicating -switch, (see No. 17 a). c. Must have the attachments of feed wires to the heaters in plain sight, easily accessible, and protected from interference, accidental or otherwise. d. The flexible conductors for portable apparatus, such as irons, etc., must have an approved insulating covering (see No. 45 h). e. Must each be provided with name-plate, giving the maker's name and the normal capacity in volts and amperes. EOW POTEUJTIAE SYSTEMS. 550 VOLTS OR LESS. Any circuit attached to any machine, or combination of machines, which develops a difference of potential, between any tivo wires, of over ten volts and less than 550 volts, shall be considered as a low-potential circuit, and as coming under this class, unless an approved transform- ing device is used, which cuts the difference of potential down to ten volts or less. The primary circuit not to exceed a potential of 3,500 volts. 24. Wires — GEXERAL RULES. (See also Nos. 14, 15, and 16.) jt. Must not be laid in plaster, cement, or similar finish. b. Must never be fastened with staples. c. Must not be fished for any great distance, and only in places where the inspector can satisfy himself that the rules have been complied with. d Twin wires must never be used, except in conduits, or where flexible conductors are necessary. c. Must be protected on side walls from mechanical injury. When cross- ing floor-timbers in cellars or in rooms, where they might be exposed to injury, wires must be attached by their insulating supports to the under side of a wooden strip, not less than one-half inch in thickness, and not less than three inches in width. 772 NATIONAL ELECTRICAL CODE. Suitable protection on side walls may be secured by a substantial boxing, retaining an air space of one inch around the conductor, closed at the top (the wires passing through bushed holes), and extending not less than live feet from the floor ; or by an iron-armored or metal-sheathed insulating conduit sufficiently strong to withstand the strain it will be subjected to ; or plain metal pipe, lined with insulating tubing which must extend one- half inch beyond the end of the metal tube. The pipe must extend not less than five feet above the floor, and may extend through the floor in place of a floor bushing. If iron pipes are used with alternating currents, the two or more wires of a circuit must be placed in the same conduit. In this case the insulation of each wire must be reinforced by a tough conduit tubing projecting beyond the ends of the iron pipe at least two inches. /. When run immediately under roofs, or in proximity to water tanks or pipes, will be considered as exposed to moisture. SPECIAL RULES. For open work : In dry places : g. Must have an approved rubber or " slow-burning" waterproof insula- tion (see Nos. 41 and 42). h. Must be rigidly supported on non-combustible, non-absorptive insula- tors, which separate the wires from each other and from the surface wired over in accordance with following table : VOLTAGE. DISTANCE FROM SURFACE. DISTANCE BETWEEN WIRES. 0 to 225 I inch. 1\ inches. 225 "550 1 " 4 " Rigid supporting requires under ordinary conditions, where wiringalong flat surfaces, supports at least every four and one-half feet. If the wires are liable to be disturbed, the distance between supports should be shortened. In buildings of mill construction, mains of No. 8 B. & S. wire or over, where not liable to be disturbed, may be separated about four inches, and run from timber to timber, not breaking around, and may be supported at each timber only. This rule will not be interpreted to forbid the placing of the neutral of a three-wire system in the center of a three-wire cleat, provided the outside wires are separated in accordance with above table. In damp places, such as Breweries, Sugar Houses, Packing Houses, Stables, Dye Houses, Paper or Pulp Mills, or buildings specially liable to moisture, or acid, or other fumes liable to injure the wires or their insu- lation, except tvhere used for pendants : i. Must have an approved rubber insulating covering (see No. 41). j. Must be rigidly supported on non-combustible, non-absorptive in; -t. la- tors, which separate the wire at least one inch from the surface wired over, and they must be kept apart at least two and one-halt inches. Rigid supporting requires under ordinary conditions, where wiring over flat surfaces, supports at least every four and one-half feet. If the wires are liable to be disturbed, the distance between supports should be shortened. In buildings of mill construction, mains of No. 8 B. & S. wire or over, where not liable to be disturbed, may bf> separated about four inches, and run from timber to timber, not breaking around, and may be supported at each timber only. k. Must have no joints or splices. JFor molding- work : I. Must have approved rubber insulation covering (see No. 41). m. Must never be placed in molding in concealed or damp places. Cor conduit work : n. Must have an approved rubber insulating covering (see No. 47). o. Must not be drawn in until all mechanical work on the building has been, as far as possible, completed. LOW POTENTIAL SYSTEMS. 773 p. Must, for alternating systems, have the two or more wires of a circuit drawn in the same conduit. It is advised that this be done for direct-current systems also, so that they may be changed to alternating systems at any time, induction troubles preventing such a change unless this construction is followed. Tor concealed 4i knob and tube " work: q. Must have an approved rubber insulating covering (see No. 41). r. Must be rigidly supported on non-combustible, non-absorptive insula- tors which separate the Avire at least one inch from the service wired over, and must be kept at least ten inches apart, and, when possible, should be run singly on separate timbers or studding. Kigid supporting requires under ordinary conditions, where wiring along flat surfaces, supports at least every four and one-half feet. If the wires are liable to be disturbed, the distance between supports shoxild be shortened. s. When, from the nature of the case, it is impossible to place concealed wiring on non-conbustible, insulating supports of glass or porcelain, an ap- proved armored cable with single or twin conductors (see No. 48) may be used where the difference of potential between wires is not over 300 volts, provided it is installed without joints between outlets, and the cable armor properly enters all fittings and is rigidly secured in place ; or, if the differ- ence of potential between wires is not over 300 volts, and if wires are not exposed to moisture, they may be fished on the loop system if separately incased throughout in approved flexible tubing or conduits. Tor fixture work : t. Must have an approved rubber insulating covering (see No. 46), and shall not be less in size than No. 18 B. & S. u. Supply conductors, and especially the splices to fixtures wires, must be kept clear of the grounded part of gas-pipes ; and, where shells are used, the latter must be constructed in a manner affording sufficient area to allow this requirement. r. Must, Avhen fixtures are wired outside, be so secured as not to be cut or abraded by the pressure of the fastenings or motion of the fixture. 35. Interior Conduits. (See also Nos. 24 n top, and 49.) The object of a tube or conduit is to facilitate the insertion or extraction of the conductors to protect them from mechanical injury and, as far as possible, from moisture. Tubes or conduits are to be considered merely as raceways, and are not to be relied upon for insulation between wire and wire, or between the wire and the ground. a. No conduit tube having an internal diameter of less than five-eights of an inch shall be used. (If conduit is lined, measurement to be taken inside of lining.) b. Must be continuous from one junction box to another or to fixtures, and the conduit tube must properly enter all fittings. c. Must be first installed as a complete conduit system, without the con- ductors. d. Must be equipped at every outlet with an approved outlet box. e. Metal conduits, where they enter junction boxes, and at all other out- lets, etc., must be fitted with a capping of approved insulating material, fitted so as to protect wire from abrasion. /. Must have the metal of the conduit permanently and effectively grounded. SO. Fixtures — (See also No. 24 t to v.) a. Must, when supported from the gas-piping of a building, be insulated from the gas-pipe system by means of approved insulating joints (see No. 59) placed as close as possible to the ceiling. It is recommended that the gas outlet pipe be protected above the insulat- ing joint by a non-combustible, non-absorptive insulating tube, having a flange at the lower end where it comes in contact with the insulating joint ; 774 NATIONAL ELECTRICAL CODE. and that, where outlet tubes are used, they be of sufficient length to extend below the insulating joint, and that they be so secured that they will not be pushed back when the canopy is put in place. Where iron ceilings are used, care must be taken to see that the canopy is thoroughly and perma- nently insulated from the ceiling. b. Must have all burs, or tins, removed before the conductors are drawn into the fixture. c. The tendency to condensation within the pipes should be guarded against by sealing the upper end of the fixture. d. No combination fixture in which the conductors are concealed in a space less than one-fourth inch between the inside pipe and the outside casing will be approved. e. Must be tested for " contacts " between conductors and fixture, for " short circuits," and for ground connections before it is connected to its supply conductors. /. Ceiling blocks for fixtures should be made of insulating material ; if not the wires in passing through the plate must be surrounded with nun- combustible non-absorptive, insulating material, such as glass or porcelain. g. Under no conditions shall there be a difference of potential of more than 300 volts between wires contained in or attached to the same fixture. 2?. Sockets. (For construction rules, see No. 55.) a. In rooms where inflammable gases may exist the incandescent lamp and socket must be inclosed in a vapor-tight globe, and supported on a pipe hanger, wired with approved rubber-covered wire (see No. 41) soldered directly to the circuit. b. In damp or wet places, or over specially inflammable stuff, waterproof sockets must be used. When waterproof sockets are used, they should be hung by separate stranded rubber-covered wires, not smaller than No. 14 B. & S., which should preferably be twisted together when the drop is over three feet. These wires should be soldered direct to the circuit wires, but supported independently of them. 28. Flexible Cord — a. Must have an approved, insulation and covering (see No. 45). b. Must not be used where the difference of potential between the two wires is over 300 volts. c. Must not be used as a support for clusters. d. Must not be used except for pendants, wiring of fixtures, and port- able lamps or motors. e. Must not be used in show windows. /. Must be protected by insulating bushings where the cord enters the socket. g. Must be so suspended that the entire weight of the socket and lamp will be born by knots under the bushing in the socket, and above the point where the cord comes through the ceiling-block or rosette, in order that the strain may be taken from the joints and binding screws. 20. Arc trig-lit* on tow-Potential Circuits — a. Must have a cutout (see No. 17a) for each lamp of each series of lamps. The branch conductors should have a carrying capacity about fifty per cent in excess of the normal current required by the lamp to provide for heavy current required when lamp is started or when carbons become stuck without overfusing the wires. b. Must only be furnished with such resistances or regulators as are in- closed in non-combustible material, such resistances being treated as sources of heat. Incandescent lamps must not be used for resistance de- vices. c Must be supplied witn globes and protected by spark arresters and wire netting around globe, as in the case of arc lights on high-potential circuits (see Nos. 19 and 58). LOW POTENTIAL SYSTEMS. 775 30. Economy Coils. a. Economy and compensator coils for arc lamps must be mounted on non-conbustible, non-absorptive insulating supports, such as glass or porce- lain, allowing an air space of at least one inch between frame and support, and in general to be treated like sources of beat. 31. Decorative Series JLaiiips. a. Incandescent lamps run in series sball not be used for decorative pur- poses inside of buildings, except by special permission in writing from tbe Inspection Department baving jurisdiction. 33. Car- Wiring- — a. Must be always run out of reach of tbe passengers, and must have an approved rubber-insulating covering (see No. 41). 33. Car-Houses — a. Must have the trolley wires securely supported on insulating hangers. 6.' Must have the trolley hangers placed at such distance apart that, in case of a break in tbe trolley wire, contact cannot be made with the floor. c. Must have cutout switch located at a proper place outside of the building, so that all trolley circuits in tbe building can be cut out at one point and line circuit-breakers must be installed, so that when this cutout switch is open the trolley wire will be dead at all points within 100 feet of the building. The current must be cut out of the building whenever the same is not in use or the road not in operation. d. Must have all lamps and stationary motors installed in such a way that one main switch can control the whole of each installation — lighting or power— independently of main feeder-switch. No portable incandes- cent lamps or twin wire allowed, except that portable incandescent lamps may be used in the pits, connections to be made by two approved rubber- covered flexible wires (see No. 41), properly protected against mechanical injury ; the circuit to be controlled by a switch placed outside of the pit. e. Must have all wiring and apparatus installed in accordance with rules under Class " C " for constant potential systems. /. Must not have any system of feeder distribution centering in the building. g. Must have the rails bonded at each joint with no less than No. 2 B. & S. annealed copper wire, also a supplementary wire to be run for each track. h. Must not have cars left with trolley in electrical connection with the trolley wire. 34. Ug-fitiiis* and Power from Railway "Wires — a. Must not be permitted, under any pretense, in the same circuit with trolley wires with a ground return, except in electric railway cars, electric car houses and their power stations ; nor shall the same dynamo be used for both purposes. HICJKf-T»©TE]¥TIAE SYSTEMS. 550 TO 3,500 Volts. Any circuit attached to any machine, or combination of machines, which de- velops a difference of potential , between any two wires, of over 300 volts and less than 3,500 volts, shall be considered as a high-potential cir- tuit, and as coming under that class, unless an approved transforming device is used, which cuts the difference of potential down to 300 volts or less. 35. Wires — {See also Nos. 14, 15, and 16.) a. Must have an approved rubber-insulating covering (see No. 41). 6. Must be always in plain sight and never incased, except Avhere re- quired by the Inspection Department having jurisdiction 776 NATIONAL ELECTRICAL CODE, c. Must be rigidly supported on glass or porcelain insulators, which raise the wire at least one inch from the surface wired over, and must be kept apart at least four inches for voltages up to 750 and at least eight inches for voltages over 750. Rigid supporting requires under ordinary conditions, where wiring along flat surfaces, supports at least about every four and one-half feet. If the wires are unusually liable to be disturbed, the distance between supports should be shortened. In buildings of mill construction, mains of No. 8 B. & S. wire or over, where not liable to be disturbed, may be separated about six inches for voltages up to 750 and about ten inches f or voltages above 750 ; and run from timber to timber, not breaking around, and may be supported at each timber only. d. Must be protected on side walls from mechanical injury by a substan- tial boxing, retaining an air space of one inch around the conductors, closed at the top (the wires passing through bushed holes) and extending not less than seven feet from the floor. When crossing floor-timbers, in cellars or in rooms, where they might be exposed to injury, wires must be attached by their insulating supports to the under side of a wooden strip not less than one-half an inch in tbickness. 3G. Xraiasformei-s (when permitted inside buildings, see No. 13) — (For construction rules, see No. 62.) a. Must be located at a point as near as possible to that at which the primary wires enter the building. b. Must be placed in an inclosure constructed of or lined with fire- resisting material : the inclosure to be used only for this purpose, and to be kept securely locked, and access to the same allowed only to responsible persons. c. Must be effectually insulated from the ground, and the inclosure Li which they are placed must be practically air-tight, except that it shall be thoroughly ventilated to the outdoor air, if possible, through a chimney or flue. There should be at least six inches air space on all sides of the trans- former. 31. Series lamps. a. No system of multiple-series or series-multiple for light or power will be approved. b. Under no circumstances can lamps be attached to gas fixtures. EXTRA HlftH 1POTE1* TIJUL SYSTJEMS. Ovek 3,500 Volts. Ami circuit attached to any machine or combination of machines , which de- velops a difference of potential, between any two wires, of over 3,500 volts shall be considered as an extra high-potential circuit, and as cominq under that class, unless an approved transforming device is used, which cuts the difference of p>otential down to 3,500 volts or less. 3S. Primary Wires — a. Must not be brought into or over building, except power and sub- stations. ___. 30. Secondary Wires — a Must be installed under rules for high-potential systems, when their immediate primary wires carry a current of oyer 3,500 volts, unless the primary wires are entirely underground, within city and village limits. The presence of wires carrying a current with a potential of over 3,500 volts in the streets of cities, towns, and villages is considered to increase the. fire hazard. Extra high potential circuits are also objectionable in any location where telephone, telegraph, and similar circuits run in proximity to them. As the underwriters have no jurisdiction over streets and roads they can only take this indirect way of discouraging such systems ; but fur- ther, it is strongly urged that municipal authorities absolutely refuse to grant any franchise for right of way for overhead wires carrying a current of extra high potential through streets or roads which are used to any great extent for public travel or for trunk-line, telephone, or telegraph circuits. CLASS D. FITTINGS, MATERIALS, AND DETAILS. 777 CJ.JLSS ». JTITTI^OS, MATEKIAIS, .AJtfl* DETAILS o*1 coj*s>xucTioar. All Systems and Voltages. Insulated Wires — Rules 40 to 48. 40. General Rules. a. Copper for insulated conductors must never vary in diameter so as to be more than two one- thousandths of an inch less than the specihed size. b. Wires and cables of all kinds designed to meet the following speciiica tions must be plainly tanked or marked as follows : 1. The maximum voltage at which the wire is designed to be used. 2. The words " National Electrical Code Standard." 3. Name of the manufacturing company, and, if desired, trade-name of the wire. 4. Month and year when manufactured. 41. Bnblier-Covered. a. Copper for conductors must be thoroughly tinned. Insulation for voltag-es between © and GOO. b. Must be of rubber or other approved substance, and be of a thickness not less than that given in the following table for B. & S. gauge sizes : rom 18 to 16, inclusive, ^ 14 to 8, " B34 7 to " 1 to 0000, " BV " 0000 to 500,000, c. m. gy " 500,000 to 1,000,000, " BV Larger than 1,000,000, " i' Measurements of insulating wall are to be made at the thinnest portion of the dielectric. c. The completed coverings must show an insulation resistance of at least 100 megohms per mile during thirty days' immersion in water at seventy degrees Fahrenheit. d. Each foot of the completed covering must show a dieletric strength sufficient to resist throughout five minutes the application of an electro- motive force of 3,000 volts per one-sixty-fourth of an inch thickness of in- sulation under the following conditions : The source of alternating electro-motive force shall be a transformer of at least one kilowatt capacity. The application of the electro-motive force shall first be made at 4,000' volts for five minutes, and then the voltage in- creased by steps of not over 3,000 volts, each held for five minutes, until the rupture of the insulation occurs. The tests for dielectric strength shall be made on a sample of wire which has been immersed for seventy-two hours in water, one foot of which is submerged in a conducting liquid" held in a metal trough, one of the transformer terminals being connected to the wire and the other to the metal of the trough. Insulations for voltag-es between 600 and 3.5»©0: e. The thickness of the insulating walls must not be less than those given in the following table for B. & S. gauge sizes : From 14 to 1, inclusive, gV From 0 to 500,000, C. M., ^" covered by a tape or a braid. Larger than 500,000, C. M., ^" covered by a tape or a braid /. The requirements as to insulation and break-down resistance for wires for low potential systems shall apply, with the exception that an insulation resistance of not less than 300 megohms per mile shall be required. g. Wire for arc-light circuits exceeding 3,500 volts potential shall have an insulating wall not less than six-thirty-seconds of an inch in thickness, and shall withstand a break-down test of at least 30,000 volts, and have an insulation of at least 500 megohms per mile. The tests on this wire to be made under the same conditions as for low- potential wires. Specifications for insulations for alternating currents exceeding 3,500 ( id NATIONAL ELECTRICAL CODE. volts have been considered, but on account of tbe somewhat complex con- ditions in such work it has so far been deemed inexpedient to specify gen- eral insulations for this use. h. All of the above insulations must be protected by a substantial braided covering properly saturated with a preservative compound and suffi- ciently strong to withstand all t,he abrasion likely to be met with in prac- tice, and sufficiently elastic to permit all wires smaller than No. 7 B. & S. gauge to be bent around a cylinder with twice the diameter of the wire, without injury to the braid. 42. Slow-lrarning- Weatherproof. a. The insulatioi shall consist of two coatings, the inner one to be fire- proof in character, the outer to be weatherproof. The inner fireproof coat- ing must comprise at least six-tenths of the total thickness of the wall. The completed covering must be of a thickness not less than that given in the following table for 13. & S. gauge sizes : rorn 14 to 8, inclusive, &ri 7 to 2, A" " 2 to oooo, By " 0000 to 500,000, CM., 5y' " 500,000 to 1,000,000, '• B'4y/ arger than 1,000,000, " i" Measurements of insulating wall are to be made at the thinnest portion of the dielectric. b. The inner fireproof coating shall be layers of cotton or other thread, the outer one of which must be braided. All the interstices of these layers are to be filled with the fireproofing compound. This is to be material whose solid constituent is not susceptible to moisture, and which will not burn even when ground in an oxidizable oil, making a compound which, while proof against fire and moisture, at same time has considerable elasti- city, and which when dry will suffer no change at a temperature of 250 degrees Fahrenheit, and which will not burn at even higher temperature. c. The weatherproof coating shall be a stout braid thoroughly satu- rated with a dense moistureproof compound thoroughly slicked down, applied in such manner as to drive any atmospheric moisture from the cotton braiding, thereby securing a covering to a greater degree waterproof and of high insulating power. This compound to retain its elasticity at zero Fahrenheit, and not to drip at 160 degrees Fahrenheit. This wire is not as burnable as the old " weatherproof," nor as subject to softening under heat, but still is able to repel the ordinary amount of moisture found indoors. It would not usually be used for outside work. 43. Slow-burning-. a. The insulation shall be the same as the " slow-burning weatherproof," except that the outer braiding shall be impregnated with a fireproofing compound similar to that required for the interior layers, and with the outer surface finished smooth and hard. This " slow-burning" wire shall only be used with special permission of the Inspection Department having jurisdiction. This is practically the old " Underwriters' " insulation. It is specially useful in hot, dry places where ordinary insulations would perish, also where wires are bunched, as on the back of a large switchboard or in a wire tower so that the accumulation of rubber or weatherproof insulation would result in an objectionably large mass of highly inflammable material. Its use is restricted, as its insulating qualities are not high and are dam- aged by moisture. 44. Weatherproof. a. The insulating covering shall consist of at least three braids thoroughly impregnated with a dense moisture repellent, which will not drip at a tem- perature lower than 180 degrees Fahrenheit. The thickness of insulation shall be not less than that of "slow-burning weatherproof." The outer surface shall be thoroughly slicked down." This wire is for outdoor use where moisture is certain and where fireproof qualities are not necessary. CLASS D. FITTINGS, MATERIALS, AND DETAILS. 779 45. flexible Cord — a. Must be made of stranded copper conductors, each strand to be not larger than No. 26 or smaller than No. 30 B. & S. gauge, and each stranded conductor must be covered by an approved insulation and protected from mechanical injury by a tougb braided outer covering. For pendent lamps: In this class is to be included all flexible cord which under usual condi- tions bangs freely in air, and whicb is not likely to be moved sufficiently to come in contact with surrounding objects. b. Each stranded conductor must have a carrying capacity equivalent to not less than a No. 18 B. & S. gauge wire. c. The covering of each stranded conductor must be made up as follows : 1. A tight, close wind of fine cotton. 2. The insulation proper, which shall be either waterproof or slow- burning. 3. An outer cover of silk or cotton. The wind of cotton tends to prevent a broken strand puncturing the insu- lation and causing a short circuit. It also keeps the rubber from corroding the coppei d. Waterproof insulation must be solid, at least one-thirty-second of an inch thick, and must show an insulation resistance of fifty megohms per mile throughout two weeks' immersion in water at 70 degrees Fahrenheit, and stand the test prescribed for low-tension wires as far as they apply. e. Slow-burning insulation must be at least one-thirty-second of an inch in thickness, and composed of substantial, elastic, slow-burning materials, which will suffer no damage at a temperature of 250 degrees Fahrenheit. /. The outer protecting braiding should be so put on and sealed in place that when cut it will not fray out, and where cotton is used, it should be impregnated with a flameproof paint, which will not have an injurious effect on the insulation. For portables: In this class is included all cord used on portable lamps, small portable motors, etc. g. Flexible cord for portable use must have waterproof insulation as re'quired in section d for pendent cord, and in addition be provided with a reinforcing cover especially designed to withstand the abrasion it will be subject to in the uses to which it is to be put. For portable beating* apparatus: h. Must be made up as follows : — 1. A tight, close wind of fine cotton. 2. A thin layer of rubber about one-one-hundredth of an inch thick, or other cementing material. _ 3. A layer of asbestos insulation at least three-sixty-fourths of an inch thick. 4. A stout braid of cotton. 5. An outer reinforcing cover especially designed to withstand abrasion. This cord is in no sense waterproof, the thin layer of rubber being speci- fied in order that it may serve merely as a seal to help hold in place the fine cotton and asbestos, and it should be so put on as to accomplish this. 4©. Fixture Wire — a. Must have a solid insulation, with a slow-burning, tough, outer cover- ing, the whole to be at one-thirty-second of an inch in thickness, and show an insulation resistance between conductors, and between either conductor and the ground, of at least one megohm per mile, after one week s submer- sion in water at seventy degrees Fahrenheit, and after three minutes electrification with 550 volts. 4*. Conduit Wire — Must complv with the following specifications : a. For metal conduits, having a lining of insulating material, single wires 780 NATIONAL ELECTRICAL CODE. must comply with Xn. 41, and all duplex, twin, and concentric conductors must comply with No. 41, and must also have each conductor separately braided or taped and a substantial braid covering the whole. b. For unlined metal conduits, conductors must conform to the specifica- tions given for lined conduits, and in addition have a second outer fibrous covering at least one-thirty-second of an inch in thickness, and sufficiently tenacious to withstand tlie abrasion of being hauled through the metal conduit. The braid required around each conductor in duplex, twin, and concen- tric cables is to hold the rubber insulation in place and prevent jamming and flattening. 48. .Armored Cable. a. The armor of such cables must be at least equal in thickness and of equal strength to resist penetration by nails, etc., as the armor of metal covering of metal conduits (see No. 49 b). b. The conductors in same, single wire or twin conductors, must have an insulating covering as required by No. 41, any filler used to secure a round exterior must be impregnated with a moisture repellent, and the whole bunch of conductors and fillers must have a separate exterior covering of insulating material at least one-thirty-second of an inch in thickness, con- forming to the insulation standard given in No. 41, and covered with a sub- stantial braid. Very reliable insulation is specified, as such cables are liable to hard usage, and in part of their length may be subject to moisture, while they may not be easily removable, so that a breakdown of insulation is likely to be expensive. -!■«>. Interior Conduits. {For wiring rules, see Nos. 24 and 25.) a. Each length of conduit, whether insulated or uninsulated, must have the maker's name or initials stamped in the metal or attached thereto in a satisfactory manner, so that the inspectors can readily see the same. METAL CONDUITS WITH LINING OF INSULATING MATERIAL. b. The metal covering or pipe must be equal in strength to the ordinary commercial forms of gas-pipe of the same size, and its thickness must be not less than that of standard gas-pipe, as shown by the following table : Size, nches. Thickness of Wall — Inches. Size. Inches. Thickness of Wall — Inches. h 1 1 .109 .111 .113 if 2" .140 .145 .154 1 -.134 An allowance, of two one-hundredths of an inch for variation in manu- facturing and loss of thickness by cleaning will be permitted. c. Must not be seriously affected externally by burning out a wire inside the tube when the iron pipe is connected to one side of the circuit. d. Must have the insulating lining firmly secured to the pipe. e. The insulating lining must not crack or break when a length of the conduit is uniformly bent at temperature of 212 degrees Fahrenheit to an angle of ninety degrees, with a curve having a radius of fifteen inches, for pipes of one inch and less, and fifteen times the diameter of pipe for larger pipes. /. The insulating lining must not soften injuriously at a temperature below 212 degrees Fahrenheit, and must leave water in which it is boiled practically neutral. (j. The insulating lining must be at least one-thirty-second of an inch in thickness ; and the materials of which it is composed must be of such a nature as will not have a deteriorating effect on the insulation of the con- ductor, and be sufficiently tough and tenacious to withstand the abrasion test of drawing long lengths of conductors in and out of same. CLASS D. FITTINGS, MATERIALS, AND DETAILS. 781 h. The insulating lining must not be mechanically weak after three days' submersion in water, and when removed from the pipe entire must not absorb more than ten per cent of its weight of water during 100 hours of submersion. *. All elbows or bends must be so made that the conduit or lining of same will not be injured. The radius of the curve of the inner edge of any elbow not to be less than three and one-half inches. Must have not more than the equivalent of four quarter bends from outlet to outlet, the bends at the outlets not being counted. UXLIXED METAL CONDUITS. ?". Plain iron or steel pipes of equal thickness and strengths specified for lined conduits in No. 49 b may be used as conduits, provided their interior surfaces are smooth and free from burs ; pipe to be galvanized, or the interior surfaces coated or enameled, to prevent oxidation, with some sub- stance which will not soften so as to become sticky and prevent wire from being withdrawn from the pipe. k. All elbows or bends must be so made that the conduit will not be injured. The radius of the curve of the inner edge of any elbow not to be less than three and one-half inches. Must have not more than the equiva- lent of four quarter bends from outlet to outlet, the bends at the outlet not being counted. 5©. Wooden Moldiiag-s — (For wiring rules, see No. 24.) a. Must have, both outside and inside, at least two coats of waterproof paint, or be impregnated with a moisture repellent. b. Must be made of two pieces, a backing and capping, so constructed as to thoroughly incase the wire, and provide a one-half inch tongue between the conductors, and a solid backing, which, under grooves, shall not be less than three-eighths of an inch in thickness, and must afford suitable protec- tion from abrasion. It is recommended that only hardwood molding be used. 51. Switches — (See Nos. 17 and 22.) a. Must be mounted on non-combustible, non-absorptive, insulating bases, such as slate or porcelain. b. Must have carrying capacity sufficient to prevent undue heating, c. Must, when used for service switches, indicate, on inspection, whether the current be " on " or " off." d. Must be plainly marked, Avhere it will always be visible, with the name of the maker and the current and voltage for which the switch is designed. e. Must, for constant potential systems, operate successfully at fifty per cent overload in amperes, Avith twenty-rive per cent excess voltage under the most severe conditions they are liable to meet with in practice. /. Must, for constant potential systems, have a firm and secure contact ; must make and break readily, and not stop when motion has once been imparted by the handle. g. Must, for constant current systems, close the main circuit and discon- nect the branch wires when turned" off " ; must be so constructed that they shall be automatic in action, not stopping between points when started, and must prevent an arc between the points under all circumstances. They must indicate, upon inspection, whether the currents be " on " or " off." 52. Cutouts and Circuit-Breakers — (For installation rules, see Nos. 17 and. 21.) a. Must be supported on bases of non-combustible, non-absorptive insu- lating material. b. Cutouts must be provided with covers, when not arranged in approved cabinets, so as to obviate any danger of the melted fuse metal coming in contact with any substance which might be ignited thereby. t OJ NATIONAL ELECTRICAL CODE. c. Cutouts must operate successfully, under the most severe conditions they are liable to meet with in practice, on short circuits with fuses rated at fifty per cent above, and Avith a voltage twenty-five per cent above the current and voltage for which they are designed. d. Circuit-breakers must operate successfully, under the most severe conditions they are liable to meet with in practice, on short circuits when set at fifty per cent above the current, and with a voltage twenty-five per cent above that for which they are designed. e. Must be plainly marked, where it will always be visible, with the name of the maker, and current and voltage for which the device is de- signed. 53. fuses — (For installation rules, see Kos. 17 and 21.) a. Must have contact surfaces or tips of harder metal having perfect electrical connection with the fusible part of the strip. b. Must be stamped with about eighty per cent of the maximum current they can carry indefinitely, thus allowing about twenty-five per cent over- load before fuse melts. With naked open fuses, of ordinary shapes and not over 500 amperes capacity, the maximum current which will melt them in about five minutes may be safely taken as the melting point, as the fuse practically reaches its maximum temperature in this time. With larger fuses a longer time is necessary. Inclosed fuses where the fuse is often in contact with substances having good conductivity to heat and often of considerable volume, require a much longer time to reach a maximum temperature, on account of the surrounding material which heats up slowly. These data are given to facilitate testing. c. Fuse terminals must be stamped with the maker's name, initials, or some known trade-mark. 54. Cutout Caliiuets — a. Must be so constructed, and cutouts so arranged, as to obviate any danger of the melted fuse metal coming in contact with any substance which might be ignited thereby. A suitable box can be made of marble, slate, or wood, strongly put together, the door to close against a rabbet so as to be perfectly dust-tight ; and it should be hung on strong hinges, and held closed by a strong hook or catch. If the box is wood, the inside should be lined with sheets of asbestos board about one-sixteenth of an inch in thickness, neatly put on, and firmly secured in place by shellac and tacks. The wire should enter through holes bushed with porcelain bushings ; the bushings tightly fitting the holes in the box, and the wires tightly fitting the bushings (using tape to build up the wire, if necessary) so as to keep out the dust. 55. Sockets. (See No. 27.) Sockets of all kinds, including wall receptacles, must be constructed in accordance with the following specifications : — a. Standard Sizes. — The standard lamp socket shall be suitable for use on any voltage not exceeding 250 and with any size lamp up to fifty candle-power. For lamps larger than fifty candle-power a standard keyless socket may be used ; or if a key is required, a special socket designed for the current to be used must be made. Any special sockets must follow the general spirit of these specifications. b. Marking. — The standard socket must be plainly marked fifty candle- power, 250 volts, and with either the manufacturer's name or registered trademark. Special large sockets must be marked Avith the current and voltage for which they are designed. c. Shell. — Metal used for shells 'must be moderately hard, but not hard enough to be brittle or so soft as to be easily dented or knocked out of place. Brass shells must be at least 0.013 inch in thickness, and shells of any other material must be thick enough to give the same stiffness and strength of brass. CLASS D. — FITTINGS, MATERIALS, AND DETAILS. 78o d. Lining. — The inside of the shells must he lined with insulating material, which shall absolutely prevent the shell from becoming a part of the circuit, even though the wires inside the socket should start from their position under binding screws. The material used for lining must be at least one thirty-second of an inch in thickness, and must be tough and tenacious. It must not be in- juriously affected by the heat from the largest lamp permitted in the socket, and must leave the water in which it is boiled practically neutral. It must be so firmly secured to the shell that it will not fall out with ordinary handling of the socket. It is preferable to have the lining in one piece. e. Cap. — Caps when of sheet brass 'must be at least 0.013 inch in thick- ness, and when cast or made of other metals must be of equivalent strength. The inlet piece, except for special sockets, must be tapped and threaded for ordinary one-eight-inch pipe. It must contain sufficient metal for a full, strong thread, and, when not of the same piece as the cap, must be joined to it in a way to give the strength of a single piece. There must be sufficient room in the cap to enable the ordinary wireman to easily and quickly make a knot in the cord, and push it into place in cap without crowding. All parts of the cap upon which the knot is likely to bear must be smooth and well insulated. /. Frame and Screws. — The frame holding moving parts must be sufficiently heavy to give ample strength and stiffness. Brass pieces containing screw threads must be at least 0.06 of an inch in thickness. Binding-post screws must not be smaller than No. 5 wire and about forty threads per inch. g. Spacing. —Points of opposite polarity must everywhere be kept not less than three sixty-fourths of an inch apart unless separated by a reliable insulation. h. Connections. — The connecting points for the flexible cord must be made to very securely grip a No. 16 or 18 B. & S. conductor. A turned-up lug, arranged so that the cord may be gripped between the screw and the lug in such a way that it cannot possibly come out, is strongly advised. i. Lamp-Holder. — The socket must firmly hold the lamp in place so that it cannot be easily jarred out, and must provide a contact good enough to prevent undue heating with maximum current allowed. The holding- pieces, springs and the like, if a part of the circuit, must not be sufficiently exposed to allow them to be brought in contact with anything outside of lamp and socket. j. Base. —The inside parts of the socket, which are of insulating material, except the lining, must be made of porcelain. k. Key. — The socket key-handle must be of such a material that it will not soften from the heat of a fifty candle-power lamp hanging downwards in air at seventy degrees Fahrenheit from the socket, and must be securely, but not necessarily" rigidly, attached to the metal spindle it is designed to turn. /. Sealing. — All screws in porcelain pieces, which can be firmly sealed in place, must be so sealed by a waterproof compound which will not melt below 200 degrees Fahrenheit. to. Putting Together. — The socket must, as a whole, be so put together that it will not rattle to pieces. Bayonet joints or equivalent are recommended. * n. Test. — The socket when slowly turned "on and off," at the rate of about two or three times per minute, must " make and break " the circuit 6,000 times before failing, when carrying a load of one ampere at 220 volts. o. Keyless Sockets.— Keyless sockets of all kinds must comply with requirements for key sockets as far as they apply. p. Sockets of Insulating Materials. — Sockets made of porcelain or other insulating material must conform to the above requirements as far as they apply, and all parts must be strong enough to withstand a moderate amount of hard usage without breaking. q. Inlet Bushing. — When the socket is not attached to fixtures the threaded inlet must be provided with a strong insulating bushing, having a smooth hole of at least fifteen sixty-fourths of an inch in diameter. The corners of the bushing must be rounded, and all inside fins removed, so that in no place will the cord be subjected to the cutting or wearing action of a sharp edge. 784 NATIONAL ELECTRICAL CODE. 5G. Hanger-boards. a. Hanger-boards must be so constructed that all wires and current- carrying devices thereon shall be exposed to view, and thoroughly insu- lated by being mounted on a non-combustible, non-absorptive insulating substance. All switches attached to the same must be so constructed that they shall be automatic in their action, cutting off both poles to the lamp, not stopping between points when started, and preventing an arc betwesn points under all circumstances. »?. Arc liamps. (For installation rules, see No. 19.) a. Must be provided with reliable stops to prevent carbons from falling out in case the clamps become loose. b. Must be carefully insulated from the circuit in all their exposed parts. c. Must, for constant-current systems, be provided with an approved hand switch, also an automatic switch that will shunt the current around the carbons, should they fail to feed properly. The hand switch to be approved, if placed anywhere except on the lamp itself, must comply with requirements for switches on hanger-boards as laid down in No. 56. 58. Spark Arresters. (See No. 19c.) a. Spark arresters must so close the upper orifice of the globe that it will be impossible for any sparks thrown off by the carbons to escape. 50. Insulating1 Joints - (See No. 26 a.) a. Must be entirely made of material that will resist the action of illumi- nating gases, and will not give way or soften under the heat of an ordinary gas-flame, or leak under a moderate pressure. They shall be so arranged that a deposit of moisture will not destroy the insulating effect, and shall have an insulating resistance of at least 250,000 ohms between the gas-pipe attachments, and be sufficiently strong to resist the strain they will be liable to be subjected to in being installed. Insulating Joint for Gas Pipes. b. Insulating joints having soft rubber in their construction will not be approved. GO. Resistance Boxes and Equalizers — (For installation rules, see No. 4.) a. Must be equipped with metal or with other non-combustible frames. The word " frame " in this section relates to the entire case and sur- roundings of the rheostat, and not alone to the upholding supports. CLASS D. FITTINGS, MATERIALS, AND DETAILS. 785 Gl. Reactive Coils and Condensers. a. Reactive coils must be made of non-combustible material, mounted on non-combustible bases, and treated, in general, like sources of heat. b. Condensers must be treated like apparatus operating with equivalent voltage and currents. They must have non-combustible cases and supports, and must be isolated from all combustible materials, and, in general, treated like sources of heat. G2. Transformers — (For installation rules, see Nos. 11, 13, and 33.) a. Must not be placed in any but metallic or other non-combustible cases. b. Must be constructed to comply with the following tests : 1. Shall be run for eight consecutive hours at a full load in watts under conditions of service, and at the end of that time the rise in temperature, as measured by the increase of resistance of the primary coil, shall not exceed 135 degrees Fahrenheit. 2. The insulation of transformers when heated shall withstand con- tinuously for five minutes a difference of potential of 10,000 volts (alternating) between primary and secondary coils and core, and between the primary coils and core and a no-load " run " at double voltage for thirty minutes. G3. liig-litning; Arresters. (For installation rules, see No. 5.) a. Must be mounted on non-combustible bases, and must be so con- structed as not to maintain an arc after the discharge has passed, and must have no moving parts. CLA§i E. — JfEISCEIiEiAMEOUS. G4. Sig-naling- Systems (governing wiring for telephone, telegraph, district messenger, and call-bell circuits, fire and burglar alarms, and all similar systems) — a. Outside wires should be run in undergrouna ducts or strung on poles and, as far as possible, kept off of buildings, and must not be placed on the same cross-arm with electric light or power wires. b. When outside wires are run on same pole with electric light or power wires, the distance between the two inside pins of each cross-arm must not be less than twenty-six inches. c. All aerial conductors and underground conductors which are directly connected to aerial wires must be provided with some approved protective device, which shall be located as near their point of entrance to the build- ing as possible, and not less than six inches from curtains or other inflam- mable material. d. If the protector is placed inside of building, wires, from outside sup- ports to binding-posts of protector, shall comply with the following require- ments : 1. Must be of copper, and not smaller than ISio. 16 B. & S. gauge. 2. Must have an approved rubber insulating covering (see No. 41). , 3. Must have drip loops in each wire immediately outside the building. 4. Must enter buildings through separate holes sloping upward from the outside ; when practicable, holes to be bushed with non-absorptive, non-combustible insulating tubes extending through their entire length. Where tubing is not practicable, the wires shall be wrapped with two layers of insulating tape. 5. Must be supported on porcelain insulators, so that they will not come in contact with anything other than their designed supports. 6. A separation between wires of at least two and one-half inches must '"be maintained. In case of crosses these wires may become a part of a high-voltage circuit, so that similar care to that given high-voltage circuits is needed in placing them. Reliable porcelain bushings at the entrance holes are desirable, and are only waved under adverse conditions, because the state of the art in this type of wiring makes an absolute requirement inadvisable. 78G NATIONAL ELECTRICAL CODE. e. The ground wire of the protective device shall be run in accordance with the following requii ements : 1. Shall be of copper, and not smaller than No. 16 B. & S. 2. Must have an approved rubber insulating covering (See No. 41). 3. Shall run in as straight a line as possible to a good permanent ground, to be made by connecting to water- or gas-pipe, preferably water-pipe. If gas-pipe is used, the connection, in all cases, must be made between the meter and service pipes. In the absence of other good ground, the ground shall be made by means ol a metallic plate or bunch of wires buried in permanently moist earth. 4. Shall be kept at least three inches from all other conductors, and sup- ported on porcelain insulators so as not to come in contact with anything other than its designated supports. In attaching a ground wire to a pipe, it is often difficult to make a thoroughly reliable solder joint. It is better, therefore, where possible, to carefully solder the wire to a brass plug, which may then be firmly screwed into a pipe fitting. Where such joints are made under ground, they should be thoroughly painted and taped to prevent corrosion. f. The protector to be approved must comply with the following require- ments : 1. Must be mounted on non-combustible, non-absorptive insulating bases, so designed that when the protector is in place, all parts which may be alive will be thoroughly insulated from the wall holding the protector. 2. Must have the following parts : A lightning arrester which will operate with a difference of potential between wires of not over 500 volts, and so arranged that the chance of accidental grounding is reduced to a minimum. A fuse designed to open the circuit in case the wires become crossed with light or power circuits. The fuse must be able to open the circuit without arcing or serious flashing -"'hen crossed with any ordinary commercial light or power circuit. A heat coil which will operate before a sneak current can damage the instrument the protector is guarding. The heat coil is designed to warm up and melt out with a current large enough to endanger the instruments if continued for a long time, but so small that it would not blow the fuses ordinarily found necessary for such instruments. These smaller currents are often called " sneak " currents. 3. The fuses must be so placed as to protect the arrester and heat coils, and the protector terminals must be plainly marked "line," "in- strument," " ground." g. Wires beyond the protector, except where bunched, must be neatly arranged and securely fastened in place in any convenient, workmanlike manner. They must not come nearer than six inches to any electric light or power wire in the building, unless incased in approved tubing so secured as to prevent its slipping out of place. The wires would ordinarily be insulated, but the kind of insulation is not specified, as the protector is relied upon to stop all dangerous currents. Porcelain tubing or circular loom conduit may be used for incasing wires where required as above. h. Wires connected with outside circuits, where bunched together within any building, or inside wires, where laid in conduits or ducts, with electric light or power wires, must have fire-resisting coverings, or else must be inclosed in an air-tight tube or duct. It is feared that if a burnable insulation were used, a chance spark might ignite it and cause a serious fire, for many installations contain a large amount of very readily burnable matter. G;1. Electric *-ii«. lag-hting-. Where electric gas lighting is to be used on the same fixture with the electric light : a. No part of the gas-piping or fixture shall be in electric connection with the gas-lighting circuit. CLASS E. MISCELLANEOUS. 787 b. The wires used with the fixtures must have a non-inflammable insula- tion, or, where concealed between the pipe and shell of the fixture, the insulation must be such as required for fixture wiring for the electric light. c. The whole installation must test free from " grounds." d. The two installations must test perfectly free from connection with each other. OO. Insulation .Resistance. The wiring in any building must test free from grounds ; i. e., the com- plete installation must have an insulation between conductors and between all conductors and the ground (not including attachments, sockets, recep-. tacles, etc.) of not less than the following : Up to 5 ampere: 10 25 50 100 200 400 " 800 " 1,600 " 4,000,000 ohms 2,000,000 4 800,000 4 400,000 4 200,000 4 100,000 4 25,000 ' 25,000 4 12,500 4 All cutouts and safety devices in place in the above. Where lamp sockets, receptacles, and electroliers, etc., are connected, one-half of the above will be required. tit. Soldering- Fluid. a. The following formula for soldering fluid is suggested : Saturated solution of zinc chloride 5 parts Alcohol 4 parts Glycerine . 1 part CLAS§ F.— BEARIIVi: WOJtlfc. ft§. Generators — a. Must be located in a dry place. ft. Must have their frames insulated from their bed-plates. c. Must each be provided with a waterproof cover. (I. Must each be provided with a name-plate, giving the maker's name, the capacity in voltage and amperes and normal speed in revolutions per minute — «©. Wires — a. Must have an approved insulating covering. The insulation for all conductors, except for portables, to be approved, i must be at least one-eighth-inch in thickness and be covered with a substan- I tial waterproof and flameproof braid. The physical characteristics shall > not be affected by any change in temperature up to 200 degrees Fahrenheit. After two weeks' submersion in salt water at seventy degrees Fahrenheit it must show an insulation resistance of one megohm per mile after three minutes' electrification, with 550 volts. b. Must have no single wire larger than No. 12 B. & S. "Wires to be 1 stranded when greater carrying capacity is required. No single solid wire smaller than No. 14 B. & S. /except in fixture wiring, to be used. ; Stranded wires must be soldered before being fastened under clamps or binding screws, and when they have a conductivity greater than No. 10 B. & S. copper wire they must be soldered into Ligs. c. Must be supported in approved molding, except at switchboards and portables. Special permission may be given for deviation from this rule in dynamo- rooms. d. Must be bushed with hard-rubber tubing one-eighth of an inch in thickness when passing through beams and non-water-tight bulkheads. 788 NATIONAL ELECTRICAL CODE. e. Must have, when passing through water-tight bulkheads and through all decks, a metallic stuffing-tube lined with hard rubber. In case of deck tubes they shall be boxed near deck to prevent mechanical injury. f. Splices or taps in conductors must be avoided as far as possible. Where it is necessary to make them they must be so spliced or joined as to be both mechanically and electrically secure without solder. They must then be soldered, to insure preservation, covered with an insulating compound equal to the insulation of the wire, and further protected by a waterproof tape. The joint must then be coated or painted with a waterproof compound. a©. Portable Conductors — a. Must be made of two stranded conductors, each having a carrying capacity equivalent to not less than No. 14 B. & S. wire, and each covered with an approved insulation and covering. Where not exposed to moisture or severe mechanical injury, each stranded conductor must have a solid insulation at leas! one-thirty-second of an inch in thickness, and must show an insulation resistance between conductors, and between either conductor and the ground, of at least one megohm per mile after one week's submersion in water at seventy degrees Fahrenheit and after three minutes' electrification, with 590 volts, and be protected by a slow-burning, tough-braided outer covering, Where exposed to moisture and mechanical injury — a? for use on decks, holds, and fire-rooms — each stranded conductor shall have a solid insula- tion to be approved, of at least one-thirty-second of an inch in thickness and protected by a tough braid. The two conductors shall then be stranded together, using a jute filling. The whole shall then be covered with a layer of flax, either woven or braided, at least one-thirty-second of an inch' in thickness, and treated with a non-inflammable waterproof compound. After one week's submersion in water at seventy degrees Fahrenheit, at 55C volts and a three minutes' electrification, must show an insulation between the two conductors, or between either conductor and the ground, of one megohm per mile. 91. Bell or Other Wires — a. Shall never run in same duct with lightning or power wires. ll as. Tal»le of Capacity of Wires. t: B. & S. G. Area Actual No. of Size of Strands CM. Strands. B. &S. G. Amperes. L 19 1,288 - 18 1,624 3 in 17 2,048 ;' 16 2,583 *6 ?i 15 3,257 ! 14 4,107 12 £' 12 6,530 17 i 9,016 i 19 21 H 11,368 7 18 25 i, 14,336 7 17 30 ::t 18,081 7 16 35 (, 22,799 7 15 40 K. 30,856 19 18 50 "Il 38,912 19 17 60 If 49,077 19 16 70 •' 60,088 37 18 85 k 75,776 37 17 100 iiiii 99,064 61 18 120 i 124,928 61 17 145 his 157,563 61 16 170 198,677 61 15 200 250,527 61 14 235 296,387 91 15 270 J, 373,737 91 14 320 ■> 413,639 127 15 340 1, CLASS F. MARINE WORK. 789 "When greater conducting area than that of a single wire is required, the conductor shall be stranded in a series of *, lO, 31, Ol, Ol, or 12*, vires as may be required ; the strand consisting of one central wire, the remainder laid around it concentrically, each layer to be twisted in the opposite direction from the preceding 73. Switchboard's — a. Must be made of non-combustible, non-absorbtive insulating material, such as marble or slate. 6. Must be kept free from moisture, and must be located so as to be accessible from all sides. c. Must have a main switch, main cutout, and ammeter for each gen- erator. Must also have a voltmeter and ground detector. d. Must have a cutout and switch for each side of each circuit leading from board. "S4r. [Resistance Boxes — a. Must be made of non-combustible material. b. Must be located on switchboard or away from combustible material. When not placed on switchboard they must be mounted on non-inflam- mable, non-absorptive insulating material. c. Must be so constructed as to allow sufficient ventilation for the uses to which they are put. "33. Switches — a. Must have non-combustible, non-absorptive insulating bases. b. Must operate successfully at fifty per cent overload in amperes with twenty-five per cent excess voltage under the most severe conditions they are liable to meet with in practice, and must be plainly marked, where they will always be visible, with the name of the maker and the current and voltage for which the switch is designed. c Must be double pole when circuits which they control supply more than six sixteen-candle-power lamps or their equivalent. d. When exposed to dampness, they must be inclosed in a water-tight ?G. Cutouts — a. Must have non-combustible, non-absorptive insulating bases. b. Must operate successfully, under the most severe conditions they are liable to meet with in practice, on short circuit with fuse rated at fifty per cent above, and with a voltage twenty-five per cent above the current and voltage they are designed for, and must be plainly marked, where they will always be visible, with the name of the maker and current and voltage for which the device is designed. c. Must be placed at every point where a change is made in the size of the wire (unless the cutout in the larger wire will protect the smaller). d. In places such as upper decks, holds, cargo spaces, and fire-rooms a water-tight and fireproof cutout may be used, connecting directly to mains when such cutout supplies circuits requiring not more than' 660 watts energy. e. Wben placed anywhere except on switchboards and certain places, as cargo spaces, holds, fire-rooms, etc., where it is impossible to run from center of distribution, they shall be in a cabinet lined with fire-resisting material. /. Except for motors, searchlights, and diving-lamps shall be so placed that no group of lamps, requiring a current of more than six amperes, sball ultimately be dependent upon one cutout. A single-pole covered cutout may be placed in the molding when same con- tains conductor supplying circuits requiring not more than 220 watts energy. •J1?. [Fixtures — . Shall be mounted on blocks made from Avail-seasoned lumber treated with two coats of white lead or shellac. b. Where exposed to dampness, the lamp must be surrounded by a vapor- proof globe. 790 NATIONAL ELECTRICAL CODE. e. Where exposed to mechanical injury the lamp must he surrounded hy a glohe protected by a stout wire guard. cL Shall be wired -with, same grade of insulation as portable conductors which are not exposed to moisture or mechanical injury. «•*». Sockets. a. No portion of the lamp socket or lamp base exposed to contact with outside objects shall be allowed to come into electrical contact with either of the conductors. tO. Wooden HEoulding-s — a. Must be made of well-seasoned lumber and be treated inside and out with at least two coats of white lead or shellac. b. Must be made of two pieces, a backing and a capping, so constructed as to thoroughly incase the wire, and provide a one-half inch tongue between the conductors, and a solid backing which, under grooves, shall not be less than three-eighths of an inch in thickness. c. Where molding is run over rivets, beams, etc., a backing strip must first be put up and the molding secured to this. d. Capping must be secured by brass screws. $©. Motors — a. Must be wired under the same precautions as with a current of same volume and potential for lighting. The motor and resistance box must be protected by a double-pole cutout, and controlled by a double-pole switch, except in cases where one-quarter horse-power or less is used. The leads or branch circuits should be designed to carry a current at least fifty per cent greater than that required by the rated capacity of the motor to provide for the inevitable overloading of the motor at times. b. Must be thoroughly insulated. Where possible, should be set on base frames made from filled, hard, dry, wood, and raised above surrounding deck. On hoists and winches they shall be insulated from bed-plates by hard rubber, fiber, or similar insulating material. c. Shall be covered with a waterproof cover when not in use. d. Must each be provided with a name-plate giving maker's name, the capacity in volts and amperes, and the normal speed in revolutions per minute. GMERAI fcl'CiGJESTIOJfS. In all electric work conductors, however well insulated, should always be treated as bare, to the end that under no conditions, existing or likely to exist, can a grounding or short circuit occur, and so that all leakage from conductor to conductor, or between conductor and ground, may be reduced to the minimum. In all wiring special attention must be paid to the mechanical execution of the work. Careful and neat running, connecting, soldering, taping of conductors and securing and attaching of fittings, are specially conducive to security and efficiency, and will be strongly insisted on. In laying out an installation, except for constant-current systems, the work should, if possible, be started from a center of distribution, and the switches and cutouts, controlling and connected with the several branches, be grouped together in a safe and easily accessible place, where they can be readily got at for attention or repairs. The load should be divided as evenly as possible among the branches, and all complicated and unnecessary wiring avoided. The use of wire-ways for rendering concealed wiring permanently acces- sible is most heartily indorsed and recommended ; and this method of accessible concealed construction is advised for general use. Architects are urged, when drawing plans and specifications, to make pro- vision for the channeling and pocketing of buildings for electric light or power wires, and in specifications for electric gas lighting to require a two- wire circuit, whether the building is to be wired for electric lighting or not, so that no part of the gas fixtures or gas-piping be allowed to be used for the gas-lighting circuit. FOUNDATIONS AND STRUCTURAL MATERIALS. rOWEH 8IATIOI COflfSlRrCTIOHf. Chart. (By E. P. Roberts & Co.) fFoundafcion (A Setting [.Stack Sta- tion Steam Plant Link y En- gines j ^Source J Pumps and injectors, valves ] and gauges (^Heaters fSediment ( Blow off I Mud drum {Steam pipe and valve to heater Entrained water, separator Placing in building Placing in boiler Removal of coke and ashes ^Removal of soot Supply to surface Piping and valves Coverings Drains and drips ^Supports 'Foundation . Steam to cylinder Oil to cylinder Steam from cylinder Water from cylinder Oil to engine Oil from engine Engine indicator ( Steam to condenser ■{ Water to condenser I^Water from condenser | Foundations ( Lubrication f Belts ^Connecting links . . . -< Shafts * ^Pulleys ("Foundation I Lubrication Insulation Governing devices Measuring devices [Safety devices Elec- [Dynamos to switchboard trical -i Wire <; Switchboards to line plant I Track to dynamo [Distribution devices I Dynamo governing devices J Dynamo measuring devices ^Switchboard 1 Feeder to measuring devices I Safety devices '^Cut-out and lightning arrester ( Weatherproof | Fireproof Build- J. Ventilated ing Light (^Provisions for cranes or other strains foreign to its func- tions as a shelter. 791 792 FOUNDATIONS AND STRUCTURAL MATERIALS. FOUNDATIONS. The term foundation designates the portion of a structure used as a base on which to erect the superstructure, and must be so solid that no move- ment of the superstructure can take place after its erection. As all foundations or structures of coarse masonry, whether of brick or stone, will settle to some extent, and as nearly all soils are compressible under heavy weight, care must be taken that the settlement be even all over the structure in order to avoid cracks or other haws. Although it is quite general to make the excavation for all the sub-foundation without predetermining in mure than a general way the nature of the subsoil, and then adapting the base of the foundation to the nature of soil found ; yet in large undertakings, where there may be question as to the bearing, borings are made and samples brought up in order to determine the different strata and distance of rock below the surface. Where foundations are not to be deep, or the soil is of good quality, a trench or pit is often sunk alongside the location of the proposed foundation, and the quality of the soil deter- mined in that way. Foundations on Rock. The surface of rock should be cleaned and dressed, all decayed portions removed, crevices filled with grouting or concrete, and where the surface is inclined, it should be cut into a series of level steps before commencing the structure In such cases of irregular levels, all mortar joints must be kept as close as possible, iu order to prevent unequal settlement. A still better way is to bring all such uneven surfaces to a common level with a good thick bed of concrete, which, if properly made, will become as incom- pressible as stone or brick. The load on rock foundation should never exceed one-eighth its crushing- load. Baker says " the safe bearing power of rock is certainly not less than one-tenth of the ultimate crushing strength of cubes. That is to say, the safe bearing power of solid rock is not less than 18 tons per square foot for the softest rock, and 180 for the strongest. It is safe to say that almost any rock, from the hardness of granite to that of a soft crumbling stone easily worn by exposure to the weather or to running water, when well bedded will bear the heaviest load that can be brought upon it by any masonry construction." Rankine gives the average of ordinary cases as 20,000 pounds per square foot on rock foundations. Later in this chapter will be found a table that gives the crushing load in pounds per square inch for most of the substances used in foundations and building-walls. Foundations on Sand or Gravel. Strong gravel makes one of the best bottoms to build on; it is easily leveled, is almost incompressible, and is not affected by exposure to the atmosphere. Sand confined so that it cannot escape forms an excellent foundation, and is nearly incompressible. It has no cohesion, and great care must be used in preparing it for a foundation. Surface water must be kept from running into earth foundation beds, and the beds themselves must be well-drained and below frost-line. Baker says that a rather thick bed of sand or gravel, well protected from running water, will safely bear a load of 8 to 10 tons per square foot. Of course the area of the surface must be proportioned to the weight of the superstructure, and to the bearing resistance of the material, and for this reason it is common practice to spread the subfoundation to give it the proper area. Rankine sjives 2,500 to 3,500 lbs. per square foot as the greatest allowable pressure on firm earths. Foundation on Clay. A good stiff clay makes a very good foundation bed, and will support great Aveight if care is taken in its preparation. Water must be kept away from it, and the foundation level must be below the frost-line. The less clay is exposed to the atmosphere the better will be the result. Baker gives as safe bearing power for clay 3,000 or 4,000 pounds per square foot. Gaudard says a stiff clay will support in safety 5,500 to 11,000 pounds per square foot. FOUNDATIONS. 793 foundation on Soft Earth* Where the earth is too soft to support the superstructure, the trench is ; excavated to a considerable width, and to a considerable depth below the frost-line ; then a bed is prepared of stones, sand, or concrete, the latter 1 being most in use to-day. In fact, it is a common thing to cover the whole .; area of the basement of large power stations with a heavy layer of concrete 'I of a thickness sufficient to sustain not only the building-walls, but all ma^ 1. chine foundations. I Sand makes a good foundation bed over soft earth, if the earth is of a .quality that will retain the sand in position. Sand may be rammed in 19-mcli layers in a soft earth trench, or it can be used as piles instead of ,1 wooden ones, by boring holes 6 or 8 inches in diameter and say six feet deep ■ and ramming the sand in wet. It is necessary to cover the surface with .1 planking or concrete to prevent the earth pressing upward. Alluvial soils J are considered by Baker safe under a load of one-half to one ton per square J foot. M- ! foundation on files. When the earth is unsuitable in nature to support foundations, it is com- Imon to drive piles, on the tops of which the foundation is then built. II When possible the piles are driven to bed rock, otherwise they are made of fcsuch length and used in such number as to support the superstructure by ■ reason of the friction of their surfaces in the soil. Where the soil is quite iisoft it is also common to drive piles in large number all over the basement I area iu order to consolidate the earth, and make all parts of a better bearing I" quality. Piles must be driven and cut off below the water level, and a grillage of heavy timbers or a layer of broken stone and a capping of concrete must be placed on top of them for supporting the foundation. ■ The woods most used for piles are spruce and hemlock in soft or medium- § soft soils, or when they are to be always under water, hard pine, elm, and I beech in firmer soils, and oak in compact soils. When piles are liable to be I alternately wet and dry, white oak or yellow pine should be employed. I Piles should not be less than 10 inches in diameter at the small end, nor l| more than 14 inches at the large end. They should be straight-grained, and have the bark removed. The point is frequently shod with an iron shoe, to prevent the pile from splitting, and the head is hooped with an iron band to prevent splitting or brooming. Safe load on Piles. Rankine gives as safe loads on piles 1,000 pounds per square inch of head, If driven to firm ground; 200 pounds, if in soft earth, and supported by friction. Major Sanders, IT. S. Engineers, gives the following rule for finding the safe load for a wooden pile driven until each blow drives it short and nearly equal distances: c, - , , . , Weight of hammer in pounds x fall in inches Safe load m pounds = — ^ : — = 5— = , — ; — rrr, ^ 8 X inches driven by last blow Trautwine's rule is as follows : _ 3VFall in feet x Lbs. wgt. of hammer x .023 Extreme load in gross tons = - inches driven by last blow -4- 1 He recommends as safe load one-half the extreme load where driven in firm soils, and one-sixth when driven in soft earths or mud. The last blow should be delivered on solid wood, and not on the " broomed " head. Piles under Trinity Church, Boston, support two tons each. Piles under the bridge over the Missouri River at Bismarck, Dakota, were driven into sand to a depth of 32 feet, and each sustained a load of 20 tons. A pile under an elevator at Buffalo, N. Y., driven into the soil to a depth of 18 feet, sustained a load of 35 tons. 794 FOUNDATIONS AND STRUCTURAL MATERIALS. Arrangement of Piles. Under walls of a building piles are arranged in rows of two or three, spaced 24 inches or 30 inches on centers. Under piers or machine founda- tions they are arranged in groups, the distance apart being determined by the weight to be supported, but usually, as above, from two to three feet apart on centers. Concrete foundation Bed. As mentioned in a previous paragraph, concrete is now used to a very great extent for foundation beds, not only in soft earths, but to level up all kinds of foundation beds. Good proportions are by measure, using Portland cement: Cement, 1 part, Coarse sand, 2 parts, Broken stone, 5 parts. Only hard and sharp broken stone that will pass through a 1£- or 2-inch ring should be used ; and the ingredients should be thoroughly mixed dry, and after mixing, add just as little water as will fully wet the material. Concrete should be placed carefully. It is never at its best when dropped any distance into place. It should be thoroughly rammed in six- or nine- inch layers, and after setting the top of each layer should be cleaned, wet, and roughened before depositing another layer over it. It is common prac- tice to place side-hoards in trenches and foundation excavations in order to save concrete.- This is economical, but not good practice, if the earth is even moderately firm, as filling out the inequalities makes the foundation much firmer and steady in place. Weight of good concrete per cubic foot is 130 to 160 lbs. dry. Permissible JLoadw on foundation Beds. Piles, in firm soil, each pile 30,000 to 140,000 lbs. Piles in made ground, each pile, 4,000 " Clay, 4,000 " Coarse gravel and sand, 2,500 to 3,500 " Rock foundations, average, 20,000 " Concrete, 8,000 " New York City laws, no pile to be weighted with a load exceeding, 40,000 " New York City rule for solid nat- ural earth per superficial foot, 8,000 " Concrete foundations. One of the best foundations for engines or other heavy machinery is con structed wholly of concrete, rammed in a mold of planking. The mould can be made of any desired shape; the holding-down bolts placed by tem- plate, and the material rammed in layers not exceeding 12 inches thick. Brick foundations. Only the best hard-burned brick should be used for foundations, and they should be thoroughly wet before laying. To insure a thorough wetting, the bricks should be deposited in a tub of water. Bricks should be push placed in a good rich cement mortar. Grouting should never be used, as it takes too long to dry. Joints should be very small. A well constructed brick foundation will break as easily in the brick as at the joints after it has been built for some time. Stone foundations. Rubble stone foundations should start with large flat stones on the bot- tom. Care must be taken that all are well bedded in mortar, and that the work is well tied together by headers. MORTARS. 795 Dimension stone foundations are always laid out with the heavy and thick stones at the bottom, and gradually decreasing in height, layer by layer, to the top. A large cap-stone, or several if the size is too great for one, is often placed on top of the foundation. Care must be taken to bed each stone in cement mortar, so that the joints will be thin and yet leave all the spaces between the stones completely tilled with mortar to prevent any unequal strains on the stone. In all large foundations use plenty of headers; and if the backing or center is of rubble, see that all stones are well bedded, and the crevices rilled with spawls and cement. I-Beam Foundations. One of the best and now most common methods of constructing founda- tions for piers, walls, columus, etc., is the use of steel I-beams set in con- crete. Knowing the weight to be supported and the bearing value of the soil, excavation is made of the right dimensions to get the proper area of bearing, then I-beams of predetermined dimensions are laid parallel along the bottom, and beld in place with bolts from one beam to the next. Con- crete is rammed in all the spaces to a level with the top of the beams. An- other similar layer of beams is then laid on top of the first, and at right angles thereto, and the spaces also tilled with concrete. The column base, or footing course, is then set on the structure ready to receive the column. For method of calculation of dimensions of I-beams for use in foundations for piers and walls, the reader can consult the hand-book of the Carnegie Steel Company. ;viomt Alt*. lime JtEortar. Good proportions are : 1 measure or part quicklime, 3 measures of sand, well mixed, or tempered with clean water. Quantity required". — Trautwine. 20 cu. ft. sand and 4 cu. ft. of lime, making about 22J cu. ft. mortar, will lay 1,000 bricks with average coarse joints. Weig-lit. — 1 bbl. weighs 230 lbs. net, or 250 lbs. gross; 1 heaped bushel of lump lime weighs about 75 lbs.; 1 struck bushel ground quick lime, loose, weighs about 70 lbs. Average hardened mortar weighs about 105 to 115 lbs. per cu. ft. Tenacity. —Ordinary good lime mortar 6 months old has cohesive strength of from 15 to 30 lbs. per square inch. Adhesion to common l»ricks or rubble.- At 6 months old, 12 to 24 lbs. per sq. inch. Cement Mortar. Good proportions are: 1 measure cement, 2 measures sand, h measure water. The above is rich and strong, and for ordinary work will allow in- crease of sand to 3 or 4 measures. Quantity required. — Trautwine. 1 bbl. cement, 2 bbls. sand, will lay 1 cu. yd. of bricks with § inch joints or 1 cu. yard rubble masonry. WeigTit. — American Rosendale, ground, loose, average, 56 lbs. per cu. ft. " " U. S. struck bushel, 70 " " " " English Portland, 81 to 102 " " " " per struck bushel, 100 to 128 " " " " " " per bbl., 400 to 430 " " " " 796 FOUNDATIONS AND STRUCTURAL MATERIALS. Average Streng-th of Heat Cement after © I>ays in Water. Tensile, Lbs. per sq. in. Compress, Lbs, per sq. in. Compress, Tons per sq. ft. Portland, artificial . . . " Saylor's natural U. S. common hydraulic . 200 to 350 170 to 370 40 to 70 1400 to 2400 1100 to 1700 250 to 450 90 to 154 70 to 10 9 16 to 29 Cements are weakened by the addition of sand somewhat as shown in the following table : calling neat cement 1. Sand. 0 h 1 1* 2 3 4 5 6 Strength. 1 * \ .4 i .3 1 5 i JLdraesion to Bricks or Rul»l»le. Adhesion of cement, either neat or mixed with sand, will average about three-fourths the tensile strength of the mortar at the same age. S.OD A]V» CEMEHTT. Recommendations of Am. Soc. Civil Engineers. (Sand. — To be crushed quartz only. To pass, 1st sieve, 400 meshes per square inch. 2d " 900 Sand to pass the 400 mesh, but be caught by the 900 mesh, all finer parti- cles to be rejected. Portland. Cement. — For fineness, to pass, 1st sieve, 2500 meshes per square inch. 2d " 5476 3d " 10000 " " Should be stored in bulk for at least 21 days to air-slake and free it from lime, as lime swells the bulk, and if not removed is apt to crack the work. IMOJV AJ¥I» §TEEL. Iron, weig-lit of: cu. in. Cast, .2604 Lbs. Wrought^ .'2777 " a = sectional area wrought-iron bar. x =r weight per foot " cu. ft. 450 Lbs. 3x Steel, weig-lit of: _10rt X— g ■ cu. in. cu. ft. .2831 Lbs. 489.3 Lbs Cast Iron. Test. Bar an inch square, supported on edges 1 foot apart, must sustain 1 ton at center. WEIGHT OF FLAT ROLLED IRON. 797 h r, ; * S 0B -d fl .■s ft s 3 ft - • N .3 o 1 »_ N •" i ^ « w © s 3 S2 ^ W- - 2 a s£ g H 5~ M *> g M ^ "* © 3" P & ^ s ! M £ 1 .3 .990 1.98 2.97 3.96 4.95 5.94 6.93 7.92 8.91 9.90 10.89 11.88 12.86 13.85 14.84 15.83 16.82 17.81 18.80 19.79 20.78 21.77 22.76 23.75 24.74 25.73 26.72 27.71 28.70 30^68 31.67 3 5 .938 1.88 2.81 3.75 4.69 5.63 6.56 7.50 8.44 9.38 10.31 11.25 12.19 13.13 14.06 15.00 15.94 16.88 17.81 18.75 19.69 20.63 21.56 22.50 23.44 24.38 25.31 26.25 27.19 28.13 29.06 30.00 d .885 1.77 2.66 3.54 4.43 5.31 6.20 7.08 7.78 8.85 9.74 10.63 11.51 12.40 13.28 14.17 15.05 15.94 16.88 17.71 18.59 19.48 20.36 21.25 22.14 23.02 23.91 24.79 25. (IS 26.56 27.45 28.33 ■* .833 1.67 2.50 3.33 4.17 5.00 5.83 6.67 7.50 8.33 9.17 10.00 10.83 11.67 12.50 13.33 14.17 15.00 15.83 16.67 17.50 18.33 19.17 20.00 20.83 21.67 22.50 23.33 24.17 25.00 25.83 26.67 CO .781 1.56 2.34 3.13 3.91 4.69 5.47 6.25 7.03 7.81 8.59 9.38 10.16 10.94 11.72 12.50 13.28 14.06 14.84 15.63 16.41 17.19 17.97 18.75 19.53 20.31 21.09 21.88 22.66 23.44 24.22 25.00 5 .729 1.46 2.19 2.92 3.65 4.38 5.10 5.83 6.56 7.29 8.02 8.75 9.48 10.21 10.94 11.67 12.40 13.13 13.85 14.58 15.31 16.04 16.77 17.50 18.23 18.96 19.69 20.42 21.15 21.88 22.60 23.33 s* .677 1.35 2.03 2.71 3.39 4.06 4.74 5.42 6.09 6.77 7.45 8.13 8.80 9.48 0.16 0.83 1.51 2.19 2.86 L3.54 4.22 L4.90 5.57 16.25 6.93 7.60 18.28 18.96 19.64 20.31 20.99 21.67 i .625 1.25 1.88 2.50 3.13 3.75 4.38 5.00 5.63 6.25 6.88 7.50 8.13 8.75 9.38 0.00 0.63 1.25 1.88 2.50 3.13 3.75 4.38 5.00 5.63 6.25 6.88 7.50 8.13 18.75 9.38 20.00 .3 .573 1.15 1.72 2.29 2.86 3.44 4.01 4.58 5.16 5.73 6.30 6.88 7.45 8.02 8.59 9.17 9.74 10.31 10.89 11.46 12.03 12.60 13.18 13.75 14.32 14.90 15.47 16.04 16.61 17.19 17.76 18.33 n .521 1.04 1.56 2.08 2.60 3.13 3.65 4.17 4.69 5.21 5.73 6.25 6.77 7.29 7.81 8.33 8.85 9.38 9.90 10.42 10.94 11.46 11.98 12.50 13.02 13.54 14.06 14.58 15.10 15.63 16.15 16.67 | .469 .938 1.41 1.88 2.34 2.81 3.28 3.75 4.22 4.69 5.16 5.63 6.09 6.56 7.03 7.50 7.97 8.44 8.91 9.38 9.84 10.31 10.78 11.25 1.1.72 12.19 12.66 13.13 13.59 14.06 14.53 15.00 ifl W Cl O '/. 1- C " : 1 O X i- l". " C [- :o o t> » o t- p !-h c 1 co ^ 5 in p 1^ x p p p — 0 1 : : p p x p r-< co in p us^rS^ x m co o x 1- co © x ■ >n co — x 1- co o n c c © m o m c x 1- p p co ci t-h p x L- p in co 0 1 1—, p l- p ci p 1- in oi o r-5 cb in 1-- ci r-5 oo m' ©' x ©' ci-t'o x' o w t- h l* cc 010c — -H S i-h th 01 oi ci ci oi co co co -* ■* ** m m « t- -+< t-h 00 m co © 1- -t< h x 10 oi 0-.. ocooofiBO-joont; t- in co © x © -r — - 1 - -r 0 1 © 1 - >o co x -1- © m oohb r-5 cd id t-: 06 © ci -r o t- © ~ co -3 © x j-j £ x ci o 0 co jo o co ^ -* •# m m t> co o n n © t- co 00 t- co © t- co © t- © co i~ o co c- © co p co © p co 0; p :: © © :: © © co © © © co p © co p p co r-5 cd in © 00 © -5 co in © x ©' -^ co m" © © co © ©' co' © © co ©co©mi-HX-H©©co©in^x-+(©coinx©coin'X© in th © -oi co cc © o © p -h i-.«co-i;oh 01 co p © l- x © r-5 cd "*' © t>5 ci © oi -+' in t- x ~ -- co' 1 0' x — -+ t- © co' © © HriHHHr- 1 CI 0)01 01 01 00 CO CO ■"■f^^lO © cm x cc © m »h t^ co x -t< © © 0 1 x co 10 t- x © 01 co m t- •+ © co x 01 1- 01 © — m co in © -+- x co oi — © © © co l- p i-< ci <# in t^ 00 © t-h 00 -+ ©' 1- co © 1-^ 00' © ©' 0 i lol^ © co © rlHrlHHHHCltlClflCl CO CO CC ■* ■* ■* 1-1 oi t-h co co -p -f m © © t>- x x © © © 1-1 co "**! in i* CO CI © © •* CO 01 p © -+ CO CI tSOlOtOnOl-; r-i ci -* in l- 06 ci T-.' oi -P m' © x © -5 ci m' «' i-icc-*in©x©© 00: 1^ c o cc 00 co ■* -# o©i*©in©©iHt-xx©©T-icicccc <=> © j— 1 i-H 1-1 ih i-( ci ci ci ci 00 00 cc co co -t< -+i m m m © © t- cc p © ci m co i-j ■# t>; © co © p 01 1.0 x -h © © 01 x -t< © © i-5 ci cd in" © t^ ci © i-H cc •* in © co ci © cd ©' 06 1-5 cd ©' ci 1 " M Cl CO CC CO CC Td m © m © m © in © m © 0 © m © m © © © © © © © © © ci m l- © 01 m t- © ci 101-0:1 p 1- © m © m © un m m © i-5 ci cd in © t^-' cd ©' ih ci cd in © t-^ 00 ©' ci in t- © ci in t~ ©" rH ri r->. rH HrtHH(Md(M!M«MWm-# OOOfflQC.OWOOCOCCOOt-t-Nt-tOOl-Ol.OOif-fM ci -fiqi-OH^ p t- p rH co p l- © t-h m p 00 i~ hoqw i-5 ci ci tj5 un t- cd ci © i-5 cd -r in ©' 1 - © r-' 00 ©' x -5 cc in cd T-H lH l-H rH l-H 1-1 1-H T-H ci ci oi 01 co co :: :: O CO 00 Ol N H O O l-O O i) OtlHOOfflCCh rH CI -t LO L-_ X © H :o -+ © t'-^ © © rH 00 p ci ci p f^ C i-5 ci cd "*" in © 00 ci ©' ri ci co -r" © 1--' cd © ci in t-^ ci ci -f oi in t-^ ci 1-" TH^HTjn^h-a^r.^i-'^-.iu.^c;!^; WEIGHT OF BARS OF IRON. 799 WEIGHTS OF lariBE AID HOOD BARS ©J? WROIGHT JLR09T I]¥ POmiDi raja JLJOTEAI, JFOOT. Iron weighing 480 lbs. per cubic foot. For steel add 2 per cent. © s U ID cS u 03 3 u s .5 • 03&P ^ si -3 fl jL A • Mi 03 3 . a . 2^ 12 8 A « II §a A $ S o SA SJ cS O PA C O gA «1 S3 co o M 9 co o Ao °1 tf o o£ |£ ei-i O oA 03 fl . © © b- © © b- cq © t>- co C 1 © © co d © co d © d d d d d © d d © d ::' © d ::' ©' d d d d c: © © © : I -C LO LO LO © © © L- L- I- X X -X © © © © © i-i 01 01 CO -* -V LO © © I- X X © C - — - .'J — — . .' w -^ - I l. 5 -*-*•* LO LO 1.0 LO © © ?J V3 % ~ P « d©looo©©loco©©©o: -M - © ::" ©' d oi x i.o — ft d © d oi x £ X © © © © ,-h i-i o 1 CO CO -)' ) - i 0 © - © : i :o iO i - © © — © io t - © © — co uo i - © © co c- © © 00 t-; q 10 q q q ci q © q © x . q q q cmhown W CO CO ■*"*'* ^ i'O i.0 l© © © © ft L- ft L~^ X X X © cK © 2S? -I 01 CO © LO 1-0 LO t~ CO CO © I © © CO © I- q rH 00 q CO ©©CI ©•# © -f 10 i- q x 1 0 0 o ro -tf •* -t< ■* ir ooLodddftftftft^cc-Sb©;-:-?; )©©©©©©© ©©©©©©©©©©©OO© > q © q © q © q © q © q © q © q © q © © © © 0 CO CO -# -* "# i~~ © ci lo h c oi i o i ~ © oi rdi'idi^oiA •# o o lo lo © © © © ft ft ft x x © © 5 © ©©©©©©©©© © d ~ o"i © ':: © d i"o 3 © CO 00 t— © 1-0 -t1 CO CO 01 — © © X X I- © 10 CO -h © © c~ lo o i- © q © © oi q q q q q © oi q q q q q co © o © © oi ~ q I" © q © io © io o lo © io © io © lo © © © © © © © © © 5 © 5 9 2 2 = _ o i q q q : i i- i - q oi io i- © oi lo i- © oi lo © lo © lo © lo © lo © lo © lo — l. , _ q ft © oi lo ft © o t lo © lo © lo © 2 " 2 i d d d Lo d © — 3 oi co lo d ft © ci i o" oi co rJH Tfi •* LO LO LO LO © © W © b- t- t- 2 "2 -2 .:2 Jr rn lo © co oi © © -h oo co i- -< 10 co oi © ©. i- ;- q © © © © i-i i-i oi oi q q lo lo_ © t- ©' i-H Oi CO lO* ©' I - d © © — ' CO 10 ft © rH o- 01 01 01 01 0 1 0 1 M 0 1 0 1 CO CO CO CO CO © i* Ti +< 01 © 00 t^ LO CO 01 © o © © © q o i co q lo o ft © oi d © x b o_i ti -r< i.O LO i.O LO LO © © - ^ :" '.: 2 !2 '~ S c- OI>-©t— CO©t— CO©t- o — oo © co © © q © « v io © d © ih d lo c: ■ i ; i : i oi :0 co co CO c 0 00 t- CO © t- CO © : © © o © © co © c © — d 1.0 d x © o co © © © ^h -r-i oi © d -+i lo io © © © lo © ip © lo © in © ft oJOH oi d lo d ft oioqeococococococc © 0 ! CO lO t- CO © O © q q oi q - io r lo lo d d © ft ft t- 0 C- CO © 01 OO IO t~ 00 © CO t- OtONO - q lo © q q oi © © io_ co q © oq © lo o © © © © © -3 -^ oi oi co tjh lo qq©q©Loqq© OA Ol Ol ?1 01 OI 01 u © ^H Ol © CO •* LO © t- CO CO © © — 0 I CO -o -n LO t- 0O O Ol CO LO C- CO © OI CO IO C- CO © „to j lo l- © q co i.o i - © i-h co lo l- © oi -t* © x © oi © © i o © co t ; lo © q oq oi © © q ' © © © © © i-3 --< oi oi c © © i-i oi co -I lo © r- oo © © oi -t< co co © oi ■* © OI Ol Ol 01 Ol CO CO © © © tP tJ< •* ^ © oo © oi -f cc: GAUGE FOR SHEET AND PLATE IRON. 801 HT.OUARD «A.UCJE EOR SHEET -AJ¥» PJLAXJE IROI AID STEEJL. 1S03. 6 «H ^ ^ ^; M CD *- « O H "-* "£ • l| I 0*03.2 "II ^3* co S Pi t per Meter unds upois. X xn ee <2.J •3 *< m£,2 CO .3 1 H*3 0 F * f 0000000 1-2 0.5 12.7 320 20. 9.072 97.65 215.28 000000 15-32 0.46875 11.90625 300 18.75 8.505 91.55 201.82 00000 7-16 0.4375 11.1125 280 17.50 7.938 85.44 188.37 0000 13-32 0.40625 10.31875 260 16.25 7.371 79.33 174.91 000 3-8 0.375 9.525 240 15. 6.804 73.24 161.46 00 11-32 0.34375 8.73125 220 13.75 6.237 67.13 148.00 0 5-16 0.3125 7.9375 200 12.50 5.67 61.03 134.55 1 9-32 0.28125 7.14375 180 11.25 5.103 54.93 121.09 2 17-64 0.265625 6.746875 170 10.625 4.819 51.88 114.37 3 1-4 0.25 6.35 160 10. 4.536 48.82 107.64 4 15-64 0.234375 5.953125 150 9.375 4.252 45.77 100.91 5 7-32 0.21875 5.55625 140 8.75 3.969 42.72 94.18 6 12-64 0.203125 5.159375 130 8.125 3.685 39.67 87.45 7 3-16 0.1875 4.7625 120 7.5 3.402 36.62 80.72 8 11-64 0.171875 4.365625 110 6.875 3.118 33.57 74.00 9 5-32 0.15625 3.96875 100 6.25 2.835 30.52 67.27 10 9-64 0.140625 3.571875 90 5.625 2.552 27.46 60.55 11 1-8 0.125 3.175 80 5. 2.268 24.41 53.82 12 7-64 0.109375 2.778125 70 4.375 1.984 21.36 47.09 13 3-32 0.09375 2.38125 60 3,75 1.701 18.31 40.36 14 5-64 0.078125 1.984375 50 3.125 • 1.417 15.26 33.64 15 9-128 0.0703125 1.7859375 45 2.8125 1.276 13.73 30.27 16 1-16 0.0625 1.5875 40 2.5 1.134 12.21 26.91 17 9-160 0.05625 1.42875 36 2.25 1.021 10.99 24.22 18 1-20 0.05 1.27 32 2. 0.9072 9.765 21.53 19 7-160 0.04375 1.11125 28 1.75 0.7938 8.544 18.84 20 3-80 0.0375 0.9525 24 1.50 0.0804 7.324 16.15 21 11-320 0.034375 0.873125 22 1.375 0.6237 6.713 14.80 22 1-32 0.03125 0.793750 20 1.25 0.567 6.103 13.46 23 9-320 0.028125 0.714375 18 1.125 0.5103 5.493 12.11 24 1-10 0.025 0.635 16 1. 0.4536 4.882 10.76 25 7-320 0.021875 0.555625 14 0.875 (i.. •;:»<;<) 4.272 9.42 26 3-160 0.01875 0.47625 12 0.75 0.3402 3.662 8.07 27 11-640 0.0171875 0.4365625 11 0.6875 0.3119 3.357 7.40 28 1-64 0.015625 0.396875 10 0.625 0.2835 3.052 6.73 29 9-640 0.0140625 0.3571875 9 0.5625 0.2551 2.746 6.05 30 1-80 0.0125 0.3175 8 0.5 0.2268 2.441 5.38 31 7-640 0.0109375 0.2778125 7 0.4375 0.1984 2.136 4.71 32 13-1280 0.01015625 0.25796875 6* 0.40625 0.1843 1.983 4.37 33 3-320 0.009375 0.238125 6 0.375 0.1701 1.831 4.04 34 11-1280 0.00859375 0.21828125 5} 0.34375 0.1559 1.678 3.70 35 5-640 0.0078125 0.1984375 5 0.3125 0.1417 1.526 3.36 36 9-1280 0.00703125 0.17859375 ih 0.28125 0.1276 1.373 3.03 37 17-2560 0.006640625 0.168671875 4 0.2(55625 0.1205 1.297 2.87 38 1-160 0.00625 0.15875 4 0.25 0.1134 1.221 2.69 802 FOUNDATIONS AND STRUCTURAL MATERIALS. COLUMlfi, PILLARN, Oil STJRUTS. Hodgfkinson'ii Formula for Columns, P = crushing weight in pounds ; d = exterior diameter in inches ; dl - interior diameter in inches ; L = length in feet. Kind of Columns. Both ends rounded, the length of the column exceeding 15 times its diameter. Both ends flat, the length of the column exceeding 30 times its diameter. Solid cylindrical col- umns of cast iron . Hollow cylindrical columns of cast Solid cylindrical col- : unms of wrought iron ' Solid square pillar of ] Dantzic oak (dry) . J P = 33,380 P = 29,120 P = 95,850 - $3.76 (J 3.70 P — 98,920 - rf3.5S _ rfi3.B $3.55 W- These formula? apply only to cases of breakage caused by bending rather than mere crushing. Where the column is short, or say five times its diam- eter in length, then the following formula applies. Let P z= value given in preceding formulae, K— transverse section of column in square inches, C= ultimate compressive resistance of the material, W = crushing strength of the column. Then P CK P + l CK' Hodgkinson's experiments were made upon columns the longest of which for cast iron was 60^ inches, and for wrought iron 90| inches. The following are some of his conclusions : 1. In all long pillars of the same dimensions, when the force is applied in the direction of the axis, the strength of one which has Hat ends is about three times as great as one with rounded ends. 2. The strength of a pillar with one end rounded and the other flat is an arithmetical mean between the two given in the preceding case of the same dimensions. 3. The strength of a pillar having both ends firmly fixed is the same as one of half the length with both ends rounded. 4. The strength of a pillar is not increased more than one-seventh by en- larging it at the middle. Gordon's formulae, deduced from Hodgkinson's experiments, are more generally used than Hodgkinson's own. They are : Columns with both ends fixed or flat P — P' 1 + a- Columns with one end flat, the other end round, P =z fS 72 Columns with both ends round or hinged, P - Z2' 1+ 4a -£ STRENGTH OF MATERIALS. 803 S= area of cross-section in inches ; P = ultimate resistance of column in pounds ; f zz crushing strength of the material in pounds per square inch ; . . ,. ,. ,. . . . , Moment of inertia r =s least radius or gyration, in inches, r2 = — : : area of section I = length of column in inches ; a = a coefficient depending upon the material ; / and a are usually taken as constants ; they are really empirical varia- bles, dependent upon the dimensions and character of the column as well as upon the material. (Burr.) For solid wrought-iron columns, values commonly taken are : /= 36,000 to 40,000 ; a r- New York City Building Laws 1897-1898 give the following values for/: Cast iron f — 80,000 lbs. Rolled steel .... f = 48,000 lbs. Wrought or rolled iron / = 40,000 lbs. American oak . . . f= 6,000 lbs. Pitch or Georgia pine . /■=. 5,000 lbs. White pine and spruce f = 3,500 lbs. For solid cast-iron column s,/ = 80,000, a = 0400* 80 000 For hollow cast-iron columns, fixed ends, p = j9, I = length and 1 + 800 -j d =r diameter in the same unit, andp = strength in lbs. per square inch. Sir Benjamin Baker gives, For mild steel / = 67,000 lbs., a = . For strong steel /= 114,000 lbs., a — tt-ttja- STRENGTH OF IHATERUI§. The terms stress and strain are generally used synonymously, authorities differing as to which is the proper use. Merriman defines st?-ess as a force which acts in the interior of a body, and resists the external forces which tend to change its shape. A deformation is the amount of change of shape of a body caused by the stress. * The word strain is often used as synony- mous with stress, and sometimes it is also used to designate the deforma- tion. Merriman gives the following general laws for simple tension or compression, as having been established by experiment. a. When a small stress is applied to a body, a small deformation is pro- duced, and on the removal of the stress the body springs back to its original form. For small stresses, then, materials may be regarded as perfectly elastic. b. Under small stresses the deformations are approximately proportional to the forces or stresses which produce them, and also approximately pro- portional to the length of the bar or body. c. When the stress is great enough, a deformation is produced which is partly permanent; that is, the body does not spring back entirely to its original form on removal of the stress. This permanent part is termed a set. In such cases the deformations are not proportional to the stress. d. When the stress is greater still, the deformation rapidly increases, and the body finally ruptures. e. A sudden shock or stress is more injurious than a steady stress, or than a stress gradually applied. 804 FOUNDATIONS AND STRUCTURAL MATERIALS. .Elastic liimit. The elastic limit of a material under test for tensile strength is defined as the point where the rate of stretch begins to increase, or where the defor- mations cease to be proportional to the stresses, and the body loses its power to return completely to its former dimensions when the stress is re- moved. Modulus of Elasticity. The modulus or coefficient of elasticity is the term expressing the relation of the amount of extension or compression of a material under stress to the load producing that stress or deformation. It is the load per unit of section divided by the extension per unit of length. If P == applied load, k = sectional area of piece, I = length of the part extended, A. = amount of extension, M = modulus of elasticity, M- P ' k— Pl k ' l~ kk Following are the Moduli of elasticity for various materials. Brass, cast 9,170,000 wire 14,230,000 Copper 15,000,000 to 18,000,000. Lead 1,000,000 Tin, cast 4,600,000 Iron, cast 12,000,000 to 27,000,000 (?) Iron, wrought 22,000,000 to 29,000,000 Steel 26,000,000 to 32,000,000 Marble 25,000,000 Slate 14,500,000 Glass 8,000,000 Ash 1,600,000 Beech 1,300,000 Birch 1,250,000 to 1,500,000 Fir 869,000 to 2,191,000 Oak 974,000 to 2,283,000 Teak 2,414,000 Walnut 306,000 Pine, long-leaf (butt-logs) . . 1,119,200 to 3,117,000 Average, 1,926,00 Factor of Safety. This may be defined as the factor by which the breaking strength of a material is divided to obtain a safe working-stress. The factor of safety is sometimes a rather indefinite quantity, owing to lack of information as to the strength of materials, and it is now becoming common to name a defi- nite stress which is substantially the result of dividing the average strengths by a factor. The following factors are found in the " Laws Relating to Building in New York City," 1897-1898. For beams, girders, and pieces subject to transverse strains, factor of safety — 4. For wrought-iron or rolled-steel posts, columns, or other vertical sup- ports, 4. For other materials subject to a compressive stx-ain, 5. For tie-rods, tie-beams, and other pieces subject to tensile strain, 6. MOMEUfT OF IHfEMTIA. The moment of inertia of a body about any axis, is the sum of the products of the mass of each particle of the body, into the square of its (least) dis- tance from the axis. MOMENT OF INERTIA. 805 radius or cnritATi:©]*. The radius of gyration of a section is the square root of the quotient of the moment of inertia, divided by the area of the section, or Radius of gyrations /Moment of inertia V Area of section. The radius of gyration of a solid about an axis is equal to the . /Moment of Inertia V Mass of the Solid Use in the formulae for Streng-th of Girders and Columns. The strength of sections to resist strains, either as girders or as columns, depends on the form of the section and its area, and the property of the section which forms the basis of the constants used in the formulae for strength of girders and columns to express the effect of the form, is its moment of inertia about its neiitral axis. Thus the moment of resistance of any section to transverse bending is its moment of inertia divided by the distance from the neutral axis to the fibers farthest removed from the axis ; or ...- .:, . , Moment of inertia ., i" Moment of resistance r= • Distance of extreme fiber from axis" e ' moment of Inertia of Compound Shapes. (Pencoyd Iron "Works.) The moment of inertia of any section about any axis is equal to the I about a parallel axis passing through its center of gravity -f- (the area of the sec- tion X the square of the distance between the axes). By this rule, the moments of inertia or radii of gyration of any single sec- tions being known, corresponding values may be obtained for any combina- tion of these sections. Radius of gyration of Compound Shapes. In the case of a pair of any shape without a web the value of R can always be found without considering the moment of inertia. The radius of gyration for any section round an axis parallel to another axis passing through its center of gravity is found as follows : Let r = radius of gyration around axis through center of gravity ; R =: radius of gyration around another axis parallel to above ; d =. distance be- tween axes : R — Vrf2 + r When r is small, R may be taken as equal to d without material error. ELEHXWTI OF USUAX SECTIONS. Moments refer to horizontal axis through center of gravity. This table is intended for convenient application where extreme accuracy is not impor- tant. Some of the terms are only approximate ; those marked * are cor- rect. Values for radius of gyration in flanged beams apply to standard miiiimum sections only. A = area of section ; b = breadth ; h = depth ; D = diameter. 806 FOUNDATIONS AND STRUCTURAL MATERIALS. Shape of Section. Moment of Inertia. Moment of Resistance. Square of Least Radius of Gyration. Least Radius of Gyration. bh** 12 bh** 6 /Least \2* V Side ) *"" Solid Rect- angle. Least side* fc 12 F e-6— » Hollow Rect- angle. bW—bJtf * bW — bJi** h* + V * _. h+h1 12 6h 12 4.89 -&-* Solid Circle. AD** 16 AD* 8 D** "16" D* 4 - - D- -H Hollow Circle A, area of large section ; a, area of small section. AD*— ad* AD*— ad* Z>2+rt2* 16 D + d 16 8D 5.64 -> / \_ "^ Solid Triangle. bhs 36 bh* 24 The least of the two: h2 b* 18 °r 24 The leist of the two : h b -6— H IM °r 479 Even Angle. Ah* 10.2 Ah 7.2 63 25 b ~5~ -rft — o Uneven Angle Ah* 9.5 Ah 6.5 (hb)* /t6 fcr- 13(/i2+62) 2.6 (h -4- 6) HB Even Cross. Ah* 19 Ah 9.5 /t2 22.5 ft 4.74 M Even Tee. Ah* ILT 8 b* 22.5 6 4J4 ^m I-Beam. Ah* 6.66 3.2 62 21 b 4.58 f i Channel. Ah* 7.34 Ah 3.67 62 12.5 3.54 .^ 11 I 4 fe) Deck Beam. Ah* 6.9 4 b* 36.5 6 6 Distance of base from center of gravity, solid trii -— — ; uneven angle, -— ; even tee, -^; deck beam, given in the table, ^ or ~z ■ even angle, 3.3' ELEMENTS OF USUAL SECTIONS. 807 Solid Cast-iron Columns. Table, based on Hodgkinson's formula (gross tons). The figures are one-tenth of the breaking weight in tons, for solid col- umns, ends flat and fixed. .5 S3 Length of Column n Feet. 6. 8. 10. 12. 14. 16. 18. 20. 25. ii .82 .50 .34 .25 .19 .15 .13 .11 .07 if 1.43 .87 .60 .44 .34 .27 .22 .18 .13 2 2.31 1.41 .97 .71 .55 .44 .36 .30 .20 21 1 3.52 2.16 1.48 1.08 .83 .67 .54 .46 .31 5.15 3.16 2.16 1.58 1.22 .97 .80 .66 .56 7.26 4.45 3.05 2.23 1.72 1.37 1.12 .94 .64 3 9.93 6.09 4.17 3.06 2.35 1.87 1.53 1.28 .88 3J 17.29 10.60 7.26 5.32 4.10 3.26 2.67 2.23 1.53 4 27.96 17.15 11.73 8.61 6.62 5.28 4.32 3.61 2.47 4* 42.73 26.20 17.93 13.15 10.12 8.07 6.60 5.52 3.78 5 62.44 38.29 26.20 19.22 14.79 11.79 9.65 8.06 5.52 5* 88.00 53.97 36.93 27.09 20.84 16.61 13.60 11.37 7.78 6" 120.4 73.82 50.51 37.05 28.51 22.72 18.60 15.55 10.64 6J 160.6 98.47 67.38 49.43 38.03 30.31 24.81 20.74 14.19 7 209.7 128.6 87.98 64.53 49.66 39.57 32.30 27.08 18.53 7i 268.8 164.8 112.8 82.73 . 63.66 50.73 41.53 34.72 23.76 8 339.1 207.9 142.3 104.4 80.31 64.00 52.39 43.80 29.97 8i 421.8 258.6 177.0 129.8 99.90 79.61 65.16 54.48 37.28 9 518.2 317.7 217.4 159.5 122.7 97.80 80.05 66.92 45.80 9* 629.5 386.0 264.2 193.8 149.1 118.8 97.25 81.70 55.64 10 757.2 464.3 317.7 233.1 179.3 142.9 117.0 97.79 66.92 lOJ 902.6 553.5 378.7 277.8 213.8 170.3 139.4 116.6 79.77 11 1067.1 654.4 447.8 328.5 252.7 201.4 164.9 137.8 94.31 11* 1252.3 767.9 525.5 385.4 296.6 236.4 193.5 161.7 110.7 12 1459.6 895.1 612.5 449.3 345.7 275.5 225.5 188.5 129.0 Where the length is less than 30 diameters, Strength in tons of short columns = SC 10£+fC" S being the strength given in the above table, and C= 49 times the sec- tional area of the metal in inches. Hollow Columns. The strength nearly equals the difference between that of two solid col- umns, the diameters of which are equal to the external and internal diam- eters of the hollow one. More recent experiments carried out by tbe Building Department of New York City on full-size cast-iron columns, and other tests made at the Watertown Arsenal on cast-iron mill columns, show Gordon's formula, based on Hodgkinson's experiments, to give altogether too high results. The following table, from results of the New York Building Department tests, as published in the Engineering N'eics, January 13-20, 1898, sIioav actual results on columns such as* are constantly used in buildings. Applying Gordon's formula to the same columns gives the following as the breaking load per square inch. For 15-inch columns, 57,000 lbs.; for 8-inch and 6-inch columns, 40,000 lbs., all of Avhich are much too high, as shown by the table. Prof. Lanza gives the average of 11 columns in the Watertown tests as 29,600 pounds per square inch, and recommends that 5,000 pounds per square inch be used as the maximum safe load for crushing strength. 808 FOUNDATIONS AND STRUCTURAL MATERIALS. Tests of Cast-iron Colun I11S. Thickness. Breaking Load. Diam. Inches. Max. Min. Average. Pounds. Pounds per sq. in. 1 15 1 1 1,356,000 30,8300 2 15 Its 1 li 1,330,000 27,700 3 15 1 li 1,198,000 24,900 4 15| 1 li 1,246,000 25,200 5 15 3* 1 itt 1,632,000 32,100 6 15 li 1t3S 2,082,000+ 40,400+ 7 7| to 81 li f 1 651,00 31,900 8 8 1* 1 l£ 612,800 26,800 9 6A 1* H 1& 400,000 22,700 10 653s ^1 1A 1* 455,200 26,300 Ultimate Streng-th of Hollow, Cylindrical VTroug-ht and Cast-iron Columns, when fixed at the Ends. (Pottsville Iron and Steel Co.) f Computed by Gordon's formula, p ■=. — - — - 1 + Cl p = Ultimate strength in lbs. per square inch ; I = Length of column, * ) , ,, . xmita- h = Diameter of column, \ both m same umt8» f f 40,000 lbs. for wrought iron; ) J ~~ \ 80,000 lbs. for cast iron; } C = 1/3000 for wrought iron, and 1/800 for cast iron. _ . . 80.000 For cast iron, p — 1 + For wrought iron, p = 40,000 800 V h ) +— i-V ^ 3,000 V hj Hollow Cylindrical Columns. Ratio of Maximum Load per sq. in. Safe Load pei • Square Inch. Length to Diameter. Z h Cast Iron. Wrought Iron. Cast Iron, Factor of 6. Wrought Iron, Factor of 4. 8 74075 39164 12346 9791 10 71110 38710 11851 9677 12 67796 38168 11299 9542 14 64256 37546 10709 9386 16 60606 36854 10101 9213 18 56938 36100 9489 9025 20 53332 35294 8889 8823 22 49845 34442 8307 8610 24 46510 33556 7751 8389 26 43360 32642 7226 8161 28 40404 31712 6734 7928 30 37646 30768 6274 7692 ELEMENTS OF USUAL SECTIONS. 809 Hollow Cylindrical Columns.- Ratio of Maximum Load per Sq. In. Safe Load pei Square Inch. Length to Diameter. 1 h Cast Iron. Wrought Iron. Cast Iron, Factor of 6. Wrought Iron, Factor of 4. 32 35088 29820 5848 7455 34 32718 2S874 5453 7218 36 30584 27932 5097 6983 38 28520 27002 4753 6750 40 26666 26086 4444 6522 42 21962 25188 41G0 6297 44 23396 24310 3899 6077 46 21946 23454 3658 5863 48 2061S 22620 3430 5655 50 19392 21818 3262 5454 52 18282 21036 3047 5259 54 17222 20284 2S70 5071 56 16260 19556 2710 4889 58 153G8 18856 2361 4714 60 14544 181S0 2424 4545 intimate Streng-tla of Wroug^ht-iron ColunniM. p — ultimate strength per square inch; I— length of column in inches; r — least radius of gyration in inches. For square end-bearings, p =- 1 + mo\rJ For one pin and one square bearing, For two pin bearings, 40000 40000 T 30000 V*"/ 40000 1- / I \ 20000 \rj For safe working-load on these columns use a factor of 4 when used in buildings, or when subjected to dead load only; but wben used in bridges the factor should be 5. Wrought-Iron Columns. Ultimate Strength in Lbs. Safe Strength in Lbs. per 1 per Square In ch. I r Square Inch — Factor of 5. r Square Pin and Pin Square Pin and Pin Ends. Sq. End. Ends. Ends. 7888 Sq.End. Ends. 10 39944 39866 39800 10 7973 7960 15 39776 39702 39554 15 7955 7940 7911 20 39604 39472 39214 20 7021 7894 7843 25 39384 39182 38788 25 7887 7836 7758 30 39118 38S34 38278 30 7821 7767 7656 35 38810 38430 37690 35 7762 7686 7538 40 38460 37974 37036 40 7692 7595 7407 45 38072 37470 36322 45 7614 7494 7264 50 37646 36928 35525 50 7529 7386 7105 55 37186 36336 34744 55 7437 7267 6949 60 36697 35714 33898 60 7339 7143 6780 65 3bl82 34478 33024 65 7236 6896 6605 70 3o634 34384 32128 70 7127 6877 6426 75 350^6 33682 31218 75 7015 6736 6244 80 3*482 32066 30288 80 6896 6593 6058 85 33883 32236 29384 85 6777 6447 5877 90 33264 31496 28470 90 6653 6299 5694 95 32636 30750 27562 95 6527 6150 5512 100 32000 30U00 2b666 100 6400 6000 5333 105 31357 29250 25786 105 6271 5850 5157 810 FOUNDATIONS AND STRUCTURAL MATERIALS. TMASrSVEMSE S1REXGTH. Transverse strength of bars of rectangular section is found to vary di- rectly as the breadth of the specimen tested, as the square of its depth, and inversely as its length. The deflection under load varies as the cube of the length, and inversely as the breadth and as the cube of the depth. Alge- braically, if S = the strength and D the deflection, I the length, b the breadth, and d the depth, S varies as —r- and JJ varies as j-^t. I bd3 To reduce the strength of pieces of various sizes to a common standard, the term modulus of rupture (li) is used. Its value is obtained by experi- ment on a bar of rectangular section supported at the ends and loaded in the middle, and substituting numerical values in the following formula : 2 bd* in which P = the breaking load in pounds, I = the length in inches, b the breadth, and d the depth. fundamental Forinnla' for flexure of ISeain*. (Merriman.) Resisting shear ~ vertical shear ; Resisting moment = bending moment ; Sum of tensile stresses = sum of compressive stresses ; Resisting shear = algebraic sum of all the vertical components of the in- ternal stresses at any section of the beam. If A be the area of the section and ,S'S the shearing unit stress, then resist- ing shear = ASs ; and if the vertical shear = V, then V— ASa. The vertical shear is the algebraic sum of all the external vertical forces on one side of the section considered. It is equal to the reaction of one sup- port, considered as a force acting upward, minus the sum of all the vertical downward forces acting between the support and the section. The resistiug moment = algebraic sum of all the moments of the inter- nal horizontal' stresses at any section with reference to a point in that sec- tion, = — , in which S = the horizontal unit stress, tensile or compressive as the case may be, upon the fiber most remote from the neutral axis, c = the shortest distance from that fiber to said axis, and /= the moment of inertia of the cross-section with reference to that axis. The bending moment Mis, the algebraic sum of the moment of the external forces on one side of the section with reference to a point in that section = moment of the reaction of one support minus sum of moments of loads be- tween the support and the section considered. The bending moment is a compound quantity — product of a force by the distance of its point of application from the section considered, the distance being measured on a line drawn from tbe section perpendicular to the direc- tion of the action of the force. Concerning the above formula, Prof. Merriman, Eng. News, July 21, 1894, says : The formula just quoted is true when the unit-stress S on tlie part of the beam farthest from the neutral axis is within the elastic limit of the material. It is not true when this limit is exceeded, because then the neutral axis does not pass through the center of gravity of the cross-section, and because also the different longitudinal stresses are not proportional to their distances from that axis, these two requirements being involved in the de- duction of the formula. But in all cases of design the permissible unit- stresses should not exceed the elastic limit, and hence the formula applies rationally, without regarding the ultimate strength of the material or any of the circumstances regarding rupture. Indeed, so great reliance is placed upon this formula that tlie practice of testing beams by rupture has been almost entirely abandoned, and the allowable unit-stresses are mainly de- rived from tensile and compressive tests. TRANSVERSE STRENGTH. 811 = *fi o © 0 si^l^^K^ ^lfe]-lf*: g& hi5 p p ,-mco-uco-I^io|* + fijc? mini 2^|§ 2 1 ^ CO . O i . p-e £<£ Sl,,S|,,Si.1^i„N| ,,Si,,S £ g§|| fci|«ft;|»ft3.|0ftq|ttfcj|cfi;|«ft5 o«^loft?h 51° o 1 2 II II II II II II 1 II II 11 P go ,-^s A £> £=. ^ £> u *2£ oS OCC A s- 5; fe ^7«> ^ S^ & | L. ^,^,^■00 + rt,C0^ i»~|S iA 1-1 100 ^S A. i-i I-* CO 100 *"bJ l§ 2 oS w oS o o £3 ^co « l^«|g + ^|S ^I8~ % 5 <=>l ii i-* bJO -d oS ce o hi B;| 03 | ft3| 0=| C3| C=| | 1-1 icoth i coin ico-* ico-t< in-t ico^ 05 1 03 1 TfH ICO P bfi II II II II Jl II 1 II II s A L. ft, t^ ^ ft, ft U oS + P A • T5 -p O a • 0 0 s '2 ■a p 0 73 (D ^3 OP 0) ^ s ■> ^ p= T3 a •2 =2 3 0 '3 2 ® ^ p 0 -s OS JH y c3 -P OS •S ^ £ Tl P 0> 43 0) -p g © * P ^ 1 ■§ >, OT 1 1 'P eve ® as -a C3 •£ O P 'P 5 C. 3 Pa. ■p -;§»*" > 2 0 P "oS^ P ® ■ § 1 § § © i L 1880^Z> L 2,2 AD2 mA& Hollow 940 (AD — ad) L lS80(AD—ad) L wLs wL* Rectangle. 32(AD2—ad2) 52(AD2—ad2) Solid Cylinder. 700AB L 1400.4/) L 2AAD2 wL3 38AD2 Hollow 700 (AD — ad) L 1400(AD—ad) L wL* wlfl Cylinder. 2\(AD2—ad2) 3S(AD2—ad2) APPROXIMATE GREATEST SAFE LOAD IN LBS. 813 Shape Greatest Safe Load, in Lbs. Deflection in Inches. of Section. Load in Middle. Load Distributed. Load in Middle. Load Distributed. Even- legged Angle 01- Tee. 930 AD L 1860 A D L wL3 32Alf- wL3 52AD* Channel or Z Bar. 160(U£> L 3200AD L wL3 53AD2 wL3 85AD2 Deck Beam. 145fUZ> 290(U D L Eoaip ivL3 80ZD2 I-Beam. 1780AD L 3560AD L ioL3 58ZZJ2 wL3 93ZD2 I II III IV Y The rules for rectangular and circular sections are correct, while those for the flanged sections are approximate, and limited in their application to the standard shapes as given in the Pencoyd tables. The calculated safe loads will be approximately one-half of loads that "would injure the elasticity of the materials. The rules for deflection apply to any load below the elastic limit, or less than double the greatest safe load by the rules. If the beams are long, without lateral support, reduce the loads for the ratios of width to span as follows : Length of Beam. Proportion of Calculated Load forming Greatest Safe Load. 20 times flange width. 60 70 Whole calculated load. 9-10 8-10 " " 7-10 " 6-10 " " 5-10 " These rules apply to beams supported at each end. For beams supported otherwise, alter the coefficients of the table as described below, referring to the respective'columns indicated by number. Changes of Coefficients for Special Forms of Beams. Kind of Beam. Coefficient for Safe Load. Coefficient for Deflec- tion. Fxed at one end, loaded at the other. One-fourth of the coeffi- cient of col. II. One-sixteenth of the co- efficient of col. IY. 814 FOUNDATIONS AND STRUCTURAL MATERIALS. Changres of Coefficients — Continued. Kind of Beam. Coefficient for Safe Load. Coefficient of Deflec- tion. Fixed at one end, load evenly distributed. One-fourth of the coeffi- cient of col. 111. Five forty-eighths of the coefficient of col. V. Both ends rigidly fixed, or a continuous beam, with a load in middle. Twice the coefficient of col. IE. Four times the coeffi- cient of col. IV. Both ends rigidly fixed, or a continuous beam, with load evenly dis- tributed. One and a half times the coefficient of col. III. Five times the coeffi- cient of col. V. Modulus of Elasticity and Elastic Resistance. P = tensile stress in pounds per square inch at the elastic limit e = elongation per unit of length at the elastic unit ; E = modulus of elasticity = P -f- e ; e = P -j- E. j pi Then elasticity resilience per cubic inch = \Pe = 2 E' THRODGHOIT BEAMS OE MIFORM STBEI^TH THEIR LEHGTH. The section is supposed in all cases to be rectangular throughout. The beams shown in plan are of uniform depth throughout. Those shown in elevation are of uniform breadth throughout. B = breadth of beam. D = depth of beam. Fixed at one end, loaded at the other ; curve parabola, vertex at loaded end ; BD2 proportional to distance from loaded end. The beam may be reversed so that the up- per edge is parabolic, or both edges may be parabolic. Fixed at one end, loaded at the other ; tri- angle, apex at loaded end ; BD2 proportional to the distance from the loaded end. Fixed at one end ; load distributed ; tri- angle, apex at unsupported end ; BD2 pro- portional to square of distance from unsup- ported end. Fixed at one end ; load distributed ; curves two parabolas, vertices touching each other, at unsupported end ; BD2 proportional to dis- tance from unsupported end. Supported at both ends ; load at any one point ; two parabolas, vertices at the points of support, bases at point loaded ; BJJ2 pro- portional to distance from nearest point of support. The upper edge or both edges may also be parabolic. Supported at both ends ; load at any one point ; two triangles, apices at points of sup- port, bases at point loaded ; BD2 propor- tional to distance from the nearest point of support. Supported at both ends ; load distributed ; curves two parabolas, vertices at the middle of the beam ; bases center line of beam ; BD2 proportional to product of distances from points of support. Supported at both ends ; load distributed ; curve semi-ellipse ; BD2 proportional to the product of the distances from the points of support. TRENTON BEAMS AND CHANNELS. 815 XUJEJffTOJtf BEAMS AJD CHA^aTEES. (Trenton Iron Works.) To find which beam, supported at both ends, will be required to support with safety a given ■uniformly distributed load : Multiply the load in pounds by the span in feet, and take the beam whose " Coefficient for Strength " is nearest to and exceeds the number so found. The weight of the beam itself should be included in the load. The deflection in inches for such distributed load will be found by divid- ing the square of the span taken in feet, by seventy (70) times the depth of the beam taken in inches for iron beams, and by 52.5 times the depth for steel. Example. — Which beam will be required to support a uniformly distrib- uted load of 12 tons (= 24,000 lbs.) on a span of 15 feet ? 2-±,O0O X 15= 360,000, which is less than the coefficient of the 12i-inch 125- lb. iron beam. The weight of the beam itself would be 625 lbs., which, added to the load and multiplied by the span, would still give a product less than the coefficient; thus, The deflection will be : 24,625 X 15=369,375. z 0.26 inch. 70 X T2i The safe distributed load for each beam can be found by dividing the coefficient by the span in feet, and subtracting the weight of the beam. When the load is concentrated entirely at the center of the span, one-half of this amount must be taken. The beams must be secured against yielding sideways, or the safe loads will be much less. TREITOH ROLLED STEEL BEAMS. Designation of Beam. 12 10 10 10 2 Weight per Yard in Lbs. 150 123 120 96 39 30 190 160 150 125 160 125 100 105 85 Width of Flanges in Inches. 5.5 5.25 5.25 5.0 4.75 4.75 2.75 2.62 .75 1.50 Thickness of Stem. .45 .37 Coefficient for Strength in Lbs., Minimum Weight. 753,000 603,000 500,000 407,000 461,000 344,000 264,000 232,000 200,000 192,000 154,000 151,000 118,000 104,000 83,300 67,000 52,900 41,200 31,400 2,660 2,300 816 FOUNDATIONS AND STRUCTURAL MATERIALS. TBE]¥TO]¥ IROJY JlKATlft A]¥» CHAHrHEl§. .3 I W 2J= ah 3| oj Coefficient in Lbs. for Transverse Strength. a .5 W 2-S ^ !h 5? m an % ft.1* J OOlOLOlOOlOOLOlOlOLOLOCOOOOOOOHOM M ^iniQOtst-coo2:u::i:ii: i- i- L- — co — ■ — i^- — t- oiot-ooMOOci^^irii^utc^occooo M " » o o u si ft c3 O S »>oq»iocoqccco-*LOoqq©cooociLoa)rt>a j^ co •* CO -# CO -<# CO id CO •#" CO -#' CO ■* CO CO CO 00 CO -# CO ■* ■*" id to by j OOOIOLOOOOOOC IO l- ■ o io uO o O O O O O O oo os o co o c co oi ' o i o i ' o i io o'V lo lo lo lo © © o £ O o ft ■2s a; co ft 1 o ® -2 • 3* Jo ^ ©coot^^iqcor^t^-t^c^cq^iqi^-iqoioccoi^ococ^ H CO^COCO00ri<00^CO^^^c6^C0^00-^CO»l0^cdc6lO to ■Sfe: ►* S- 8§§S!S88g8§8§S8ia88§g88888 i-l rtrlHr-rirlrlClClfllMOl lQffltoONoocoocc c^icci^ii.t :u*i-i:i-'30ioo ^ ,-^^^^^^c,^, o o © M ft1-1 e3 O « 3 • lis! H to '""' ^ 'I * "? <°. °7. L~ ,;r- - ] ~: —. L~ ~. °. °> ,~i ^ n. r1 ^ °5 ^ 1 H co^^iCCO^cOT^co^co^^O^^^ior^iO^OCOiJO 05 -3, H v S-Sft^ J O00cooioioi0iocioc->oio:oi0i0'0>ocooooo w co co -1- 10 10 io io — — l- x co co oi co oi oi o I : i io '0 r 10 - Tt* ia CD «o co t- 1- -o CO CO CO co C 0 I 01 O I 0 1 10 10 10 10 LO LO LO Szi o o M si ■03 c3 O O _; ^ ■ h J ?-§ - fcj CD O CO 00 03 OI O t^ © CO CO CO O ■OOOHH 1 LO | CO | W "*ioc6^-*iaicc5iocDrtii6cD *'i6i-OC5lo 'irj ' lo , -i o o c o o o io io io © o io o "oisoioia no " o 1-1 CO CO CO ^5 ^ io LO cs vr L- l- CO - _ - - r. 0101 _oi _S ■* io ia ia c- co n co oo o o o c oocici'ci i« io 03 0) 5 3 c"§ 02 ^ O01-*C0C0OCM-*C000O 00 t-i, rH H rH iH, STRUCTURAL MATERIALS. •" ^ ~£ c 5 o o in g - c c o O c o o P4;£J o 1 c 1 L-j oo 1 rS M += 03 1 1 8 CO IQ rh i^s-s o OS o o o O ■Stofi'S? 00 o -- r : i o SrH-t i- ,-, co o i oo -t< i-> t- u".(MOt- o oo cm en h o n ci t~ o •* co n h c g a » co t- i» t» » to to ;o io Ot-LOCOCOCMCM-- rH CO t- CO CM CI -+ ^ CM i-l L- -+i I- CM I- t-h i-< CO CO i-l OO CI CM t- ^ -# rf « — r : : r : : ■ - -t- r — co c-. :: — i - r i- ~ > - -- x -t — c re -+ o i r v. CO X lO -f l- CM CI t- lO CO CM t* O CS CI OO OO L~ L- CD CO CO' LO ICO LO LO LO -I CO CO-* CO CMC- Hori««'riHHrHHrid6666666dd6d666o'6o'd 0 CO CO i-l CO CM OO IO CM O L- lei CO 5- ~ l&~ C^\:'-f- £ o q co in q q -ch c; --i q q oo l - l- q q q q q q •* -p -ch -t; -+ co co co co co didco'ciciHHririrHodoooodddddddddddddd 5 q q co t- t^ q q q lco ^ --ch -ch -oh *-h co co co co co 'c6 co cm t>. CCTFcicMTMi--5-MT--10©©0©0©d©ddddddddddddd f CM LCi CO CS C~ CM OO CM -J 1-1 cs -J CO co -ch CO CO W t— i-l H CM cs 00 r> t— 00 OS iH v i - .— co x o cs co l- cs co l- co cs co co r X or; ~\- -- ,-- — rr , r. — J>. rY K rr r. -r :o l- -o v-, q x . - q q q q -i- -r -f -# q q co co co co cm cm ci ci oi ci oi onciHriHooooododdddddddddddddddd lONOfflocogngWHoncjL'jNHMtoHajtoioitioteojHto o i i— l- r o i co o o cm -co i-i co co o l~ i~ co i-H cs oo co lo -f cc- ^ i r; — - - co x x -tHOco l- q q q -ch -ch -h co q co co ci oi cm cm ci ci 6i cm oi cm S ^c4rHiHrHOOOOOOOOOCDOoooOo^^^<='C'C'<='G' hi tredg -h cm i-h i-i x — i r; lo -h -h -ch o oi t- co oo i-i tr- -ck oi o i oi co t-cfrtirca • -*C}rHT--i ©©©©©©©©©© ©'©©©©©©©©©©©©© • _ l-HCO co co co oi oj oi 6 DOCOLOOt-NClX'OI-COClC.t-Ot-OCl LO K-tC-fOtO o io co o i r i- io v-; i- io oi o co l- co lo -f co co oi i-i i-i cD r r cs OClCOOlOtWCOClClCKIHrtHHHrtHHHrtHrfHO spiddoddddddddoddddddooe'ddo tHOOOOOOOOOOOOOOOOOOOOOOOO c To co o i — cd ccs co t- i. ICO OI OO CO ICO -H CS OO CS 01 I- 01 X I- 01 CS I- ICO CO r- CCO N oi holociO coi-cocoi-c i~ -k -* -h co co co co ro re oi oi oMNrtHiHoqc o cccqcccccccc r do'ddddo'dddcdcddcddddccc ^ ~ ^ ,~ -5 — C= l-O —-< -v- i - -- .1 C Cl f» CC LO LO -H CO CO - ci h q q q cq c q q q q cbooooo qq q q dddddo'ddddddddddoddddd " qdqcqc q q q q q q occoceoq " * o o d d d cd cd cd d cd d o" r c r" r" cd r d d ' ' 3 t» 00 CS O i-< fl « "f LO C3 N CO CS D "-< Ol CC •+ l' CC HMHrtrtHHrtHlCKMClClClClC wood. 821 Rule. — To find the safe uniformly distributed load in tons for white pine or spruce beams, multiply the number given in the above table by the thick- ness of the beam in inches. For beams of other wood, multiply also by the following numbers : White Oak. Hemlock. White Cedar. Yellow Pine. Chestnut. 1.45 .99 .60 1.50 1.08 Formula' for White JPine Beams. Subject to vibration from live loads. w = safe load in pounds, less weight of beam. I = length of beam in inches. d = depth of beam in inches. b = breadth of beam in inches. For a beam fixed at one end and loaded at the other: 1000 6rf2 w=— 6T— For a beam fixed at one end and' uniformly loaded : 1000 bd2 W=-JT— For a beam supported at both ends and loaded at the middle: 2000 bd* W=—3f— For a beam supported at both ends and uniformly loaded: 4000 bd2 w = —sr- Note. — In placing very heavy loads upon short, but deep and strong beams, care should be taken that the beams rest for a sufficient distance on their supports to prevent all danger from crushing or shearing at the ends. Ordinary timbers crush under 6,000 lbs. per square inch. To assure a safety of beam against crushing at the end, divide half of the load by 1000; the quotient will be the least number of square inches of base that should be allowed for each end to rest on. Table of Safe Load for Moderately Seasoned White JPine Struts or I*illars. The following table, exhibiting the approximate strength of white pine struts or pillars, with flat ends, is outlined and interpolated from the rule of Rondolet, that the safe load upon a cube of the material being regarded as unity, the safe load upon a post whose height is, 12 times the side will be I 60 72 700 pounds per square inch is assumed as the safe load upon a cube of white pine. The strength of each strut is considered with reference to the first-named dimension of its cross-section, so that if the second dimension is less than the first, the strut must be supported in that direction, to fulfill the condi- tions of the computation. The strength of pillars, as well as of beams of timber, depends much on their degree of seasoning. Hodgkinson found that perfectly seasoned blocks 2 diameters long, required in many cases twice as great a load to crush them as when only moderately dry. This should be borne in mind when building with green timber. 822 FOUNDATIONS AND STRUCTURAL MATERIALS. I. Safe Distributed Xioads upon Southern Pine Beams One Inch in Width. (C. J. H. Woodbury.) (If the load is concentrated at the center of the span, the beams will sus- tain half the amount as given in the table.) CD Depth of Beam in Inches. fc i|s|« * • | T| 8 9 10 11 | 12 | 13 1 "1 ,5 | 16 Load in Founds per Foot of Span. 5 38 86 154 240 34fi 470 614 778 960 6 27 60 107 L67 240 327 427 540 0(17 807 7 20 44 78 122 176 240 314 397 41)0 593 705 828 8 15 34 60 94 135 1S4 240 304 375 454 540 634 735 9 27 47 74 107 145 190 240 296 359 427 501 581 667 759 10 22 38 60 86 11« 154 194 240 290 34! ; 40(1 470 540 614 11 32 50 71 97 127 161 HIS 240 286 335 38!) 446 508 12 27 42 60 82 107 135 1(17 202 240 282 327 375 474 13 36 51 70 90 115 142 172 205 240 278 320 364 14 31 44 60 78 99 123 148 176 207 240 276 314 15 27 38 52 68 86 107 129 154 ISO 209 240 273 16 34 46 CO 76 94 113 135 158 184 211 240 17 30 41 53 67 83 101 120 140 163 187 217 18 36 47 60 74 90 107 125 145 167 190 19 43 54 66 80 96 112 130 150 170 20 38 49 60 73 86 101 118 135 154 21 44 54 06 78 92 107 122 139 22 50 60 71 84 97 112 127 23 45 55 65 77 89 102 116 24 50 60 70 82 94 107 25 46 55 65 75 86 98 DiNtrihuted Load* upon Southern Pine Beams Suf- ficient to Produce Standard Limit of Betlection. (0. J. H. Woodbury.) % Depth of Beam in Inches. £ fr p- 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 * o Load in Founds per Foot of Span. 5 3 10 23 44 77 122 182 259 .0300 6 2 7 16 31 53 85 126 180 247 .0432 7 5 12 23 39 62 93 132 181 241 .0588 8 4 9 17 30 48 71 101 139 185 240 305 .0768 9 7 14 24 38 56 80 110 146 190 24! 301 .0972 10 6 11 19 30 46 65 89 118 154 195 244 300 .1200 11 9 16 25 38 54 73 98 127 161 202 248 301 .1452 12 13 21 32 45 62 82 197 136 10!) 208 253 .1728 13 11 18 27 38 53 70 91 116 144 178 215 .2028 14 16 23 33 45 60 78 100 124 153 186 .2352 15 14 20 29 40 53 68 87 108 133 162 .2700 16 18 25 35 46 60 76 95 117 147 .3,072 17 16 ' 22 31 41 53 68 84 104 126 .3468 18 20 27 37 47 60 75 93 112 .3888 19 18 25 33 43 54 68 83 101 .433,2 20 22 30 38 49 61 75 91 .4800 21 20 27 35 44 55 68 83 .5292 22 24 32 40 50 62 75 .5808 23 22 29 37 46 57 69 .6348 24 27 34 42 52 63 .6912 25 •• .. | .. 25 31 39 48 58 .7500 823 MASONRY. Brick work is generally measured by 1000 bricks laid in the wall. In con- sequence of variations in size of bricks, no rule for volume of laid brick can be exact. Tbe following scale is, bowever, a fair average. 7 common bricks to a super, ft. 4-incb wall. 14 " " " " 9-incb " 24 " " " " 13-incb " 28 " " " " 18-incb " 35 " " " " 22-incb " Corners are not measured twice, as in stone- work. Openings over 2 feet square are deducted. Arcbes are counted from tbe spring. Fancy work counted l£ bricks for 1. Pillars are measured on tbeir face only. One tbousand bricks, closely stacked, occupy about 56 cubic feet. One tbousand old bricks, cleaned and loosely stacked, occupy about 72 cu- bic feet. One cubic foot of foundation, with one-fourtb inch joints, contains 21 bricks. In some localities 24 bricks are counted as equal to a cubic foot. One superficial foot of gaviged arches requires 10 bricks. Stock bricks commonly measure 8| inches by 4^ incbes by 2| inches, and weigh from 5 to 6 lbs. each. Paving bricks should measure 9 inches by 4i inches by If inches, and weigh about 4i lbs. each. One yard of paving requires 36 stock bricks, of above dimensions, laid flat, or 52 on edge^ and 35 paving bricks, laid flat, or 82 on edge. The following table gives the usual dimensions of the bricks of some of the principal makers. Description. Inches. Description. Inches. Baltimore front . Philadelphia front Wilmington front Trenton front Croton .... Colabaugh . . . >- 8i X 4i X 2f 8J X 4 X 1\ 8i X 3f X 2§ Maine .... Milwaukee . . North River . Trenton . . . Ordinary . . . 7\ X 3f X 21 %\ X 4i X 2f 8 X Sk X 2J 8 X 4 X21 ( 7f X 3f X 2\ { 8 X 4i X 2i ( Valentine's (Woodbridge, N. J.) ( Downing's (Allentown, Pa.) . . 8| X 4f X 2|- inches 9 X 4J X 2| inches To compute the number of bricks in a square foot of wall. — To the face dimensions of the bricks used, add the thickness"of one joint of mortar, and multiply these together to obtain the area. Divide 144 square inches by this area, and multiply by the number of times which the dimension of the brick, at right angles to its face, is contained in the thickness of the wall. Example. — How many Trenton bricks in a square foot of 12-inch wall, the joints being J inch thick ? 8-4-J X 2J + \ — 20.62 ; 144 -=- 20.62 = 7 ; 7 X 3 = 21 bricks per square ft. S24 FOUNDATIONS AND STRUCTURAL MATERIALS. W^igli* and Bulk of It rick*. Number of Bricks, by itself. in wall w Tons. Pounds. Cu. ft. C. Brick. F. Brick. C. Brick. 1 2240 22.4 448 416.6 381 0.044.64 100 1 20 18.6 17 2.23 5000 50.00 lOOO 930 850 2.4 5376 53.76 1075 lOOO 914 2.62 5872 58.72 1130 1100 lOOO 2.88 6451 64.51 1240 1200 1100 One perch of stone is 24.75 cubic feet. In New York City laws a cubic foot of brick-work is deemed to weigh 115 lbs. Building-stone is deemed to weigh 160 lbs. per cubic foot. The safe load for brick-work according to the New York City Laws is as folio \vs : — In tons per superficial foot, For good lime mortar 8 tons. For good lime and cement mortar mixed . ll£ tons. For good cement mortar 15 tons. Average Ultimate Crushing--Uoad in Pounds per Square Inch for Jtricks, Stones, mortars, and Cements. Lbs. per Sq. In. Brick, common (Eastern) Brick, best pressed Brick (Trautwine) Brick, paving, average of 10 varieties (Western) Brick-work, ordinary Brick-work, in good cement Brick-work, first-class, in cement Concrete (1 part lime, 3 parts gravel, 3 weeks old) Lime mortar, common Portland cement, best English, Pure, three months old Pure, nine months old 1 part sand, 1 part cement, Three months old Nine months old Granites, 7750 to 22,750 Blue granite, Fox Island, Me Blue granite, Staten Island, N. Y Gray granite, Stony Creek, Conn North River (N. Y.) flagging Limestones, 11,000 to 25,000 Limestone from Glen's Falls, N. Y. ... Lake limestone, Lake Champlain, N. Y. . . White limestone, Marblehead, O White limestone from Joliet, 111 Marbles, From East Chester, N. Y Common Italian Vermont (Souther! and Falls Co.) . . . . Vermont, Dorset, Vt Drab, North Bay Quarry, Wis 10000 12000 770 to 4660 7150 300 to 500 450 to 1000 930 620 770 3760 5960 2480 4520 12000 14875 22250 15750 13425 12000 11475 25000 11225 12775 12950 11250 10750 7612 20025 MISCELLANEOUS MATERIALS. 825 Average Ultimate Crushing-- JLoad — Continued. Lbs. per Sq. In. Sandstones Brown, Little Falls, N. Y Brown, Middletown, Conn Red, Haverstraw, N. Y Red-brown, Seneca freestone, Obio . . . Freestone, Dorcbester, N. B Longmeadow sandstone, Springfield, Mass. 6000 9850 6950 4350 9687 9150 8000 to 14000 lEKSOBULAjrBOlJS MATERIALi. freight of Hound Bolt Copper Per Foot. Incbes. Pounds. Incbes. Pounds. Inches. Pounds. t .425 3.02 If 7.99 .756 l-i 3.83 If 9.27 f 1.18 li 4.72 H 10.64 1 1.70 If 5.72 2 12.10 3 2.31 li 6.81 Weight of Sheet and Bar Brass. Thick- Sheets Square • Round Thick- Sheets Square Round ness. per Bars Bars ness. per Bars Bars Incbes. sq. ft. 1 ft. long. 1 ft. long. Inches. sq. ft. 1 ft. long. 1 ft. long. lbs. lbs. lbs. lbs. lbs. lbs. ? 2.7 .015 .011 1A 45.95 4.08 3.20 5.41 .055 .045 li 48.69 4.55 3.57 f 8.12 .125 .1 if 51.4 5.08 3.97 10.76 .225 .175 54.18 5.65 4.41 y 13.48 .350 .275 li5s 56.85 6.22 4.86 16.25 .51 .395 It 59.55 6.81 5.35 19. .69 .54 1ft 62.25 7.45 5.85 f 21.65 .905 .71 n 65. 8.13 6.37 24.3 1.15 .9 ii9s 67.75 8.83 6.92 § 27.12 1.4 1.1 if 70.35 9.55 7.48 ¥ 29.77 1.72 1.35 it' 73. 10.27 8.05 32.46 2.05 1.66 75.86 11. 8.65 if 35.18 2.4 1.85 78.55 11.82 9.29 i 37.85 2.75 2.15 H 81.25 12.68 9.95 40.55 3.15 2.48 HI 84. 13.5 10.58 l 43.29 3.65 2.85 2 86.75 14.35 11.25 Composition of Various Oracles of Boiled Brass. Trade Name. Copper. Zinc. Tin. Lead. Nickel. Common high brass Yellow metal Cartridge brass 61.5 60 66| 80 60 60 66f 61* 38.5 40 331 20 40 40 33J 20* 1 'lV l*to2 *18 Clock brass Drill rod Spring brass 18 per cent German silver . . . 826 FOUNDATIONS AND STRUCTURAL MATERIALS. So O 3 w Lbs. 1.22 1.08 .966 .860 .766 .682 .608 .541 .482 .429 .382 .340 .303 .270 .240 .214 .191 .170 .151 .135 53 o a -CMC- /.i--i:c-f-)< :o co cicicic-iHrtrt A^AA ft© © CD O 3 c3 • lis lis t> o t— w * C- CO I- 10 CO t- Cl-C-lt-QOHnfHH CO 10 .0 CO ^ _ -c ,- ,-^cii- ,o CO CO 01 CO' -fTHC. 1-10 -f CO - l ^ CO CO -P H Ci L- liS -tn COCNC-li-HrHrH©©©©©© © ft Q • cscio-f co-t«HHCioc: So oi~l- re a? ii5 -H -t< Ol — , co ■ —. x co — -f ~ i0 o i - r - co -p co oi ,o -p q lo o i co L- co -h co sciHHrtoqqooo .WNHOO Oi-iNLOCCOO-t-tOCOiOH-f _-< CO -+i 1- © O -V CO -+i i.O 0 1 0 1 10) CO CO — ~ liS. © 00 -f cr-t cci-ricc-H-ocic: : - -i- co lo . i-i -cic!Cici-Hrfrt-.rt^_ : - - HOOOCOCCCCOOOOCCOOOOO °8) o"g HNOS-PLOCCNOOCjOHSlCO-fLOONCCCSO (M LO» 'j05^ocooi^csooi>0{oid^^eoeoeco^oc50iiSoocoin ct X 1.0 liS I- H CO CO 0] 01 00 1.0 -X r- CO i-< CO 01 CO liS CO © CO © -ch £ © ^ eg b/JO • CO H t- CI i.O 01 01 CO 1 - CO liS 01 -+l CO liS CO -H lis C} SS CO lis 00 OCI cc oi co l- co -t; co in -+ cq © uo c; -# co lo q co l- q lo co q © q oiisddrtdoioc!^ id re! i - i -' — ' co co' -v A co t^ id -#° oo oi -Ol-X O :-•/. I0HCC L- LO -P CO 01 Ol tH i-l i-l CO ■* CO OS Ol ^ rl i-l o O OHCC+COtlNlO-fONOClOO • lo q q lo co — cq -v o i -r q q q -p q i- q -p x co. oi q q q - co' CO 0 i CO CO CO CO CO © CO CO CO CO i-H -f' CO L0 0 i CO t>l CO •# OS 00 ^-t;-:;o.csi.ociot-cs-*cocoisiHHH 4-1 °" o£ A o -f< o co o co oi rn -ti oi co co oo co cc^ o co co co c~i lis ci r. vr "- -f -f ., 1 - CO -t< ^h 0 1 -ti CO -f i-i ■_■ ■_ — ( -t 1 _ LO CO LO — i - co co co oi co. io oi co co -otsiiH co — x l- ■- •'. i.O -t -+ c- CO H ^ -t 00 CO 01 01 CI C) H H h H H rt .;. o c c o o o o o o OOgOHCMCO^LOCSt-COCOOi-IOlM'flOCDNCOCBO go MISCELLANEOUS MATERIAL. 827 Galvanized Iron Wire Rope. For Ships' Rigging and Guys for Derricks. CHARCOAL ROPE. CP 3.3 III Cir. of New Manila Rop of Equal Strengtli. tes'*-' • •c - °^ 11 So !-. j3 o ° a! S Cir. of New Manila Ropt of Equal Strengtli. ||| U O 5i 23} 11 43 21 51 5 9 5i 24V 10} 40 2i 4} 4f 8 5 22 10 35 2 3} 3| 7 4f 21 9} 33 H 2.T 5 4* 19 9 30 1} 2 3 3} 4i 16} 8} 26 U If 21 2} 4 i*i 8 23 H 11 3| 12J 7* 20 l * 21 3} 10| 6k 1G 1 If 1 3i 9} G 14 f i 1} 3 8 5f 12 § § H 1 2| 6i Si- 10 } 5 if Transmission an<8 Standing* Rope. With 6 Strands of 7 Wires Each. IROIf. 2 6 & ? O 0 s> a O A 03 .5 0 •31 . 03 ~ oa Proper Work- ing Load in Tons of 2000 Lbs. Circumference of new Manila Rope of Equal Strength. 05 CG g 0) ri 3fr §3.3 11 1} 4f 3.37 36 9 10 13 12 4i 3| 2.77 30 7* 9 12 13 1+ 2.28 25 6i 8} 10| 14 3| 1.82 20 n 9} 15 3 1.50 16 4 6} 8* 1G 2| 1.12 12.3 3 5f n 17 * 3 0.S8 8.8 -i 4} 6f 18 0.70 7.6 2 4} 6 19 11 0.57 5.8 1* 4 5i 20 TI if 0.41 4.1 1 3? 4* 21 if 0.31 2.83 f 93 4 22 li 0.23 2.13 2} 3i 23 H 0.19 1.65 2i 24 156 0.16 1.38 2 2} 25 32 i 0.125 1.03 if 2i C VST STEEL. 11 ji 4f 3.37 62 13 13 8} 12 If 4^ 2.77 52 10 12 8 13 If % 2.28 44 9 11 11 14 1.82 36 7} 10 64 15 3 1.50 30 6 9 5f 16 f 2f 1.12 22 4} 8 5 828 FOUNDATIONS AND STRUCTURAL MATERIALS. Transmission and Standing- Rope. — Continued. CAST STEEL. © ©" o O © s 0 a® ft 5° •is X3g in S5| > 1 g «J _, © 1 . 5 ® 1 H 5 5 p: s'> © ©EH« m.5m O - Q,© 17 1 2| 0.88 17 3£ 7 4+ 18 *t 2i 0.70 14 3 6 4" 19 s 11 0.57 11 2:1 5n- 3i 20 1 If 0.41 8 11 4f 3 21 if 0.31 6 n 4 2A 22 11 0.23 4J it 3:7 21 23 if 0.19 4 1 3| 24 1 0.16 3 j 11 25 3% i 0.12 2 j 21 li Pliable Hoisting- It ope. Wit b. 6 strar ids of 19 Wii es Each (Trenton Iron Works.) IKON. 5 © ©J ®tf| O CO 3 s^° |^8 05 $ c3 g ^ o&o 3a8fl §P© © 2 5m "bif^ * ^ ^•5 5 5 © © 5 5 2^ s § p.5 ?o Woqo fl © © 1 21 6i 8.00 74 15 14 13 2 2 6 6.30 65 13 13 12 3 If 5i 5.25 54 11 12 10 4 If 5" 4.10 44 9 11 8J 5 n 44 3.65 39 8 10 7i 5;V if 4i 3.00 33 6J- 9i 7^ G U 4 2.50 27 5i ■ 8h (U- 7 if 3£ 2.00 20 4" n 6" 8 i 3I 1.58 16 3 6| 51 9 I 93 1.20 11.50 2* 5| 4 10 0.88 8.64 lf 4-1 4 101 | 2 0.60 5.13 11 3| 3£ 10* f If 0.44 4.27 f 3i 2f 10| 1* 0.35 3.48 3 21 10 "Water-tune. — Very largely used where high steam pressures or safety from explosion are desirable. Fire passes about the exterior of tubes and in most cases under about one-half the circumference of the steam :drums. Can be built for any size or pressure. Tubes are generally placed in a slanting position, from one set of headers to another, as in the Babcock & Wilcox, Heine & Go. ; or vertically, as in the Sterling and Cahall. > Vertical Fire Tube. — Used considerably in New England. Spe- cial design by Captain Manning; tubes 15 feet long 2£ inches diameter, arranged in vertical shell with large combustion chamber surrounded by a water leg. Gases mingle in combustion chamber, and in passing through the long narrow tubes give up nearly all the heat, practicably leaving flue gases 450° to 500° F. By controlling height of water steam can be superheated. Can be built for high pressures and of large size. a^iwfr0?' Marine Boilers. — Not much used for electrical purposes, bneii or thick material, short in length and large in diameter. Furnaces : internal, with return tubes from combustion chamber to uptake. i fi /Sf are the finder boiler, of small diameter and considerable lengtn uu to 35 feet). Fired externally, and gases pass under full length to cmniney._ Flue boiler, has two or three large tubes running full length of snen, which is long and of small diameter. Fired externally under the shell, gases return through the flues to uptake. Neither of these types is now jused for electrical purposes. The Horse-Power of Steam Boiler. The committee of the A. S. M. E. on "Trials of Steam Boilers in 1884" (Trans., vol. vi. p. 265), discussed the question of the horse-power of boilers : 830 STEAM. The Committee) A.S.M.E. see Trans, vol. xxi.) approves the conclusions of the 1885 Code to the effect that the standard " unit of evaporation " should be one pound of water at '212° F. evaporated into dry steam of the same temperature. This unit is equivalent to 965.7 British thermal units. The committee recommends that, as far as possible, the capacity of a boiler be expressed in terms of the " number of pounds of water evaporated per hour from and at 212°." It does not seem expedient, however, to aban- don the widely recognized measure of capacity of stationary or land boilers expressed in terms of " boiler horse-power." The unit of commercial boiler horse-power, adopted by the Committee of 1885 was the same as that used in the reports of the boiler tests made at the Centennial Exhibition in 1876. The Committee of 1885 reported in favor of this standard in language of which the following is an extract : " Your Committee, after due consideration, has determined to accept the Centennial standard, and to recommend that in all standard trials the com- mercial horse-power be taken as an evaporation of 30 pounds of water per hour from a feed-water temperature of 100° F. into steam at 70 pounds gauge pressure, which shall be considered to be equal to 34| units of evaporation ; that is, to 34£ pounds of water evaporated from a feed-water temper- ature of 212° F. into steam at the same temperature. This standard is equal to 33,305 thermal units per hour." The present Committee accepts the same standard, but reverses the order of two clauses in the statement, and slightly modifies them to read as follows : The unit of commercial horse-power developed by a boiler shall be taken as 34£ units of evaporation per hour ; that is, 34^ pounds of water evaporated per hour from a feed-water temperature of 212° F. into dry steam of the same temperature. This standard is equal to 33,317 BritislTthermal units per hour. It is also practically equivalent to an evaporation of 30 pounds of water from a feed-water temperature of 100° F. into steam at 70 pounds gauge pressure.* The Committee also indorses the statement of the Committee of 1885 con- cerning the commercial rating of boilers, changing somewhat its wording, so as to read as follows : A boiler rated at any stated capacity should develop that capacity when using the best coal ordinarily sold in the market where the boiler is located, when fired by an ordinary fireman, without forcing the fires, while exhibit- ing good economy ; and, further, the boiler should delelop at least one- third more than the stated capacity when using the same fuel and operated by the same fireman, the full draft being employed and the fires being crowded ; the available draft at the damper, unless otherwise understood, being not less than J inch water column. Heating- Surface of Boilers. Although authorities disagree on what is to be considered the heating surface of boilers, it is generally taken as all surfaces that transmit heat from the flame or gases to the water. The outside surface of all tubes is used in calculations. Kent gives the following rule for finding the heating surface of Vertical Tubular ISoilers. — Multiply the circumference of the fire- box (in inches') by its height above the grate. Multiply the combined circum- ference of all the tubes by their length, and to these two products add the area of the lower tube sheet ; from this sum subtract the area of all the tubes, and divide by 144 : the quotient is the area of heating surface in square feet. Horizontal Return Tubular Boiler*. — (Christie). Multiply the length of that part of circumference of the shell (in inches) exposed to the fire by its length ; multiply the circumferences of the tubes by their num- ber, by their length in inches ; to the sum of these products add two-thirds of the' area of both tube sheets less twice the area of tubes, and divide the remainder by 144. The result is the herting surface in square feet. Heating* Surface of Tubes. — Multiply the number of tubes by the diameter of a tube in inches, by its length in feet, and by .2618. The diam- eter used should be that of the 'fire side of the tube. * According to the tables in Porter's Treatise on the Richards Steam, En- gine Indicator, an evaporation of 30 pounds of water from 100° F. into steam at 70 pounds pressure is equal to an evaporation of 34.488 pounds from and at 212° ; and an evaporation of 34h pounds from and at 212° F. is equal to 30.010 pounds from 100° F. into steam at 70 pounds pressure. The " unit of evaporation" being equivalent to 965.7 thermal units, the commercial horse-power = 34.5 X 965.7 = 33,317 thermal units. STEAM BOILERS. 831 , Heating* Surface per Horse-power. — There is little uniformity of practice among builders as to the amount of heating surface per horse- power, but 12 square feet maybe taken as a fair average. Babcock ci ci co co co •* ^ _,H_irH^rH_rt,HrH,-,THrH ■jBaji jt? jra[iog J° 9pjS Jtapun oj .iooii jo dox ti- -"©'©Vi .Vl'ci'co "3 lfl O IS lO ul LO O lO 13 O 13 lO » (O ffl •juojjj ye J9[iog jo 9piS Japan o; .ioot^ jo dox ts rtHHCOMMOOOOOOOOOCOW-* 3 13 13 13 13 13 13 13 13 13 CO CO © CO CO CO 6 •somunji jo qjpiAS. H r^COCOTt<-t<-#THl3 13l3COCOCOCOCOt~ ■aotjiuriji jo qjSnax w 3 CO ^ ^ •? T^ -# -# 13 CO CO CO CO CO CO t- •JQlJog jo ©pis jspufi oj saj^j*) jo dox ti 2 CO CO CO CO CO CO CO CO CO O O © O CO o 3 CI CI CI CI CI CI CI CI CI CO CO CO CO CO 00 •xeeg aqj jb saj^jQ jo dox °^ 9un -tooiil c* - r:'::\o^>-^c'-.o co oo co cc co co •-h *-i cn 3 CI CI CI CI CI CI CI 01 CI CI CI 0! CO COCO 'JTtOJ^ 9qj JT? S9JT?jr) jo dox OJ 9HJX JOOT^ Ph 2l3l3C-t-t~©©©©O©©C0C0TH 3 CI CI CI CI CI CO CO CO 00 CO CO CO CO CO CO Pi fcd 13©13©13©13©©©10 0©©13 rtUNKM-t<^inOt-t-M»0 CI 840 The draught power of a chimney varies as the square root of the height. The retarding friction of the chimney may be taken as equivalent to a diminution of its actual area by a layer of gas two inches thick all the way around the perimeter of its hue. A = actual area of flue in square feet. E — effective area of flue in square feet. H= height in feet. 1) = diameter of hue in feet. Dx = side of a square chimney equivalent to A. Then: E=A— O.&^A. (l) £)1 = "V ' E 4- 4 inches. (2) Horse-power = 3.33 E^H. (3) The above formulae are by Kent, and are based on a consumption of 5 lbs. coal per h. p. per hour. W. W. Christie, in a paper read before the A.S.M.E., Trans., vol. xviii., p. 387, gives as his opinion that all chimneys should be compared and rated by using coal capacity as a basis, not horse- poAver. In the following table, coal capacity can be found by multiplying h.p. by 4. Size of Chimneys for Steam-Boilers. (W. W. Christie.) A Height of Chimney. a 50 60 70 80 90 100 110 125 150 175 200 225 250 300 £ csS Q ft. ft. ft. ft. ft. ft. ft. ft. ft. ft. ft. ft. ft. ft. Boiler Horse-power=3.25 E^H; 4 lbs. of coal burned considered 1 H.P. 18 42 46 49 52 1 1 21 55 62 65 68 24 72 78 85 91 98 27 91 101 107 114 124 30 114 124 133 143 153 159 33 36 39 14!) 179 163 192 2<>4 172 lS'J 192 201 ior, 241 •J IS 257 22.S ••70 241 283 257 302 42 263 :;:;'' 351 •;7'i 047 683 cor, 637 774 00!) 715 797 845 809 865 965 1021 1092 72 920 1)0" nr.i 1147 1215 1300 1378 78 1131 1206 134!) 1459 1524 1619 1706 84 1310 1401 1503 1654 1768 1875 1976 2165 90 1009 1794 1898 2031 2155 2269 2486 96 ix:;o '04 1 2161 2311 2451 2584 2831 10<> 1007 '3li4 2434 2607 2766 2915 3195 108 2314 1.-S4 2734 2925 3101 3269 3578 114 ■,.s79 3045 3257 3455 3643 3991 120 :1!)1 3374 3611 3829 4037 4420 13? ;s01 4082 4368 4631 4882 5350 144 •' 4596 4859 5200 5515 5811 6367 841 The following table* will prove useful to those having to do with electric installations, and gives the horse-power of chimneys to be used in power plants having very efficient engines, such as compound or triple expansion engines, when 2 lbs. of coal burned under the boiler produce one horse- power at the engine. Size of Chimney for Steam Boilers. (W. W. Christie.) Height of Chimney S 50' 6C 70/ 807 90' 100 110' 125' 150' 175' 200' 225' 250' 300' 5 H( rse-power = G.5 Ay H. When 2 lbs coal burned per hour rr 1 H.P. 18 84 f 92 98 ,04,'.. 1 1 21 110 124 130 13(3 24 144 156 170 182 19fi 27 182 202 214 228 248 30 228 24S 266 286 301 318 33 298 358 32 s 384 448 52.3 344 410 4S2 554 728 364 431 514 592 774 384 456 540 624 S20 1034 404 482 566 662 858 10X6 36 514 604 702 916 1 158 39 4? 780 1020 1"94 48 54 1366 60 1210 1274 1548 1338 1618 1430 1730 1594 1930 1690 2042 68 2184 72 1840 1924 2102 2294 2430 2600 2756 78 2262 '412 26! ).S 2918 3048 3238 3412 84 2620 2802 3126 3308 3536 3750 3952 4330 90 5218 ■:r>sx 3796 4062 4310 4538 4972 91> 3660 4082 4322 4622 4902 5168 5662 102 . . 4134 460S 4868 5214 5532 5830 6360 108 4628 5168 5468 5850 6202 6538 7156 114 ->7.-»x 6090 6514 6910 7286 7982 12) .... 5382 6748 7222 7658 8074 8840 132 ... 7722 8164 8736 9262 9764 10700 144 "I " )192 9718 10400 11030 11622 12734 Chimney Construction. A brick chimney shaft is made up of a series of steps, each of which is of uniform thickness, but as we ascend each succeeding step is thinner than the one it rests upon. These bed joints at which the "thickness changes are the joints of least stability. The joints and the one at the ground line are the only ones to Avhich it is necessary to apply the formulas for deter- mining the stability of the stack. The height of the different steps of uniform thickness varies greatly, ac- cording to the judgment of the engineer, but 170 feet is, approximately, the extreme height that any one section should be made. This length is seldom approached even in the tallest chimneys, as the brick-work has to bear, in addition to its weight, that due to the pressure of the wind. The steps should not exceed about 90 feet, unless the chimney stack is inside a tower which protects it from the wind. In chimneys from 90 to 120 feet high the steps vary from 17 to 25 feet, the top step being one brick thick ; in chim- * " Chimney Design and Theory," W. W.Christie, D. VanNoslrand Company. 842 : ^, J=i 'Tl° PLAN OF BRICK CHIMNEY ?\%% FOR M.H. BIRGE S 'ALL PAPER FACTORY A MARYLANO 8TS. ^-J/mZ/y ? i ~' 2 I g GREEN & WICKS-j *R^" Fig. 4. 843 neys from 130 to 150 feet the steps vary from 25 to 35 feet ; in chimneys from 150 to 200 feet the steps vary from 35 to 50 feet ; in chimneys from 200 to 300 feet and over, the steps vary from 50 to 90 feet, the top step being one and one- half bricks thick. The outside dimensions of a chimney at the base should generally not be less than one-tenth of the height of the stack for square chimneys ; one-eleventh for octagonal, and one-twelfth for round. The bat- ter may be 2£ inches for every 10 feet. The foundation of a chimney is one of the most important points to be considered. When this is upon solid rock it is only necessary to excavate to a depth sufficient to prevent the heat of the gases from materially affect- ing the natural stone, and to secure the spread of the base. In cases where chimneys are to be built upon alluvial clays or made ground, it is necessary to excavate until a good stiff clay, hard sand, or rock bottom is reached. The excavation is filled with concrete in various ways, or filled according to the judgment of the engineer, so as to economize material without en- dangering the structure. Babcock and Wilcox give the following formula for the ability of brick chimneys to withstand wind pressure. w = weight of chimney in lbs. (brickwork — 100 to 130 lbs. per cubic foot.) d = average diameter' in feet, or width if square. h = height in feet. b = width of base. k = constant, for square chimneys = 56. for round chimneys — 28. for octagonal chimneys = 35. c = k and w = k —r—. w o Thin Shell IS rick Chimneys. — While the steel-plate lined stack is considerably cheaper than the ordinary heavy brick chimney, there is a design of brick chimney used by Messrs. Green & Wicks, architects, of Buffalo, N. Y., that has all the durability of the brick stack, and costs less than one of the same capacity in steel plate. The bricks must all be spe- cially selected, hard burned, laid in rich Portland cement. By courtesy of the architects we are able to show drawings of such a chimney, that was erected by them for a wall-paper factory in Buffalo, and which has success- fully withstood the most severe winds of the region (Figs. 4 and 5). Note on Thin Shell Brick Chimneys. — The fire-brick core must be kept free from the outer shell, not being tied or bonded to it in any manner. The bricks are circular, with inside diameter laid up to 4 feet. The galvanized iron-wire cables shown in the plans are for lightning protec- tion. They are soldered and bolted to the iron cap, and after passing down through staples built into the walls for the purpose, are grounded on 20-oz. copper plates 3 feet by \\ feet, set on edge ten feet away from the foot of the stack. The cables are to be soldered and riveted to the plates, and all the plates must be connected together by a f-inch galvanized iron cable soldered to all the plates, The chimney shown in the plans cost about $ 2,000, and can be built for PLAN AND SECTION SHOWING LOCATION OF CHIMNEY BIRGE FACTORY 844 Draft Power for Combustion of fuels. (R. H.Thurston.) Draft of Chim ney in Indies of Water. Draft in Ins. of Water. Wood. . . . Sawdust . . Sawdust mixed with small coal . . . Steam coal . . . Slack, ordinary . Slack, very small . 0.20 to 0.25 0.35 " 0.50 0.60 " 0.75 0.40 0.60 I 0.90 1.25 Coal-dust Semi Anthracite coal Mixture of breeze and slack Anthracite . . . Mixture of breeze and coal-dust .... Anthracite slack . . 0.80 to 1.25 0.90 " 1.25 1.00 ' 1.25 ' 1.33 1.50 1.75 1.80 Height of Chimney for Burning- Given Amounts of Coal. Professor Wood (Trans. A. S. M. E., vol. xi.) derives a formula from which he calculates the height of chimney necessary to burn stated quan- tity of coal per square foot of grate per hour, for certain temperatures of the chimney gas. Pounds of Coal per Square Foot Grate Area. Temp. Absolute Outside Temp. Chim- 16 20 24 Air. ney Gases. Height of Chimney, Feet. - 700 67.8 157.6 250.9 ■g . 800 55.7 115.8 172.4 7! .3 1000 48.7 100.0 149.1 1100 48.2 98.9 148.8 1200 49.1 100.9 152.0 1400 51.2 105.6 159.9 ©£ ss 1600 53.5 110.9 168.8 2000 63.0 132.2 206.5 Rate of Combustion Due to Height of Chimney. Prof. Trowbridge (" Heat and Heat Engines," p. 153) gives the following table, showing the heights of chimneys for producing certain rates of com- bustion per square foot of area of section of the chimney. The ratio of the grate to the chimney section being 8 to 1. Lbs. Coal burned per Lbs. Coal burned per Hour per sq. ft. of Grate. Lbs. Coal burned per Lbs. Coal Height Hour per Height in Hour per burned per in Feet. sq. ft. of Feet. sq. ft. Sec- Hour per Section of tion of sq. ft. Grate. Chimney. Chimney. 25 68 8.5 70 126 15.8 30 76 9.5 75 131 16.4 35 84 10.5 80 135 16.9 40 93 11.6 85 139 17.4 45 99 12.4 90 144 18.0 50 105 13.1 95 148 18.5 55 111 13.8 100 152 19.0 60 116 14.5 105 156 19.5 65 121 15.1 110 160 20.0 845 Dimensions and Cost of Brick Chimneys. (Buckley.) I. fa 2 fa Outside Wall. £.2 o >? tt o 5 ts ®»£ — rti« Q^ og < 5M No. Brick. C( >st us .$14 per M. Ofa if .5 HO 85 80 25 in. 7 ft. 5 in. 32,000 $ 448.00 $ 60.00 $ 90.00 $ 598.00 135 90 30 in. 8 " 3 " 40,000 560.00 82.00 144.00 786.00 200 100 35 in. 9 "10 " 65,000 910.00 118.00 198.00 1,226.00 300 110 43 in. 10 " 2 " 75,000 1,050.00 190.00 252.00 1,492.00 450 120 51 in. 11 " 2 " 87,000 1,218.00 261.00 306.00 1,785.00 750 130 61 in. 12 " 6 lk 131,000 1,834.00 334.00 360.00 2,528.00 1000 140 74 in. 13 " 11 " 151,000 2,114.00 432.00 414.00 3,060.00 1650 150 8S in. 15 " 1 " 200,000 2,800.00 482.00 468.00 3,750.00 2500 160 110 in. 17 " 10 " 275,000 3,850.00 720.00 525.00 5,095.00 Steel Plate Chimneys have long been used in the iron and coal re- gions, but have only recently come into use in the East, except in the old style thin sheet iron guyed stack, which lasts but a short time. Many of the manufacturers of steel structures are now erecting very sub- stantial steel-plate stacks lined with fire bricks, that are of artistic outline, strong, and when kept well painted are durable and need no guys, as they are spread at the base, and bolted to a heavy foundation. They are usually designed to stand a wind pressure of 50 lbs. per square foot. Sizes of foundations for Steel Chimney. (Selected from Circular of Philadelphia Engineering Works.) Half-Lined Chimneys. Diameter, clear, feet . . . Height, feet Least diameter foundation . Least depth foundation . . Height, feet Least diameter foundation . Least depth foundation . . 3 4 5 3 7 9 100 100 150 150 150 150 15'9" i6'4" 20'4" 21'IC 22/7// 23/8/- 6' 6' 9' 8' 9/ 10' 125 200 200 250 275 18'5" 23'8" 25' 29'8'' 33'6" r 10' W 12' 12'' 11 150 24/8// W 300 36' 14' Brick Lining' for Steel Stacks. Allowing 1| inches air space between stack and lining : Bricks 8\ X 4 X 2 inches, laid without mortar ; Lining B\ inches (one brick) thick ; Number of bricks per foot in diameter of stack, and per foot of height = 47. Allowing 1 inch air space between stack and lining : " Bricks 8J x 4 X 2 inches, laid without mortar ; Lining 4 inches (one brick) thick ; Number of bricks per foot in diameter of stack, and per foot of height = 25. 846 l*i mention* and Cost of Iron Stacks. (Guyed.) (Buckley.) Horse- Height, Diameter, Number of Price Stack Price Power. Feet. Inches. Iron. Complete. per Foot. 25 40 16 12 and 14 $ 61.00 $ 1.52 40 18 12 and 14 71.00 1.78 50 18 12 and 14 84.00 1.68 '75' 50 20 12 and 14 87.00 1.75 50 26 12 and 14 105.00 2.10 60 22 12 and 14 111.00 1.85 160' 60 24 12 and 14 125.00 2.08 60 26 12 and 14 133.00 2.22 60 28 12 and 14 148.00 2.45 125 60 28 10 and 12 190.00 3.18 60 32 10 and 12 203.00 3.38 'l50 60 34 12 and 14 165.00 2.75 200 60 36 10 and 12 215.00 3.58 225 60 38 10 and 12 228.00 3.80 2o0 60 42 10 and 11 257.00 4.28 300 60 46 10 and 12 286.00 4.76 400 60 52 10 and 12 340.00 5.66 For general details of construction of the various types of chimneys used in the U. S. the reader is referred to " Chimney Design and Theory," by W. Wallace Christie, published by D. Van Nostrand Co. Kinds and Ingredients of fuels. The substances which we call fuel are : wood, charcoal, coal, coke, peat, certain combustible gases, and liquid hydrocarbons. Combustion or burning is a rapid chemical combination. The imperfect combustion of carbon produces carbonic oxide (CO), and carbonic acid or dioxide (CO,). From certain experiments and comparisons Rankine concludes "that the total heat of combustion of any compound of hydrogen and carbon is nearly the sum of the quantities of heat which the hydrogen and carbon contained in it would produce separately by their combustion (CH4 — marsh gas or fire-damp excepted)." In computing the total heat of combustion of a compound, it is conven- ient to substitute for the hydrogen a quantity of carbon which would give the same quantity of heat ; this is accomplished by multiplying the weight of hydrogen by 62032 -f- 14500 = 4.28. From experiments by Dulong, Despretz, and others, " when hydrogen and oxygen exist in a compound in the proper proportion to form water (by weight nearly 1 part H to 8 parts O), these constituents have no effect on the total heat of combustion. " If hydrogen exists in a greater proportion, take into the heat account only the surplus." Dulong's formula for the total heat of combustion of carbon, hydrogen, oxygen, and sulphur, where C,H,0,and S refer to the tractions of one pound of the compound, the remainder being ash, etc. Let h = total heat of combustion in B.T.U. per pound of compound. h — 14600 C+ 62000 (h— Q\ -f 4000 S. (A.S.M.E. Trans, vol. xxi.) Rankine says : " The ingredients of every kind of fuel commonly used may be thus classed : (1) Fixed or free carbon, which is left in the form of char- coal or coke after the volatile ingredients of the fuel have been distilled away. These ingredients burn either wholly in the solid state, or part in the solid state and part in the gaseous state, the latter part being first dissolved by previously formed carbonic acid. "(2) Hydrocarbons, such as olefiant gas, pitch, tar, naphtha, etc., all of which must pass into the gaseous state before being burned. 847 " If mixed on their first issuing from amongst the burning carbon with a large quantity of air, these inflammable gases are completely burned with a transparent blue flame, producing carbonic acid and steam. When raised to a red heat, or thereabouts, before being mixed with a sufficient quantity of air for perfect combustion, they disengage carbon in fine powder, and. pass to the condition partly of marsh gas, and partly of free hydrogen ; and the higher the temperature, the greater is the proportion of carbon thus dis "ngaged. " If the disengaged carbon is cooled below the temperature of ignition be- fore coming in contact with oxygen, it constitutes, while floating in the gas, smoke, and when deposited on solid bodies, soot. " But if the disengaged carbon is maintained at the temperature of ignition, and supplied with oxygen sufficient for its combustion, it burns while float- ing in the inflammable gas, and forms red, yellow, or white flame. The flame from fuel is the larger the more slowly its combustion is effected. " (3) Oxygen or hydrogen either actually forming water, or existing in com- bination with the other constituents in the proportions which form water. Such quantities of oxygen and hydrogen are to be left out of account in de- termining the heat generated by the combustion. If the quantity of water actually or virtually present in each pound of fuel is so great as to make its latent heat of evaporation worth considering, that heat is to be deducted from the total heat of combustion of the fuel. The presence of water or its constituents in fuel promotes the formation of smoke, or of the carbona- ceous flame, which is ignited smoke, as the case may be, probably by mechanically sweeping along fine particles of carbon. " (4) Nitrogen, either free or in combination with other constituents. This substance is simply inert. " (5) Sulphuret of iron, which exists in coal and is detrimental, as tending to cause spontaneous combustion. " (6) Other mineral compounds of various kinds, which are also inert, and form the ash left after complete combustion of the fuel, and also the clinker or glassy material produced by fusion of the ash, which tends to choke the grate." Total Heat of CoiulmMtion of JFuel». (D. K. Clark.) The following table gives the total heat evolved by combustibles and their equivalent evaporative power, with the weight of oxygen and volume of air chemically consumed. Combustibles. Quantity of Air Consumed per Pound of Com- bustible. Cu. Ft. at 62°F, O oh Hydrogen Carbon making CO ...... Carbon making CO., Carbonic oxide . Light Carbureted Hydrogen . . Olefiant Gas Coal (adopted average desiccated) Coke(adopted average desiccated) Lignite, perfect Wood, desiccated Wood, 25 per cent moisture . . Petroleum Petroleum oils Sulphur 4.00 3.43 2.45 2.49 2.04 1.40 1.05 3.29 34.8 5.8 11.6 2.48 17.4 15.0 10.7 10.81 8.85 6.09 4.57 14.33 17.93 4.35 229 196 140 142 62000 4452 14&00 4325 23513 213-13 14^00 13548 13108 10974 7951 20411 27531 4000 5.00 1.48 848 -is- i''3 'oOSC su*) 'o09 J« .ny jo Apldns •JJBJQ Aauunqo *mAi Tj* OfflOONd aiqi^Stiquioo -q] ] qgiAV 'oolo W !>«« uicij pajuuo ■d'BAa jag's AV jo spunoj uj ?. § S S § £ g g i-' d i-" ci d l- t^ 10 •apq^stiquioo jo punoj jad oi pasnu jaj'BAi jo spunoj uj -jj - i- -^ ■jiy jo ^[ddng ji?oija.ioaqx aqj saiujx aaaqx tW!A\ 38 3 \ny jo X[ddng t^oi -jajoaqx aqj sox-^X miA\ 2 oooooo \iiy jo j?[ddns j-koijo.t -oaqx 8* -t< ^ in ->* ^ -o -v -+ in o 3 i. s © * 76.94 71.68 79.23 84.46 80.54 75.2 83.98 85.7 83.13 86.68 91.45 87.96 89.19 4 15.3 10.84 13.71 9.2 10.65 16.00 9.91 7.31 9.62 6.15 2.17 6.77 5.23 1 s © z £ £ sIS 6.42 5.78 3.73 5.37 7.54 7.36 4.99 5.95 5.98 5.89 5.03 2.3 1.96 CD o 1.34 1.7 3.33 .97 1.27 1.44 1.12 1.04 1.27 1.28 1.35 2.97 3.62 © o hi W.-Barre, Pa. Scliuyl. Co., Pa. Scranton, Pa. Scranton, Pa. Scranton, Pa. L. V. Region . Scranton, Pa. Scranton, Pa. Scranton, Pa. Avondale, Pa. Scranton, Pa. Drit'ton, Pa. . Cross Creek, Pa L. V. Buckwheat Jermyn . . Woodward . . Cayuga . . . Mt. Pleasant . Ii. V. Pea . . Forty Foot . . Manville Shaft Continental . Avondale . . Oxford . . . Mammoth . . Buck Mountain (3 cq 3 ° CD „£ '£ £ ^11 13700 12043 12724 10899 11827 11231 12217 12855 15255 11959 Pounds Combustible Matter in Smoke from 1 Ton Coal. 11.8 20.94 5.29 8.63 6.12 5.33 6.53 18.25 4.00 9.88 i 9 3 CP cfP 0506 L3126 3528 13756 L4150 14864 4971 5005 5094 15266 3-"S CD cS PyjJ I 275 42 32 28 31 34 34 345 $ HrtrtHHHHrir CD o 49.55 58.61 69.3 69.69 59.45 61.71 69.21 63.26 77.48 64.44 1 «i 11.74 7.83 10.9 5.73 4.22 3.80 5.3 6.1 3.65 4.03 34.94 31.29 18.51 23.55 34.22 31.41 21.4 28.71 17.62 30.42 0 m 3 s to 1 o 3.77 2.27 1.23 1.03 2.11 3.08 1.09 1.93 1.25 1.11 "c8 o o a "z £ 33 « 13 Clearfield Co., Pa. . . Monongaliela River, Pa. No. 2 Slope, Nova Scotia lleynold'sville, Pa. . . Conneilsville .... New River, Va. . . . fl § o .2 2 a: £ - '- '- Eureka . . . Turtle Creek . Nova Scotia . Reynold's ville Leisenring . . Pocahontas H i 852 Proximate A imli -sis of Coal. (Power.) Designation of Coal. ANTHRACITE. Beaver Meadow, Penn Peach Mountain, Penn Lackawanna, Penn. . Lehigh, Penn AVelsh, Wales SEMI-ANTHRACITE. Natural Coke, Virginia Cardiff, Wales Lycoming Creek, Penn Arkansas, No. 16 Geol. Survey .... SEMI-BIT 0 M1N0U S . Blossburg, Penn Mexican Port Smith, Arkansas Cliff, New South Wales, Australia . . Skagit River, State of Washington . . Cumberland, Maryland Cambria County, Penn Mount Kembla, New South Wales, Aus. Fire Creek, West Virginia Arkansas, No. 12 Geol. Survey .... BITUMINOUS. Wilkeson, Pierce County, Washington . Cowlitz, Washington New River, West Virginia Pictou, Nova Scotia Big Muddy, Illinois Bellinghain Bay, Washington .... Midlothian, Virginia Connellsville, Penn Illinois, Average Carbon Hill, Washington Clover Hill, Virginia Wellington, Vancouver Island, B.C. . . Franklin, Washington Rocky Mountains Newcastle, England Mokihinui, Westport, New Zealand . . Brunner Mine, Greymouth, New Zealand Pittsburg, Penn Nanaimo, Vancouver Island, B.C. . . . Hocking Valley, Ohio Pleasant Valley, Utah Kentucky Ellensburg, Washington Olympic Mountains, Washington . . . Scotch, Scotland Roslyn, Washington Cook's Inlet, Alaska Kootznahoo Inlet, Admiralty I., Alaska Liverpool, England Calispel. Washington Carbonado, Washington Upper Yakima, Washington MethoAV, Washington 2.96 3.91 3.28 6.25 12.44 12.: *5 13.S4 14.93 14.78 14.86 17.2 17.7 18.8 19.87 20.52 2.46 1.26 8.93 2.16 1.34 2,15 1.59 1.7 2.25 6.95 5.43 1.25 3.74 2H.9:', 22.42 24.66 26.12 26.64 27. S3 29.5 29. r»4 29. S6 30.10 30.14 31.73 32.21 34.15 34.27 34.65 34.7 34.94 35.68 36. 36.05 36.15 37.73 37.89 39.1 39.15 39.19 39.7 39.87 37.02 39.96 41.1S 42.27 42.47 43.71 NX. 94 89.02 87.74 75. 1 is 81.9 71.53 74.06 73.11 55.7 73.05 71.8 71.66 69.37 66.96 66.75 61.9 70.66 56.98 54.64 59.9 53.01 59.61 45.93 55.8 56.83 54.85 54.23 42.85 59.3 57.92 56.62 55. 51.95 51.3 49.40 50.01 54.4 47.01 48.81 52.65 49.89 45.15 54.9 42.92 52.11 52.21 49.27 853 "£■§ ^ 6-g ^S 1* 4. £ 3 ^kH Eii '_' w H r* w 32 G2.42 0. 41 62.42 9. 50 62.41 18. 59 62.38 27.01 '.',.', 62.42 l. 42 62 42 10. 51 62.41 1!). 60 G2.37 28.01 34 62.42 2. 4:5 62.42 11. 52 62.40 20. Gl G2.37 29.01 35 62.42 3. 44 62.42 12. 53 62.40 21.01 62 G2.36 30.01 36 62.42 4. 45 62.42 13. 54 02.40 22.1 H G3 62.36 31.01 3V 62.42 5. 41 ; 62.42 14. 55 62.39 23,01 64 62.35 32.01 38 62.42 6. 47 62.42 15. 56 62.39 24.01 65 62.34 33.01 39 02.42 7. 48 64.41 16. 57 62.39 25.01 66 62.34 34.02 40 62.42 8. 49 62.41 17. 58 62.38 2G.01 67 62.33 35.02 857 Weight of Water — Continued. is CO Ms to -a .2 MS no ~& ^6 ^,0 ~& rfe #o+± 3 S°-^ % « P P I > 3©^ B ft iij bDu O ■otsu O — i iJQ •HJ)0 •X dj o ■~ v z ® ft^ o> g-is-d gftk . '33 3 u o ft ~ -.1- "Z P u s pp^ p*p ^ '53 P n § £&§* s^p l> ft ftf^ ©^Q H t> ftftfr r* ftftft g~Q j£ ftftft 212 59.71 280 57.90 350 55.52 420 52.86 490 50.03 220 59.64 290 57.59 360 55.16 430 52.47 500 49.61 230 59.37 300 57.26 370 54.79 440 52.07 510 49.20 240 59.10 310 56.93 380 54.41 450 51.66 520 48.78 250 58.81 32Q 56.58 390 54.03 460 51.26 530 48.36 260 58.52 330 56.24 400 53.64 470 50.85 5*0 47.94 270 58.21 340 55.88 410 53.26 480 50.44 550 47.52 858 STEAM. Expansion of "Water. (Kopp : corrected by Porter.) Cent. Fahr. Volume. Cent. Fahr. Volume. Cent. Fahr. Volume. 40 39.2° 1.00000 35° 95° 1.00586 70° 158° 1.02241 5 41 1.00001 40 104 1.00767 75 167 1.02548 10 50 1.00025 45 113 1.00967 80 176 1.02872 15 59 1.00083 50 122 1.01186 85 185 1.03213 20 68 1.00171 55 131 1.01423 90 194 1.03570 25 77 1.00286 60 140 1.01678 95 203 1.03943 30 86 1.00425 65 149 1.01951 100 212 1.04332 Water for Boiler Feed. (Hunt and Clapp, A. I. M. E., 188S.) Water containing more than 5 parts per 100,000 of free sulphuric or nitric acid is liable to cause serious corrosion, not only of the metal of the boiler itself, but of the pipes, cylinders, pistons, and valves with which the steam comes in contact. The total residue in water used for making steam causes the interior lin- ings of boilers to become coated, and often produces a dangerous hard scale, which prevents the cooling action of the water from protecting the metal against burning. Lime and magnesia bicarbonates in water lose their excess of carbonic acid on boiling, and often, especially when the water contains sulphuric acid, produce, with the other solid residues constantly being formed by the evaporation, a very hard and insoluble scale. A larger amount than 100 parts per 100,000 of total solid residue will ordinarily cause troublesome scale, and should condemn the water for use in steam boilers, unless a bet- ter supply can be obtained. The following is a tabulated form of the causes of trouble with water for steam purposes, and the proposed remedies, given by Prof. L. M. Norton. CAUSES OF INCRUSTATION. 1. Deposition of suspended matter. 2. Deposition of deposed salts from concentration. 3. Deposition of carbonates of lime and magnesia by boiling off carbonic acid, which holds them in solution. 4. Deposition of sulphates of lime, because sulphate of /ime is but slightly soluble in cold water, less soluble in hot water, insoluble above 270° F. 5 Deposition of magnesia, because magnesium salts decompose at high temperature. 6. Deposition of lime soap, iron soap, etc., formed by saponification of grease. MEANS FOR PREVENTING INCRUSTATION. 1. Filtration. 2. Blowing off. 3. Use of internal collecting apparatus or devices for directing the circu- lation. 4. Heating feed-water. 859 5. Chemical or other treatment of water in boiler. 6. Introduction of zinc into boiler. 7. Chemical treatment of water outside of boiier. TABULAR VIEW. Trouble. Incrustation. Troublesome Substance. Sediment, mud, clay, etc. Readily soluble salts. Bicarbonates of lime, magnesia, iron. Sulphate of lime. " Chloride and sulphate of magne- ) „ sium. 8 } Corrosion. Carbonate of soda in large ) amounts. ) Acid (in mine waters). Dissolved carbonic acid and oxy- gen. Grease (from condensed water). Organic matter (sewage). Organic matter. Priming. Corrosion. Priming. Corrosion. Remedy or Palliation. Filtration, Blowing off. Blowing off. ( Heating feed. Addition of \ caustic soda, lime, or ^ magnesia, etc. | Addition of carb. soda, { barium chloride, etc. ( Addition of carbonate of ( soda, etc. (Addition of barium chlo- ( ride, etc. ("Heating feed. Addition < of caustic soda, slacked [_ lime, etc. f Slacked lime and filtering, -{ Carbonate of soda. L. Substitute mineral oil. ( Precipitate with alum or ( ferric chloride and filter. Solubilities of Scale-making- Materials. (" Boiler Incrustation," F. J. Bowan.) The salts of lime and magnesia are the most common of the impurities found in water. Carbonate of lime is held in solution in fresh water by an excess of carbonic acid. By heating the water the excess of carbonic acid is driven off and the greater part of the carbonate precipitated. At ordi- nary temperatures carbonate of lime is soluble in from 16,000 to '24,000 times its volume of water ; at 212° F. it is but slightly soluble, and at 290° F. (43 lbs. pressure) it is insoluble. The solubility of sulphate of lime is also affected by the temperature ; according to Kegnault, its greatest solubility is at 95° F., where it dissolves in 393 times its weight of water ; at 212: F. it is only soluble in 460 times its weight of water, and according to M. Coute, it is insoluble at 290° F. Carbonate of magnesia usnally exists in much smaller quantity than the salts of lime. The effect of temperature on its solubility is similar to that of carbonate of lime. Prof. R. H. Thurston, in his " Manual of Steam Boilers," p. 261, states that: The temperatures at which calcareous matters are precipitated are : Carbonate of lime betAveen 176° and 248° F. Sulphate of lime between 284c and 424° F. Chloride of magnesium between 212° and 257° F. Chloride of sodium between 324° and 304° F. 860 " Incrustation and sediment," Prof. Thurston says, " are deposited in boilers, the one by the precipitation of mineral or other salts previously held in solution in the feed-water, the other by the deposition of mineral insoluble matters, usually earths, carried into it in suspension or me- chanical admixture. Occasionally also vegetable matter of a glutinous nature is held in solution in the feed-water, and, precipitated by heat or concentration, covers the heating-surfaces with a coating almost impermea- ble to heat, and hence liable to cause an over-heating that may be very dan- gerous to the structure. A powdery mineral deposit sometimes met with is equally dangerous, and for the same reason. The animal and vegetable oils and greases carried over from the condenser or feed-water heater are also very likely to cause trouble. Only mineral oils should be permitted to be thus introduced, and that in minimum quantity. Both the efficiency and the safety of the boiler are endangered by any of these deposits. "The only positive and certain remedy for incrustation and sediment once deposited is periodical removal by mechanical means, at sufficiently frequent intervals to insure against injury by too great accumulation. Be- tween times, some good may be done by special expedients suited to the individual case. No one process and no one antidote will suffice for all cases. " Where carbonate of lime exists, sal-ammoniac may be used as a pre- ventive of incrustation, a double decomposition occurring, resulting in the production of ammonium carbonate and calcium chloride — both of which are soluble, and the first of which is volatile. The bicarbonate may be in part precipitated before use by heating to the boiling-point, and thus break- ing up the salt and precipitating the insoluble carbonate. Solutions of caustic lime and metallic zinc act in the same manner. Waters containing tannic acid and the acid juices of oak, sumach, logwood, hemlock, and other woods, are sometimes employed, but are apt to injure the iron of the boiler, as may acetic or other acid contained in the various saccharine matters often introduced into the boiler to prevent scale, and which also make the lime-sulphate scale more troublesome than when clean. Organic matters should never be used. "The sulphate scale is sometimes attacked by the carbonate of soda, the products being a soluble sodium sulphate and a pulverulent insoluble cal- cium carbonate, which settles to the bottom like other sediments and is easily washed off the heating-surfaces. Barium chloride acts similarly, producing barium sulphate and calcium chloride. All the alkalies are used at times to reduce incrustations of calcium sulphate, as is pure crude petro- leum, the tannate of soda, and other chemicals. " The effect of incrustation and of deposits of various kinds is to enor- mously reduce the conducting power of beat inn-surfaces ; so much so, that the power, as well as the economic efficiency of a boiler, may become very greatly reduced below that for which it is rated, and the supply of steam furnished by it may become wholly inadequate to the requirements of the case. " It is estimated that a sixteenth of an inch thickness of hard ' scale' on the heating-surface of a boiler will cause a waste of nearly one-eighth its efficiency, and the waste increases as the square of its thickness. The boil- ers of steam vessels are peculiarly liable to injury from this cause where using salt water, and the introduction of the surface-condenser has been thus brought about as a remedy. Land boilers are subject to incrustation by the carbonate and other salts of lime, and by tbe deposit of sand or mud mechanically suspended in the feed-water." Kerosene oil ("Boiler Incrustation," KoAvan) has been used to advantage in removing and preventing incrustation. From extended experiments made on a 100 h. p. water tube boiler, fed with water containing 6.5 grains of solid matter per gallon, it Avas found that one quart kerosene oil per day was sufficient to keep the boiler entirely free from scale. Prior to the in- troduction of the kerosene oil, the Avater had a corrosive action upon some of the fittings attached to the boiler ; but after the oil had been used for a feAV months it Avas found that the corrosive action had ceased. It should be stated, however, tbat objection has been made to the intro- duction of kerosene oil into a boiler for the purpose of preventing incrusta- 861 tion, on account of the possibility of some of the oil passing with the steam into the cylinder of the engine, and neutralizing the effect of the lubricant in the cylinder. When oil is used to remove scale from steam-boilers, too much care can- not be exercised to make sure that it is free from grease or animal oil. Nothing but pure mineral oil should be used. Crude petroleum is one thing ; black oil, which may mean almost anything, is very likely to be something quite different. The action of grease in a boiler is peculiar. It does not dissolve in the water, nor does it decompose, neither does it remain on top of the water ; but it seems to form itself into " slugs," which at first seem to be slightly lighter than the water, so that the circulation of the water carries them about at will. After a short season of boiling, these " slugs," or suspended drops, acquire a certain degree of " stickiness," so that Avhen they come in contact Avith shell and flues of the boiler, they begin to adhere thereto. Then under the action of heat they begin the process of " varnishing " the interior of the boiler. The thinnest possible coating of this varnish is suf- ficient to bring about over-heating of the plates. The time when damage is most likely to occur is after the fires are banked, for then, the formation of steam being checked, the circulation of Avater stops, and the grease thus has an opportunity to settle on the bottom of the boiler and preA-ent contact of the Avater Avith the fire-sheets. Under these circmnstances, a very low degree of heat in the furnace is sufficient to over- heat the plates to such an extent that bulging is sure to occur. Zinc as a Scale Preventive. — Dr. Corbigny gives the folloAAdng hypoth- esis : he says that " the tAvo metals, iron and zinc, surrounded by AA^ater at a high temperature, form a A'oltaic pile with a single liquid, which sloAvly decomposes the water. The liberated oxygen combines Avith the most oxy- dizable metal, the zinc, and its hydrogen equiATalent is disengaged at the surface of the iron. There is thus generated over the AArhole extent of the iron influenced a very feeble but continuous current of hydrogen, and the bubbles of this gas isolate at each instant the metallic surface from the scale-forming substance. If there is but little of the latter, it is penetrated by these bubbles and reduced to mud ; if there is more, coherent scale is produced, Avhich, being kept off by the intervening stratum of hydrogen, takes the form of the iron surface Avithout adhering to it." Zinc, in the shape of blocks, slabs, or as shaATings inclosed in a perforated vessel, should be suspended throughout the Avater space of a boiler, care being used in getting perfect metallic contact betAveen the zinc and the boiler. It should not be suspended directly over the furnace, as the oxide might fall upon the surface and be the cause of the plate being over-heated. The quantity placed in a boiler should vary Avith the hardness of the Avater, and the amount used, and should be measttred by the surface presented. Generally one square inch of surface for every 50 lbs. Avater in the boiler is sufficient. The British Admiralty recommends the reneAving of the blocks Avkenever the decay of the zinc has penetrated the slab to a depth of \ inch beloAV the surface. Purification of IVed- Water Ity Boiling-. Sulphates can be largely removed from feed-water by heating it to the tem- perature due to boiler pressure in a feed-Avater heater, or " live steam puri- fier " before introduction to boiler. This precipitates those salts in the heater and the water can then if necessary be pumped through a filter into the boiler. The feed-Avater i.3 first heated as hot as possible in the ordinary exhaust feed-water heater in Avhich the carbonates are precipitated, and then run through the purifier, Avhich is most generally a receptacle containing a number of shalloAv pans, that can be removed for cleaning, over which the feed-Avater is allowed to flow from one to the other in a thin sheet. Live steam at boiler-pressure is introduced into the purifier, heating the water to a temperature high enough to precipitate the salts Avhich form scale on the pans. This method of treating feed-Avater is said to largely increase the efficiency of a boiler plant by the almost complete avoidance of scale. Purification of f eed-AArater by filtration before introduction to the system is often practised Avith good results. 862 STEAM. Xal»l<* of Water Analyses. Grains per U. S. Gallon of 231 Cubic Inches. Where From. Buffalo, N. Y., Lake Erie .... Pittsburgh, Allegheny River . . Pittsburgh, Mononga'hela River . Pittsburgh, Pa., artesian well . . Milwaukee, Wisconsin River . . . Galveston, Texas, 1 Galveston, Texas, 2 Columbus, Ohio Washington, D. C, city supply . . Baltimore, Md., city supply . . . Sioux City, la., city supply .... Los Angeles, Cal., 1 Los Angeles, Cal., 2 Bay City, Michigan, Bay Bay City, Michigan, River .... Cincinnati, Ohio River Watertown, Conn Fort Wayne, Ind Wilmington, Del Wichita, Kansas Springfield, 111., 1 Springfield, 111., 2 Hillsboro, 111 Pueblo, Colo Long Island City, L. I Mississippi River, above Missouri River Mississippi River, below mouth of Missouri River Mississippi River at St. Louis, W. W. Hudson River, above Poughkeepsie, N. Y 5.66 0.37 1.06 23.45 6.23 13.68 21.79 20.76 2.87 2.77 19.76 10.12 3.72 8.47 4.84 3.88 1.47 8.78 10.04 14.14 12.99 5.47 14.56 4.32 4.0 Croton River, above Croton Dam N. Y Croton River water from service pipes in New York City. . . . Schuylkill River, above Philadelphia, Pa 3.32 3.78 5.12 5.71 4.67 13.52 29.15 11.74 3.27 0.65 1.24 5.84 12.59 10.36 33.06 0.78 4.51 6.22 6.02 25.91 7.40 4.31 2.97 16.15 28.0 7.41 6.94 0.58 0 58 0.64 18.41 1.76 326. 04 39S.99 7.02 Trace Trace 1.17 3.51 20*.48 120.78 1.79 1.76 3.51 4.29 24.34 1.97 1.56 2.39 1.20 16.0 0.50 1.36 1.54 0.37 0.78 1.04 20.14 Trace V.58 0.36 0.10 1.03 2.63 0.76 1.15 3.00 Trace 1.59 2.19 4.2 1.63 1.97 0.18 1.50 3.20 0.82 6.50 Trace 4.00 6.50 2.10 3.80 4.40 4.10 6.00 8.74 10.92 Trace 1.78 10.98 6.17 2.00 S.62 '..S3 1.22 1.57 10.76 1.92 1.36 1.30 15.86 9.85 39.31 353.8- 453.93 46.60 8.60 7.30 27.60 26.20 23.07 49.20 179.20 6.73 9.52 31.08 35.00 66.39 33.17 21.45 21.55 28.76 39.0 15.01 36.49 29.54 12.70 7.72 3.72 j 4.24 I 863 Feed-Pumps. These should be at least double the capacity found by calculation from tbe amount of water required for the engines, to allow for blowing off, leak- age, slip in the pumps themselves, etc., and to enable the pump to keep [down steam in case of sudden stoppage of the engines when the tires hap- pen to be brisk, and in fact should be large enough to supply the boilers when run at their full capacity. In addition, for all important plants, there [should be either a duplicate feed-pump or an injector to act as stand-by in case of accident. The speed of the plunger or piston may be 50 feet per minute and should never exceed 100 feet per minute, else undue wear and tear of the valves results, and the efficiency is reduced. If the pump be re- quired to stand idle without continually working, the plunger or piston and rod should be of brass. If D = diameter of barrel in inches, S = stroke in inches, n =z number of useful strokes per minute, w = cubic feet of water pumped per hour, W= lbs. of water pumped per hour ; w = 1.7 D2S n. n ~ 36.6 ' If S n = 50, fF =1.361)2, . and D=VZ. Rubber valves may be used for cold water, but brass, rubber composition, or other suitable material is required for hot water or oil. If a new pump will not start, it may be due to its imperfect connections or temporary stiffness of pump. Unless the suction lift and length of supply pipe be moderate, a foot-valve, a charging connection, and a vacuum chamber are desirable. The suction- pipe must be entirely free from air leakage. If the pump refuses to start lifting water with full pressure on, on account of the air in the pump-cham- ber not being dislodged, but only compressed each stroke, arrange for run- ning without pressure until the air is expelled and water flows. This is done with a check-valve in the delivery-pipe, and a waste delivery which may be closed when water flows. Pumping- Mot Water. — With a free suction-pipe, any good pump fitted with metal valves and with hot-water packing will pump water hav- ing a temperature of 212°, or higher, if so placed that the water will flow into it. Robert D. Kinney, in " Power," gives the following formula for deter- i?11?.1]1^ ,to AVnat height water of temperatures below the boiling point can be lifted by suction. D — lift in feet, A ■= absolute pressure on surface of water ; if open to air = 14.7 lbs. B and W= constants. See table. 864 Water Temp. B Water Temp. B W. Degrees F. Degrees F. 40 0.122 62.42 130 2.215 61.56 50 0.178 62.41 140 2.879 61.37 60 0.254 62.37 150 3.708 61.18 70 0.360 62.31 160 4.731 60.98 80 0.503 62.23 170 5.985 60.77 90 0.693 62.13 180 7.511 60.55 100 0.942 62.02 190 9.335 60.32 110 1.267 61.89 200 11.526 60.07 120 1.6S5 61.74 210 14.127 59.82 Speed of Witter through Pump-Passages and Valves. The speed of water flowing through pipes and passages in pumps varies from 100 to 200 feet per minute. The loss from friction will be considerable if the higher speed is exceeded. The area of valves should be sufficient to permit the water to pass at a speed hot exceeding 250 feet per minute. The amount of steam which an average engine will require per indicated horse-power is usually taken at 30 pounds. It varies widely, however, from about 12 pounds in the best class of triple expansion condensing engines up to considerably over 90 pounds in many direct-acting pumps. Where an engine is overloaded or underloaded more water per horse-power will be re- quired than when operated at rated capacity. Horizontal tubular boilers will evaporate on an average from 2 to 3 pounds of water per square foot heating-surface per hour, but may be forced up to 6 pounds if the grate sur- face is too large or the draught too great for economical working. Sizes of Direct-acting- Pumps. The two following tables are selected as representing the two common types of direct-acting pump, viz., the single-cylinder and the duplex. Efficiency of Small Direct-acting- Pumps. In "Reports of Judges of Philadelphia Exhibition," 1876, Group xx., Chas. E. Emery says : " Experiments made with steam-pumps at the Amer- ican Institute Exhibition of 1867 showed that average size steam-pumps do not, on the average, utilize more than 50 per cent of the indicated power in the steam cylinders, the remainder being absorbed in the friction of the en- gine, but more particularly in the passage of the water through the pump. Again, all ordinary steam-pumps for miscellaneous use, require that the steam-cylinder shall have three to four times the area of the water-cylinder to give sufficient power when the steam is accidentally low ; hence, as such pumps usually work against the atmospheric pressure, the net or effective pressure forms a small percentage of the total pressure, which, with the lame extent of radiating surface exposed and the total absence of expansion, makes the expenditure of steam very large. One pump tested required 120 pounds weight of steam per indicated horse-power per hour, and it is be- lieved that the cost will rarely fall below 60 pounds; and as only 50 per cent of the indicated power is utilized, it may be safely stated that ordinary steam pumps rarely require less than 120 pounds of steam per hour for each horse-power utilized in raising water, equivalent to a duty of only 15,000.000 foot pounds per 100 pounds of coal. With larger steam-pumps, particularly when they are proportioned for the work to be done, the duty will be mate- rially increased. 865 Single-Cylinder ^Direct-acting- Pump. (Standard Sizes for ordinary service.) .3 , © ft Diameter of .9 -a 1 p Capacity © © o O a per Minute £ A 3 © © © at Given .3 .3 © O ® © > o p O A ft £ Speed. M O "Si a © p © © ft 5 © ft © bo jt = © ^ MB o3 & p p* M x © rf M © O P ft ^ 6 % OQ O H H OQ P 02 s 4 3J 5 .14 300 130 18 33 9h | | 2 1J 4 4 5 .27 300 130 35 33 9h 2 li 5 4 7 .39 300 125 49 45J 15 s 1 3 2* 5} 5 7 .51 275 125 64 45J- 15 3 1 3 2J 5£ 5* 7 .72 275 125 90 45J 15 | 1 3 2* 7 7 10 1.64 250 110 180 58 17 1 1* 5 4 7£ ^2 10 1.91 250 110 210 58 17 n 5 4 7* 8 10 2.17 250 110 239 58 17 n 5 4 8 6 12 1.47 250 100 147 67 20i n 4 4 8 7 12 2.00 250 100 200 67 20i H 5 4 8 8 12 2.61 250 100 261 68 30' n 5 5 8 10 12 4.08 250 100 408 68 20J ij 8 8 10 8 L2 2.61 250 100 261 68i 30 1* 2* 2 5 5 10 10 12 4.08 250 100 408 68* 30 2 8 8 10 12 12 5.87 250 100 587 68* 30 2 8 8 12 10 12 4.08 250 100 408 64 24 21 8 8 12 10 18 6.12 200 70 428 "6Si 30 2 21 8 8 12 12 12 5.87 250 100 587 64 28* 2 2} 8 8 12 12 18 8.80 175 70 616 88 28i 2 2* 8 8 12 14 18 12.00 175 70 840 88 28l- 2 2J 8 8 14 10 12 4.08 250 100 408 69 30" 2 2* 8 8 14 10 18 6.12 175 70 428 93 25 2 2j 8 8 14 10 24 8.16 150 50 408 112 26 2 2i 2i 8 8 14 12 12 5.87 250 100 587 69 30 2 8 8 14 12 18 8.80 175 70 616 88 28J 2 2J 8 8 14 12 24 11.75 150 50 587 112 26" 2 2* 10 8 14 14 24 15.99 150 50 800 112 34 2 2J 12 10 14 16 16 13.92 175 80 1114 84 34 2 2* 12 10 14 16 24 20.88 150 50 1044 112 38 2 2* 12 10 16 14 IS 12.00 175 70 840 89 27 2 2j 8 8 16 14 24 15.99 150 50 800 109 34 2 2i 12 10 16 16 16 13.92 175 80 1114 85 34 2 24 12 10 16 16 24 20.88 150 50 1044 115 34 2 2* 12 10 16 18 24 26.43 125 50 1322 115 40 2 2} 14 12 18 16 24 20.88 125 50 1044 118 . 38 3 3* 12 10 18 18 24 26.43 125 50 1322 118 40 3 31 14 12 18 20 24 32.64 125 50 1632 118 40 3 3* 16 14 20 18 24 26.43 125 50 1322 118 40 3 3* 14 12 20 20 24 32.64 125 50 1632 118 40 3 3* 16 14 20 22 24 39.50 125 50 1975 120 40 3 3* 18 14 866 Duplex-Cylinder Direct-acting* Pump, (Standard sizes for ordinary service.) 53 I a> £ o -W M cc s o o a> a) §5 be .22 co A A i-l A 3 2 3 .04 4* 2f 4 .10 5* 3* 5 .20 6 4 6 .33 7* 7* 4* 6 .42 b 6 .51 it 10 .69 9 10 .93 10 6 10 1.22 10 7 10 1.66 12 7 10 1.66 14 7 10 1.66 12 8* 10 2.45 14 8* 10 2.45 16 8* 10 2.45 18* 8* 10 2.45 20 8* 10 2.45 12 10} 10 3.57 14 10* 10} 10 3.57 16 10 3.57 18* KM- 10 3.57 20 io|- 10 3.57 14 12 10 4.89 16 12 10 4.89 18* 12 10 4.89 20 12 10 4.89 18* 14 10 6.66 20 14 10 6.60 17 10 15 5.10 20 L2 15 7.34 20 15 15 11.47 25 15 15 11.47 J2 ® o 2 5^ 100 to 250 100 " 200 100 " 200 100 " 150 100 " 150 100 " 150 75 " 125 75 ' 125 75 " 125 125 125 125 125 125 125 77 75 " 125 75 "• 125 75 " 125 75 " 125 75 " 125 75 " 125 75 " 125 100 £i _ o . 0> d •d = t tjfc :- ~ - a-d - = f: "H £*3 a?cc P.43 ■d * bt^t ® 5° 0Q 53 33 S > -on* So| ~~Lv ^s O 53 ■- 5 50 a3 S ^"D rt^ - _ — a A 8 to 20 2-i 20" 40 4 40" 80 5 70" 100 5f 85" 125 6| 100" 150 7 100" 170 6| 135" 230 7* 180" 300 8* 245" 410 94 245 " 410 9£ 245" 410 9* 365" 610 12 365" 610 12 365" 610 12 365" 610 12 365" 610 12 530" 890 141 14} 530" S90 530" 890 14} 530" 890 14} 530" 890 14} 730" 1220 17 730" 1220 17 730" 1220 17 730" 1220 17 990" 1060 19| 990" KifiO 19| 510" 1020 14 730" 1460 17 1145 " 2290 21 1145 " 2290 21 Sizes of Pipes for Short Lengths. To be Increased as Length Increases. INJECTORS. Live Steam Injectors. W= water injected in pounds her hour. P = steam pressure in pounds per square inch. Z> = diameter of throat in inches. T= diameter of throat in millimeters. INJECTORS. 867 Then W= 1280 D*Vp = 1 .98 cP Vp The rule given by Rankine, " Steam Engine," p. 477, for finding the proper sectional area in square inches for the narrowest part of the nozzle is as follows : cubic feet per hour gross feed-water area = . 800 » pressure m atmospheres The expenditure of steam is about fourteen times the volume of water injected. The following table gives the water delivered for different sizes of injec- tors at different pressures ; but when the injector has to lift its water a de- duction must be made varying from 10 to 30 per cent according to the lift. Deliveries for Iiive Steam Injectors. o Pressure of Steam. 1 •%,$> ao5 M On a 14-foot lift : j ^ai'ts with, 2? ">f: stefm pressure. ( Works up to 175 lbs. steam pressure. On a 20-foot lift : j l^8 with, 42, "»• ste»m Pressure. ( Works up to 135 lbs. steam pressure. Whpri nnt lifHno- • i Starts with 14 lbs. steam pressure. When not lifting . j Workg up tQ 25Q lbg> gtea]£ pressure> Witli reed-Water at 100° F. On a 2-foot lift : On an 8-foot lift : Starts with 15 lbs. steam pressure. Works up to 210 lbs. steam pressure. Starts with 26 lbs. steam pressure. Works up to 160 lbs. steam pressure. r»„ ., i/i ts ~+ -\ivt- . ( Starts with 37 lbs. steam pressure. On a 14-toot lift . j Workg up to 120 lbs> gteam presSure. l Starts with 46 lbs. steam pressure. { Works up to 70 lbs. steam pressure. On a 20-foot lift : wv.™ r^*- nw„„ . S Starts witb 15 lbs. steam pressure. When not lifting : -j Workg up to 210 lbg_ gteam pressurep With Feed- Water at 130° F. /-> o -p t-Tfi- ( Starts Avith 20 lbs. steam pressure. On a 2-foot lift : j Works up to 185 lbs. steam pressure. r> a * 4- v*+ S Starts with 30 lbs. steam pressure. On an 8-toot lift : ^ Works up to 120 lbs. steam pressure. /~w ,. . .,... ( Starts with 42 lbs. steam pressure. On a 14-toot lift : j Works up to 75 lbs. steam pressure. .„ri ...... ( Starts with 20 lbs. steam pressure. When not lifting : j Works up to 185 lbs. steam pressure. WTi*li Feed- Water at 140° F. On a short lift or when not lifting, this injector will work with steam pressures from 20 'lbs. to 120 lbs., and on an 8-foot lift with steam pressures from 35 lbs. to 70 lbs. . Fxhanst Inf ectors working with exhaust steam from an engine, at about atmospheric pressure will deliver water against boiler pressure not exceeding 80 lbs. per square inch. The temperature of the Avater may I e as high as 190° F., while 12 per cent of the water delivered will be condensed steam. For pressures over 80 lbs. it is necessary to supplement the exhaust steam with a jet of live steam. Injector vs. Pump for Feeding- Boilers. The relative value of injectors, direct-acting steam pumps, and pumps driven from the engine, is a question of importance to all steam-users. The following table (" Stevens Indicator," 1888) has been calculated by D. S. Jacobs, M. F., from data obtained by experiment. It will be noticed that when feeding cold water direct to boilers, the injector has a slight economy, but when feeding through a heater a pump is much the most economical. INJECTORS. 869 Method of Supplying Feed-Water to Boiler. Relative Amount Saving of Fuel of Coal Required over the per Unit of Time, Amount Temperature of Feed-Water as the Amount for a Required delivered to the Pump or to the Direct-Acting when the Injector, 60° F. Rate of Evap- Pump, Feeding Boiler is Fed by oration of Boiler, 10 lbs. of Water at 6y Heating- Teed-"Water. (W. W. Christie.) In converting water at 32° F. into steam at atmospheric pressure, it must be raised to 212° F., the boiling point. The specific heat of water varies somewhat with its temperature, so that to raise a pound of water from 32° to 212° F. or 180° F., requires 180.8 heat units. To convert it into steam, after it has reached 212° F., requires 965.8 heat units, or in all 180.8 + 965.8 == 1146.6 units of heat, thermal units. The saving to be obtained by the use of waste heat, as exhaust steam, heating the water by transfer of some of its heat through metal walls, is calculated by this formula : » 872 r, ■ • « 100 (ft, — ft,) 100 (t2 — ty) , Gam m per cent = — r, , — - =r _r ; , ol, very nearly, ri — II i ii — f i ~r" 3-s in which H= total heat in steam at boiler pressure (above that in water at 32° F.) in B. T. U. h2 = heat in feed-water (above 32° F.) after heating. hx = heat in feed- water (above 32° F.) before beating. t2 = temperature of feed-water after heating °F. t^ =z temperature of feed-water before heating °F. given H= 1146.6, t2 = 212, fx = 112, or a difference of 100°; and we obtain by use of the above formula, gain in per cent = 9.37, or for 10° approximately .937 per cent, for 11° 1.03 per cent, so we may say that for every 11° F. added to the feed-water temperature by use of the exhaust steam, 1 per cent of fuel saving results. The table which follows is taken from " Power." Percentag-e of Saving- in Fuel tor Heating- feed-Water toy V^aste Steam, Steam at SO Pound!* Oaug-e Pressure. - o •2 3 = S ■ v Temperature of Water Entering Boiler, 120: V.HP 140° 150° 160° 170° 180° 190° 200° 210° 220° 250° 35° 7.24 8.09 8.95 9.89 10.66 11.52 12.38 13.24 14.09 14.95 15.81 19.40 40° 0.X4 7.6:) 8.56 9.42 10.28 11.14 12.00 12.87 13.73 14.59 15.45 18.89 45° 6.44 7.31) 8.16 9.03 9.90 10.76 11.62 12.49 13.36 14.22 15.09 18.37 50° (5.03 6.8!) 7.76 8.64 9.51 10.38 11.24 12.11 12.98 13.85 14.72 17.87 55° 5.63 6.4!) 7.37 8.24 9.11 9.99 10.85 11.73 12.60 13.48 14.35 17.38 60° 5.21 COS 6.96 7.84 8.72 9.60 10.47 11.34 12.22 13.10 13.98 16.8G 65° 4.80 5.(!7 6.56 7.44 8.32 9.20 10.08 10.96 11.84 12.72 13.60 16.35 70J 4.38 5.21) 6.15 7.03 7.92 8.80 9.68 10.57 11.45 12.34 13.22 15.84 75° 3.1)6 4.S4 5.73 6.62 7.51 8.40 9.28 10.17 11.06 11.95 12.84 15.33 80° 3.54 4.42 5.32 6.21 7.11 8.00 8.88 9.78 10.67 11.57 12.46 14.82 85° 3.11 4.00 4.90 5.80 6.70 7.59 8.48 9.38 10.28 11.18 12.07 14.32 90° 2. (IS 3.5S 4.48 5.38 6.28 7.18 8.07 8.98 9.88 10.78 11.68 13.81 95° 2.25 3.15 4.05 4.96 5.86 6.77 7.66 8.57 9.47 10.38 11.29 13.31 100° 1.81 2.71 3.62 4.53 5.44 6.35 7.25 8.16 9.07 9.98 10.88 12.80 Pump Exhaust. In many plants the only available exhaust steam comes from the steam pumps used for elevator service, boiler-feeding, etc. ; or in condensing plants from the air-pumps, water-supply, and boiler feed-pumps. It should also be remembered that all direct-acting steam pumps are large consumers of steam, taking several boiler h.p. for each indicated h. p., and that the ex- haust steam from them will heat about six times the same quantity by weight of cold water, from 50° to 212° F., and that these pumps, or the independent condenser pumps, are more economical when all the exhaust from them is used for heating feed-water than the best kind of triple expansion condens- ing engines. With the pumps all the heat not used in doing work can be conserved and returned to the boiler in the feed-water, whereas even with triple expansion engines at least 80 per cent of the total heat in the steam is carried away in the condensing water. While the supply of exhaust from these pumps may not be sufficient to raise the temperature to the highest point, yet the saving is large and con- stant. These results do not take any account of the purifying action in the "open" heaters on the feed-water, the improved condition of which, by di- minishing the average deposit within the boiler, materially increases both the boiler capacity and the economy ; while the more uniform temperature FUEL ECONOMIZER. 873 accompanying the use of a hot feed reduces the repairs and lengthens the life of all boilers. If the quantity of water passing through the heater is only what is re- quired to furnish steam for the engine from which the exhaust comes, more than four-fifths of this exhaust steam will remain uncondensed, and Avill thus become available for other purposes, such as heating buildings, dryer systems, etc. ; in which case the returns can be sent back to the boiler by suitable means. Fl'El ECOaiOMIZERS. Performance of a Green Economizer with a Smoky Coal. (D. K. Clark, S. E., p, 286.) From tests by M. W. Grosseteste, covering a period of three weeks on a Green economizer, using a smoke-making coal, with a constant rate of com- bustion under the boilers, it is apparent that there is a great advantage in cleaning the pipes daily — the elevation of temperature having been in- creased by it from 88° to 153°. In the third week, without cleaning, the ele- vation of temperature relapsed in three days to the level of tbe first week ; even on the first day it was quickly reduced by as much as half the extent of relapse. By cleaning the pipes daily an increased elevation of tempera- ture of 65° F. was obtained, whilst a gain of 6% was effected in the evapora- tive efficiency. The action of Green's economizer was tested by M. W. Grosseteste for a period of three weeks. The apparatus consists of four ranges of vertical pipes, 6i feet high, 3f inches in diameter outside, nine pipes in each range, connected at top and bottom by horizontal pipes. The water enters all the tubes from below, and leaves them from above. The system of pipes is enveloped in a brick casing, into which the gaseous products of combustion are introduced from above, and which they leave from below. The pipes are cleared of soot externally by automatic scrapers. The capacity for water is 24 cubic feet, and the total external heating-surface is 290 square feet. The apparatus is placed in connection witb a boiler having 355 square feet of surface. Green's Economizer. — Results of Experiments on its Efficiency as Affected by the State of the Surface. (W. Grosseteste.) Temperature of Feed- Temperatnre of Gas- water. eous Products. Time. February and March. Enter- Leav- Enter- Leav- ing Feed- ing Differ- ing ing Differ- Feed- ence. Feed- Feed- ence. heater. heater. heater. heater. Fahr. Fahr. Fahr. Fahr. Fahr. Fahr. 1st Week 73.5° 161.5° 88.0° 849° 261° 588° 2d Week 77.0 230.0 153.0 882 297 585 3d Week — Monday . . 73.4 196.0 122.6 831 284 547 Tuesday . . 73.4 181.4 108.0 871 309 562 Wednesday 79.0 178.0 99.0 Thursday . 80.6 170.6 90.0 952 329 623 Friday . . 80.6 169.0 88.4 889 338 551 Saturday 79.0 172.4 93.4 901 351 550 1st Week. Coal consumed per hour 214 lbs. Water evaporated from 32° F. per hour 1424 Water per pound of coal ...... 6.65 2d Week. 3d Week. 216 lbs. 213 lbs. 1525 1428 7.06 6.70 874 The Fuel Economizer Company, Matteawan, N.Y., describe the construc- tion of Green's economizer, thus: The economizer consists of a series of sets of cast-iron tubes about 4 inches in diameter and 9 feet in length, made in sections (of various widths) and connected by " top " and " bottom headers," these again being coupled by " top " and " bottom branch pipes " running lengthwise, one at the top and the other at the bottom, on opposite sides and outside the brick chamber which encloses the apparatus. The waste gases are led to the economizer by the ordinary flue from the boilers to the chimney. The feed-water is forced into the economizer by the boiler pump or in- jector, at the lower branch pipe nearest the point of exit of gases, and emerges from the economizer at the upper branch pipe nearest the point where the gases enter. Each tube is provided with a geared scraper, which travels continuously up and down the tubes at a slow rate of speed, the object being to keep the external surface clean and free from soot, a non-conductor of heat. The mechanism for working the scrapers is placed on the top of the econ- omizer, outside the chamber, and the motive power is supplied either by a belt from some convenient shaft or small independent engine or motor. The power required for operating the gearing, however, is very small. The apparatus is fitted with blow-off and safety valves, and a space is pro- vided at the bottom of the chamber for the collection of the soot, which is removed by the scrapers. One boiler plant equipped with the Green economizer gave, under test, these results. The total area of heating surface in the plant was 3,126 square feet, and the number of tubes in the economizer 160. The results were as follows: — Particulars of Test. Econo- mizer working, Dec. 15. Econo- mizer not working, Dec. 16. Duration of test hours Weight of dry coal consumed lbs. Percentage of' ash and refuse . . . per cent "Weight of coal consumed per hour per square foot grate surface lbs. Weight of water evaporated lbs. Horse-power developed on basis of 30 lbs. per h.p. fed at 100° and evaporated at 70 lbs., h.p. Average boiler pressure (above atmosphere), lbs. Average temperature of feed-water entering economizer deg. Fahr. Average temperature of feed-water entering boilers deg. Fahr. Number of degrees feed-water was heated by economizer deg. Fahr. Average temperature of flue gases entering economizer deg. Fahr. Average temperature of flue gases entering chimney deg. Fahr. Number degrees flue gases were cooled by econ- omizer deg. Fahr. Lbs. water evaporated per lb. of coal, as ob- served Equivalent evaporation per lb. of coal from and at 212° Percentage gained by using the economizer per cent 11.5 8,743 15.2 84,078 11.5 ,694 7.7 16.8 82,725 84.2 196.2 82.0 112. 435. 279. 452.0 156. 9.617 8.533 11.204 9.955 12.5 The steam in this test contained 1.3 per cent of moisture. I UEL ECONOMIZERS, 875 M. W. S. Hutton gives the following results of tests of a steam boiler with and without an economizer. With Econ- omizer. Without Econo- mizer. Duration of test, hours Weight of coal, pounds Steam pressure, pounds Temp, water entering economizer, degrees . . . " " " boiler, degrees Degrees feed-water heated by economizer . . . Temp, gases entering economizer, degrees . . . " " " chimney, degrees .... Degrees gases cooled by economizer Evaporation per lb. coal, from and at 212°, pounds Saving by economizer, per cent HI 7856 58 88 225 137 618 365 253 10.613 28.9 10282 57 Green's Fuel Economizer, — Clark gives the following average re- sults of comparative trials of three boilers at Wigan used with and without economizers : Coal per square foot of grate per hour . Water at 100° evaporated per hour . . Water at 212° per pound of coal . . . Without Economizers. . . 21.6 . . 73.55 . . 9.60 With Economizers. . 79.32 10.56 Showing that in burning equal quantities of coal per hour the rapidity of evaporation is increased 9.3% and the efficiency of evaporation 10% by the addition of the economizer. The average temperature of the gases and of the feed-water before and after passing the economizer were as follows : With 6-f t. grate. With 4-ft. grate. Before. After. Before. After. Average temperature of gases . . • 649 340 501 312 Average temperature of feed-water . 47 157 41 137 Taking averages of the two grates, to raise the temperature of the feed- water 100°, the gases were cooled down 250°. §EPARATORS. Carefully conducted experiments have shown that water, oil, or other liquids passing through pipes along with steam do not remain thoroughly mixed with the steam itself, but that the major portion of these liquids fol- lows the inner contour of the pipe, especially in the case of horizontal pipes. From this it would necessarily follow that a rightly designed separator to meet these conditions must interrupt the run of the liquid by breaking the continuity of the pipe, and offering a receptacle into which the liquid will flow freeiv, or fall bv gravity — that this appliance must further offer the opportunity for the liquid to come to rest out of the current of steam, for it is not enough to simplv provide a well or a tee in the pipe, since the current would jump or draw the liquid over this opening, especially if the velocity was high. It is also evident that means must be provided in this appliance for inter- rupting the progress of those particles of the liquid which are traveling in the current of the steam, and do this in such a way that these particles will 876 also be detained and allowed to fall into the receptacle provided, which receptacle must be fully protected from the action of the current of the steam ; otherwise, the separated particles of water or oil will be picked up and carried on past the separator. To prevent the current from jumping the liquid over the well, and to interrupt the forward movement of those particles traveling in or with the current, it follows that some obstruction must be interposed in the path of the current. Steam separators should always be placed as near as possible to the steam inlet to the cylinder of the engine. Oil separators are placed in the run of the exhaust pipe from engines and pumps, for the purpose of removing the oil from the steam before it is used in any way where the presence of oil would cause trouble. Prof. R. C. Carpenter conducted a series of tests on separators of several makes in 1891. The following table shows results under various conditions of moisture : A. Test with Steam of about 10% of Moisture. Tests with Varying Moisture. i Quality of Steam Before. Quality of Steam After. Efficiency per cent. Quality of Steam Before. Quality of Steam After. Average Efficiency. B A D C E F 87.0% 90.1 89.6 90.6 88.4 88.9 98.8% 98.0 95.8 93.7 90.2 92.1 90.8 80.0 59.6 33.0 15.5 28.8 66.1 to 97.5% 51.9 " 98 72.2 " 96.1 67.1 " 96.8 68.6 " 98.1 70.4 " 97.7 97.8 to 99 % 97.9 " 99.1 95.5 " 98.2 93.7 " 98.4 79.3 " 98.5 84.1 " 97.9 87.6 76.4 71.7 63.4 36.9 28.4 Conclusions from the tests were : 1. That no relation existed between the volume of the several separators and their efficiency. 2. No marked decrease in pressure was shown by any of the separators, the most being 1.7 lbs. in E. 3. Although changed direction, reduced velocity, and perhaps centrifugal force are necessary for good separation, still some means must be provided to lead the water out of the current of the steam. A test on a different separator from those given above was made by Mr. Charles H. Parker, at the Boston Edison Company's plant, in November, 1897, and the following results obtained : Length of run 3-4 hrs. Average pressure of steam 158 lbs. per sq. in. Temperature of upper thermometer in calorimeter on outlet of separator 368.5° F. Temperature of lower thermometer in calorimeter on outlet of separator 291.7° F. Normal temperature of lower thermometer, when steam is at rest 292.9° F. Degrees cooling as shown by lower thermometer . . . 1.2° F. Moisture in steam delivered by separator as shown by cooling of lower thermometer 06 per cent. "Water discharged from separator per hour 52 lbs. Steam and entrained water passing through engine, as shown by discharge from air pump of surface con- denser 7359 lbs. Steam and entrained water entering separator .... 7411 lbs. Moisture taken out by separator 72 Total moisture in steam (.06 plus .72) 78 per cent. Efficiency of separator 92.3 per cent. SAFETY VALVES. Oil SAFETY VALVES. Calculation of Weight, etc., for lever Safety- Valve. Let JF= weight of ball at end of lever, in pounds ; ■w = weight of lever itself, in pounds ; V=: weight of valve and spindle, in pounds ; L = distance between fulcrum and center of ball, in inches ; I =z distance between fulcrum and center of valve, m inches ; g — distance between fulcrum and center of gravity of lever, in inches; A = area of valve, in square inches ; . P = pressure of steam, in pounds per square inch at whicU valve will open. Then PAxl = W X L 4- w X g + VXl; whence P - WL + wg + VI Al PAl- -wg — VI L PAl- -tog — VI Example. — Diameter of valve, 4 inches ; distance from fulcrum to center of ball, 36 inches ; to center of valve, 4 inches ; to center of gravity of lever, 16 inches ; weight of valve and spindle, 6 lbs. ; weight of lever, 10 lbs. ; re- quired the weight of ball to make the blowing-olf pressure 100 lbs. per square inch ; area of 4-inch valve = 12.566 square inches. Then Pjjl — tyg— VI _ 100 X 12.566 X4— 10x16 — 6x4 _ 134.5 lbs. Rules Governing- Safety- Valves. (Rule of U. S. Supervising Inspectors of Steam-vessels as amended 1894.) The distance from the fulcrum to the valve-stem must in no case be less than the diameter of the valve-opening ; the length of the lever must not be more than ten times the distance from the fulcrum to the valve-stem ; the width of the bearings of the fulcrum must not be less than three-quarters of an inch ; the length of the fulcrum-link must not be less than four inches; the lever and fulcrum-link must be made of wrought iron or steel, and the knife-edged fulcrum points and the bearings for these points must be made of steel and hardened ; the valve must be guided by its spindle, both above and below the ground seat and above the lever, through supports either made of composition (gun-metal) or bushed with it ; and the spindle must fit loosely in the bearings or supports. Lever safety-valves to be attached to marine boilers shall have an area of not less than 1 square inch to 2 square feet of the grate surface in the boiler, and the seats of all such safety-valves shall have an angle of inclina- tion of 45° to the center line of their axes. Spring-loaded safety-valves shall be required to have an area of not less than 1 square inch to 3 square feet of grate surface of the boiler, except as hereinafter otherwise provided for water-tube or coil and sectional boilers, and each spring-loaded valve shall be supplied with a lever that will raise the valve from its seat a distance of not less than that equal to one-eighth the diameter of the valve-opening, and the seats of all such safety-valves shall have an angle of inclination to the center line of their axes af 45°. All spring-loaded safety-valves for water-tube or coil and sectional boilers required to carry a steam-pressure exceeding 175 lbs. per square inch shall be required to have an area of not less than 1 square inch to 6 square feet of the grate surface of the boiler. Nothing herein shall be construed so as to prohibit the use of two safety-values on one water-tube or coil and sectional boiler, provided the combined area of such valves is equal to that required by rule for one such valve. Rule on Safety- Valves in Philadelphia Ordinances.— Every boiler when tired separately, and every set or series of boilers when placed over one tire, shall have attached thereto, without the interposition of any other valve, two or more safety-valves, the aggregate area of which shall have such relations to the area of the grate and the pressure within the boiler as is expressed in schedule A. Schedule A. — Least aggregate area of safety-valve (being the least sec- tional area for the discharge of steam) to be placed upon all stationary boilers with natural or chimney draught (see note a). A- 22-5g P + 8.62' in which A is area of combined safety-valves in inches ; G is area of grate in square feet ; P is pressure of steam in pounds per square inch to be carried in the boiler above the atmosphere. The following table gives the results of the formula for one square foot of grate, as applied to boilers used at different pressures : Pressures per square inch : 10 20 30 40 50 60 70 80 90 100 110 120 150 175 Valve area in square inches corresponding to one square foot of grate : 1.2 .79 .58 .46 .38 .33 .29 .25 .23 .21 .19 .17 .14 .12 [Note a.] — Where boilers have a forced or artificial draught, the inspec- tor must estimate the area of grate at the rate of one square foot of grate surface for each 16 lbs. of fuel burned on the average per hour. The various rules given to determine the proper area of a safety-valve do not take into account the effective discharge area of the valve. A correct rule should make the product of the diameter and lift proportional to the weight, of steam to be discharged. Mr. A. G. Brown (The Indicator and its Practical Working) gives the fol- lowing as the lift of the lever safety-valve for 100 lbs. gauge 'pressure. Tak- ing the effective area of opening at 70 per cent of the product of the rise and the circumference Diameter of valve, inches 2 1\ 3 Zl 4 4J 5 6 Rise of valve, inches . . .0583 .0523 .0507 .0492 .0478 .0462 .0446 .043 For " pop " safety-valves, Mr. BroAvn gives the following table for the rise, effective area, "and quantity of steam discharged per hour, taking the effective area at 50 per cent of the actual on account of the obstruction which the lip of the valve offers to the escape of the steam. Dia. value, in Lift, Inches. Area, sq.in. 1 n 2 21 3 3£ 4 H 5 6 .125 .150 .175 .200 .225 .250 .275 .300 .325 .375 .196 .354 .550 .785 1.061 1.375 1.728 2.121 2.553 3.535 Gauge- press. Steam discharged per hour, lbs. 30 lbs. 474 856 1330 1897 2563 3325 4178 5128 6173 8578 50 669 1209 1878 2680 3620 4695 5901 7242 8718 120/0 70 861 1556 2417 3450 4660 6144 7596 9324 11220 15535 90 1050 1897 2947 4207 5680 7370 9260 11365 13685 18945 100 1144 2065 3208 4580 6185 8322 10080 12375 14895 20625 120 1332 2405 3736 5332 7202 9342 11735 14410 17340 24015 140 1516 2738 4254 6070 8200 10635 13365 16405 19745 27340 160 1696 3064 4760 6794 9175 11900 14955 18355 22095 30595 180 1883 3400 5283 7540 10180 l :;•_';-,() 16595 20370 2*520 33950 200 2062 3724 5786 8258 11150 14465 18175 22310 26855 37185 If we also take 30 lbs. of steam per hour, at 100 lbs. gauge-pressure = h. p., we have from the above table : Diameter inches . 1 l£- 2 2£ 3 ft . 4 4£ 5 6 Horse-power . . 38 69 107 153 206 277 336 412 496 687 RULES FOR CONDUCTING BOILER TESTS. 879 A boiler having ample grate surface and strong draft may generate double the quantity of steam its rating calls for ; therefore in determining the proper size of safety-valve for a boiler this fact should be taken into consideration and the effective discharge of the valve be double the rated steam-producing capacity of the boiler. The Consolidated Safety-valve Co.'s circular gives the following rated capacity of its nickel-seat " pop " safety-valves : Size, in . . 1 H 1* 2 2h 3 3* 4 Q 5 5i Boiler f from 8 10 20 35 60 75 100 125 150 175 200 H.P. | to 10 15 30 50 75 100 125 150 175 200 275 BUIE8 JPOK COHDUCTIJIG BOILER TESTS. The Committee of the A. S. M. E. on Boiler-tests recommended the fol- lowing revised code of rules for conducting boiler trials. (Trans, vol. xx.) Code of 1897. Preliminaries to a Trial. I. Determine at the outset the specific object of the proposed trial, whether it be to ascertain the capacity of the boiler, its efficiency as a steam gener- ator, its efficiency and its defects under usual working conditions, the econ-. omy of some particular kind of fuel, or the effect of changes of design, proportion, or operation ; and prepare for the trial accordingly. II. Examine the boiler, both outside and inside ; ascertain the dimensions of grates, heating surfaces, and all important parts ; and make a full record, describing the same, and illustrating special features by sketches. The area of heating surfaces is to be computed from the outside diameter of water-tubes and the inside diameter of fire-tubes. All surfaces below the mean water level Avhichhave water on one side and products of combustion on the other are to be considered water-heating surface, and all surfaces above the mean water level which have steam on one side and products of combustion on the other are to be considered as superheating surface. III. Notice the general condition of the boiler and its equipment, and record such facts in relation thereto as bear upon the objects in view. If the object of the trial is to ascertain the maximum economy or capa- city of the boiler as a steam generator, the boiler and all its appurtenances should be put in first-class condition. Clean the heating surface inside and outside, remove clinkers from grates and from sides of the furnace. Re- move all dust, soot, and ashes from the chambers, smoke connections, and flues. Close air leaks in the masonry and poorly-fitted cleaning-doors. See that the damper will open wide and close tight. Test for air leaks by firing a few shovels of smoky fuel and immediately closing the damper, observing the escape of smoke through the crevices, or by passing the flame of a can- dle over cracks in the brickwork. IV. Determine the character of the coal to be used. For tests of the effi- ciency or capacity of the boiler for comparison with other boilers the coal should, if possible, be of some kind which is commercially regarded as a stan- dard. For New England and that portion of the countrv east of the Allegheny Mountains, good anthracite egg coal, containing not over 10 per cent of ash, and semi-bituminous Clearfield (Pa.), Cumberland (Md.), and Pocahontas (Va.) coals are thus regarded. West of the Allegheny Mountains, Poca- hontas (Va.), and New River (W. Va.) semi-bituminous,' and Youghiogheny or Pittsburg bituminous coals are recognized as standards.* Thereds no special grade of coal mined in the Western States which is widely recog- nized as of superior quality or considered as a standard coal for boiler test- ing. Big Muddy Lump, an Illinois coal mined in Jackson County, 111., is * These coals are selected because they are about the only coals which con- tain the essentials of excellence of quality, adaptability to various kinds of furnaces, grates, boilers, and methods of firing, and icide distribution and general accessibility in the markets. 880 sug.sje.sted as being of sufficiently high grade to answer the requirements in districts where it is more conveniently obtainable than the other coals men- tioned above. For tests made to determine the performance of a boiler with a particular kind of coal, such as may be specified in a contract for the sale of a boiler, the coal used should not be higher in ash and in moisture than that speoi- fied, since increase in ash and moisture above a stated amount is apt to cause a falling off of both capacity and economy in greater proportion than the proportion of such increase. V. Establish the correctness of all apparatus used in the test for weighing and measuring. These are : 1. Scales for weighing coal, ashes, and water. 2. Tanks, or water meters for measuring water. Water meters, as a rule, should only be used as a check on other measurements. For accurate work, the water should be weighed or measured in a tank. 3. Thermometers and pyrometers for taking temperatures of air, steam, feed-water, waste gases, etc. 4. Pressure gauges, draft gauges, etc. The kind and location of the various pieces of testing apparatus must be left to the judgment of the person conducting the test; always keeping in mind the main object, i.e., to obtain authentic data. VI. See that the boiler is thoroughly heated before the trial to its usual working temperature. If the boiler is new and of a form provided with a brick setting, it should be in regular use at least a week before the trial, so as to dry and heat the walls. If it has been laid off and become cold, it should be worked before the trial until the walls are well heated. VII. The boiler and connections should be proved to be free from leaks before beginning a test, and all water connections, including blow and extra feed pipes, should be disconnected, stopped with blank flanges, or bled ' through special openings beyond the valves, except the particular pipe through which water is to be fed to the boiler during the trial. During the test the blow-off and feed-pipes should remain exposed. If an injector is used, it should receive steam directly through a felted pipe from the boiler being tested.* If the water is metered after it passes the injector, its temperature should be taken at the point at which it enters the boiler. If the quantity is deter- mined before it goes to the injector, the temperature should be determined on the suction side of the injector, and if no change of temperature occurs other than that due to the injector, the temperature thus determined is properly that of the feed-water. When the temperature changes between the injector and the boiler, as by the use of a heater or by radiation, the temperature at which the water enters and leaves the injector and that at which it enters the boiler should all be taken. The final temperature cor- rected for the heat received from the injector will be the true feed-water temperature. Thus if the injector receives water at 50° and delivers it at 12u° into a heater which raises it to 210°, the corrected temperature is 210 — (120 — 50)= 140°. See that the steam main is so arranged that water of condensation can- not run back into the boiler. VIII. Starting and Stopping a Test. — A test should last at least ten hours of continuous running, but, if the rate of combustion exceeds 25 pounds of coal per square foot of grate per hour it may be stopped when a total of 250 pounds of coal has been burned per square foot of grate surface. A longer test may be made when it is desired to ascertain the effect of widely vary- ing conditions, or the performance of a boiler under the working conditions of a prolonged run. The conditions of the boiler and furnace in all respects should be, as nearly as possible, the same at the end as at the beginning of the test. The steam pressure should be the same ; the water level the * In feeding a. boiler undergoing test with an injector talcing steam from another boiler, or the main steam pipe from several boilers, the evaporative, results may be modified by a difference in the quality of the steam from such source compared with that supplied by the boiler being tested, and in some cases the connection to the injector may act as a drip for the main steam pipe. If it is known that the steam' from the main pipe is of the same quality as that furnished by the boiler undergoing the test, the steam may be taken from such main pipe. RULES FOR CONDUCTING BOILER TESTS. 881 same ; the fire upon the grates should he the same in quantity and condi- tion ; and the walls, flues, etc., should be of the same temperature. Two methods of obtaining the desired equality of conditions of the fire may be used, viz. : those which were called in the Code of 1885 " the standard method " and " the alternate method," the latter being employed where it is inconvenient to make use of the standard method. IX. Standard Method. — Steam being raised to the working pressure, remove rapidly all the fire from the grate, close the damper, clean the ash- pit, and as quickly as possible start a new fire with weighed wood and coal, noting the time and the water level while the water is in a quiescent state, just before lighting the fire. At the end of the test remove the whole fire, which has been burned low, clean the grates and ash-pit, and note the water level when the water is in a quiescent state, and record the time of hauling the fire. The water level shuuld be as nearly as possible tbe same as at the beginning of the test. If it is not the same, a correction should be made by computation, and not by operating the pump after the test is completed. X. Alternate Method. — The boiler being thoroughly heated by a prelimi- nary run, the fires are to be burned low and well cleaned. Note the amount of coal left on the grate as nearly as it can be estimated ; note the pressure of steam and the water level, and note this time as the time of starting the test. Fresh coal which has been weighed should now be fired. The ash- pits should be thoroughly cleaned at once after starting. Before the end of the test the fires should be burned low, just as before the start, and the fires cleaned in such a manner as to leave the bed of coal of the same depth, and in the same condition, on the grates, as at the start. The water level and steam pressures should previously be brought as nearly as possible to the same point as at the start, and the time of ending of the test should be noted just before fresh coal is fired. If the water level is not the same as at. the start, a correction should be made by computation, and not by operating the pump after the test is completed. XI. Uniformity of Conditions. — In all trials made to ascertain maximum economy or capacity, the conditions should be maintained uniformly con- stant. Arrangements should be made to dispose of the steam so that the rate of evaporation may be kept the same from beginning to end. This may be accomplished in a single boiler by carrying the steam through a waste steam pipe, the discharge from which can be regulated as desired. In a battery of boilers, in which only one is tested, the draft can be regu- lated on the remaining boilers, leaving the test boiler to work under a con- stant rate of production. Uniformity of conditions should prevail as to the pressure of steam, the height of water, the rate of evaporation, the thickness of fire, the times of firing and quantity of coal fired at one time, and as to the intervals between the times of cleaning the fires. XII. Keeping the Heco?-ds. — Take note of every event connected with the progress of the trial, however unimportant it may appear. Record the time of every occurrence and the time of taking every weight and every observation. The coal should be weighed and delivered to the fireman in equal propor- tions, each sufficient for not more than one hour's run, and a fresh portion should not be delivered until the previous one has all been fired. The time required to consume each portion should be noted, the time being recorded at the instant of firing the last of each portion. It is desirable that at the same time the amount of water fed into the boiler should be accurately noted and recorded, including the height of the water in the boiler, and the average pressure of steam and temperature of feed during the time. By thus recording the amount of water evaporated by successive portions of coal, the test may be divided into several periods if desired, and the degree of uniformity of combustion, evaporation, and economy analyzed for each period. In addition to these records of the coal and the feed-water, half hourly observations should be made of the temperature of the feed-water, of the flue gases, of the external air in the boiler-room, of the temperature of the furnace when a furnace pyrometer is used, also of the pressure of steam, and of the readings of the* instruments for determining the moisture in the steam. A log should be kept on properly prepared blanks containing columns for record of the various observations. When the " standard method " of starting and stopping the test is used, 882 the hourly rate of combustion and of evaporation and the horse-power may- be computed from the records taken during the time when the tires are in active condition. This time is somewhat less than the actual time which elapses between the beginning and end of the run. This method of computation is necessary, owing to the loss of time due to kindling the fire at the beginning and burning it out at the end. XIII. Quality of Steam. — The percentage of moisture in the steam should be determined by the use of either a throttling or a separating steam calor- imeter. The sampling nozzle should be placed in the vertical steam pipe rising from the boiler. It should be made of J-inch pipe, and should extend across the diameter of the steam pipe to within half an inch of the opposite side, being closed at the end and perforated with not less than twenty £-inch holes equally distributed along and around its cylindrical surface, but none of these holes should be nearer than \ inch to the inner side of the steam pipe. The calorimeter and the pipe leading to it should be well covered with felting. Whenever the indications of the throttling or separating calorimeter show that the percentage of moisture is irregular, or occasion- ally in excess of three per cent, the results should be checked by a steam separator placed in the steam pipe as close to the boiler as convenient, with a calorimeter in the steam pipe just beyond the outlet from the separator. The drip from the separator should be caught and weighed, and the per- centage of moisture computed therefrom added to that shown by the calorimeter. Superheating should be determined by means of a thermometer placed in a mercury well inserted in the steam pipe. The degree of superheating should be taken as the difference between the reading of the thermometer for superheated steam and the readings of the same thermometer for satu- rated steam at the same pressure as determined by a special experiment, and not by reference to steam tables. XIV. Sampling the Coal and Determining its Moisture. — As each barrow load or fresh portion of coal is taken from the coal pile, a representative shovelful is selected from it and placed in a barrel or box in a cool place and kept until the end of the trial. The samples are then mixed and broken into pieces not exceeding one inch in diameter, and reduced by the process of repeated quartering and crushing until a final sample weighing about five pounds is obtained, and the size of the larger pieces is such that they will pass through a sieve with J-inch, meshes. From this sample two one-quart, air-tight glass preserving jars, or other air-tight vessels which will prevent the escape of moisture from the sample, are to be promptly filled, and these samples are to be kept for subsequent determinations of moisture and of heating value, and for chemical analyses. During the process of quartering, when the sample has been reduced to about 100 pounds, a quarter to a half of it may be taken for an approximate determi- nation of moisture. This may be made by placing it in a shallow iron pan, not over three inches deep, carefully weighing it, and setting the pan in the hottest place that can be found on the brickwork of the boiler setting or flues, keeping it there for at least twelve hours, and then weighing it. The determination of moisture thus made is believed to be approximately accurate for anthracite and semi-bituminous coals, and also for Pittsburg or Youghiogheny coal ; but it cannot be relied upon for coals mined west of Pittsburg, or for other coals containing inherent moisture. For these latter coals it is important that a more accurate method be adopted. The method recommended by the Committee for all accurate tests, whatever the char- acter of the coal, is described as follows : Take one of the samples contained in the glass jars, and subject it to a thorough air-drying in a warm room, weighing it before and after, thereby determining the quantity of surface moisture it contains. Then crush the whole of it by running it through an ordinary coffee mill, adjusted so as to produce somewhat coarse grains (less than Jginch), thoroughly mix the crushed sample, select from it a portion of from 10 to 50 grams, weigh it in a balance Avhich will easily show a variation as small as 1 part in 1,000, and drv it in an air or sand bath at a temperature between 240 and 280 degrees Fahr. for one hour. Weigh it and record the loss, then heat and weigh it again repeatedly, at intervals of an hour or less, until the minimum weight has been reached and the weight begins to increase by oxidation of a por- tion of the coal. The difference between the original and the minimum weight is taken as the moisture in the air-dried coal. This moisture should RULES FOE CONDUCTING BOILER TESTS. 883 preferably be made on duplicate samples, and the results should agree within 0.3 to 0.4 of one per cent, the mean of the two determinations being taken as the correct result. The sum of the percentage of moisture thus found and the percentage of surface moisture previously determined is the total moisture. XV. Treatment of Ashes and Refuse. — The ashes and refuse are to be weighed in a dry state. For elaborate trials a sample of the same should be procured and analyzed. XVI. Calorific Tests and Analysis of Coal. — The quality of the fuel should be determined either by heat test or by analysis, or by both. The rational method of determining the total heat of combustion is to burn the sample of coal in an atmosphere of oxygen gas, the coal to be sampled as directed in Article XIV. of this code. The chemical analysis of the coal should be made only by an expert chemist. The total heat of combustion computed from the results of the ultimate analysis may be obtained by the use of Dulong's formula (with constants modified by recent determinations), viz. : 14,600 C -f- 62,000 f H— ^ ) + 4,000 S, in which C, H, 0, and S refer to the proportions of carbon, hydrogen, oxygen, and sulphur respectively, as determined by the ultimate analysis.* It is recommended that the analysis and the heat test be each made by two independent laboratories, and the mean of the two results, if there is any difference, be adopted as the correct figures. it is desirable that a proximate analysis should also be made to determine the relative proportions of volatile matter and fixed carbon in the coal. XVII. Analysis of Flue Gases.— The analysis of the flue gases is an espe- cially valuable method of determining the relative value of different meth- ods of firing, or of different kinds of furnaces. In making these analyses, great care should be taken to procure average samples — since the compo- sition is apt to vary at different points of the flue. The composition is also apt to vary from minute to minute, and for this reason the drawings of gas should last a considerable period of time. Where complete determinations are desired, the analyses should be intrusted to an expert chemist. For approximate determinations the Orsat or the Hempel apparatus may be used by the engineer. XVIII. Smoke Observations. — It is desirable to have a uniform system of determining and recording the quantity of smoke produced Avhere bitumi- nous coal is used. The system commonly employed is to express the degree of smokiness by means of percentages dependent upon the judgment of the observer. The Committee does not place much value upon a percentage method, because it depends so largely upon the personal element, but if this method is used, it is desirable that, so far as possible, a definition be given in explicit terms as to the basis and method employed in arriving at the percentage. XIX. Miscellaneous. — In tests for purposes of scientific research, in which the determination of all the variables entering into the test is de- sired, certain observations should be made which are in general unneces- sary for ordinary tests. These are the measurement of the air supply, the determination of its contained moisture, the determination of the amount of heat lost by radiation, of the amount of infiltration of air through the setting, and (by condensation of all the steam made by the boiler) of the total heat imparted to the water. As these determinations are not likely to be undertaken except by engi- neers of high scientific attainments, it is not deemed advisable to give directions for making them. XX. Calculations of Efficiency. — Two methods of defining and calculat- ing the efficiency of a* boiler are'recommended. They are : -. -i-^ • * j.-, x. -i Heat absorbed per lb. combustible 1. Efficiency of the boiler = — — * — -—- ■ Heating value of 1 lb. combustible o ua: • * ., t, ., j . Heat absorbed per lb. coal 2. Efficiency of the boiler and grate = — : -. = „ „ .,, =• Heating value of 1 lb. coal * Favre and Silberman give 14,544 B.T.U. per pound carbon; Berthelot 14,647 B.T. U. Favre and Silberman give 62,032 B.T. U. per pound hydro- gen; Thomson 61,816 B.T.U. 884 The first of these is sometimes called the efficiency based on combustible, and the second the efficiency based on coal. The first ifi recommended as a standard of comparison for all tests, and this is the one which is understood to be referred to when the word " efficiency " alone is used without qualifi- cation. The second, however, should be included in a report of a test, together with the first, whenever the object of the test is to determine the efficiency of the boiler and furnace together with the grate (or mechanical stoker), or to compare different furnaces, grates, fuels, or methods of firing. The heat absorbed per pound of combustible (or per pound coal) is to be calculated by multiplying the equivalent evaporation from and at 212° per pound combustible (or coal) by 965.7. (Appendix XXI.) XXI. The Heat Balance. — An approximate " heat balance," or statement of the distribution of the heating value of the coal among the several items of heat utilized and heat lost, may be included in the report of a test when analyses of the fuel and of the chimney gases have been made. It should be reported in the following form : Heat Balance, or Distribution of the Heating Value of the Combustible. Total Heat Value of 1 lb. of Combustible B. T. U. 1. Heat absorbed by the boiler = evaporation from and at 212° per pound of combustible x 965.7. 2. Loss due to moisture in coal = per cent of moisture re- ferred to combustible -f 100 X [(212 — /) + 966 + 0.48 (T — 212)] (t z= temperature of air in theboi.er-room, T= that of the flue gases). 3. Loss due to moisture formed by the burning of hydro- gen zr per cent of hydrogen to combustible -f- 100 X 9 X [(212 — t) + 966 + 0.48 (T — 212)]. 4.* Loss due to heat carried away in the dry chimney gases rr weight of gas per pound of combustible x 0.24 x (T-t). CO 5.f Loss due to incomplete combustion of carbon: ~C02-\-CO per cent Cin combustible 100 X 10,150. Loss due to unconsuined hydrogen and hydrocarbons, to heating the moisture in the air, to radiation, and un- accounted for. (Some of these losses may be sepa- rately itemized if data are obtained from which they may be calculated.) Totals * The weight of gas per pound of carbon burned may be calculated from the gas analysis as follows : Dry gas per pound carbon = 11 CO., ' in which C02, CO, O, and N are the percentage* by sampling and analyses of the gases i, to considerable errors, the result of this imateone. The heat balance itself is < as well as for the fact that it is not p< centage of unburned hydrogen orhyart The weight of dry gas per pound of the dry gas per pound of carbon by tht ble. and di riding hg 100. t CO, and CO are respect, rely the and carbonic oxide in the fine gases. generated by burning to carbonic acid Ionic oxide. + 8 0+7(C0+N), .„ 3 (C02 + CO) rolumc of the several gases. Js the .i the present slate of the art are liable calculation is usually only an approx- %lso only approximate for this reason, ossible in determine accurately theper- ncarbons in the flue gases. combustible is found by multiplying " percentage of carbon in the combusti- percentage by volume of carbonic acid The quantity 10,150 =z No. heat units one pound of carbon contained in car- RULES FOR CONDUCTING BOILER TESTS. 885 XXII. Report of the Trial. — The data and results should be reported in the manner given in either one of the two following tables, omitting lines where the tests have not been made as elaborately as provided for in such tables. Additional lines may be added for data relating to the specific object of the test. The extra lines should be classified under the headings provided in the tables, and numbered, as per preceding line, witb sub let- ters, a, b, etc. The Short Form of Report, Table No. 2, is recommended for commercial tests and as a convenient form of abridging the longer form for publication when saving of space is desirable. Table Wo. 1. Data and Results of Evaporative Test. Arranged in accordance with the complete form advised by the Boiler ! Test Committee of the American Society of Mechanical Engineers. I Made by of boiler at to determine Principal conditions| governing the trial Kind of fuel i Kind of furnace State of the weather , 1. Date of trial 2. Duration of trial hours Dimensions and Proportions. (A complete description of the boiler should be given on an annexed sheet.) Grate surface . . . width . . . length . . . area . . sq. ft. Water-heating surface " 5. Superheating surface " 6. Ratio of water-heating surface to grate surface 7. Ratio of minimum draft area to grate surface Average Pressures. 8. Steam pressure by gauge lbs. 9. Force of draft between damper and boiler ins. of water 10. Force of draft in furnace " " 11. Force of draft or blast in ash-pit " " Average Temperatures. 12. Of external air 13. Offireroom 14. Of steam 15. Of feed- water entering heater . . ; 16. Of feed-water entering economizer . j 17. Of feed-water entering boiler . . . ;18. Of escaping gases from boiler. . . i|19. Of escaping gases from economizer 20. Size and condition 21. Weight of wood used in lighting fire 22. Weight of coal as fired* I * Including equivalent of wood used in lighting the fire, not including un- burnt coal iritlulrairn from furnace at times of cleaning and at end of test. One pound 'of wood is taken to 'be equal to 0.4 pound of coal , or, in case greater accuracy is desired, as having a heat value equivalent to the evaporation of 6 pounds of xoater from and at 212° per pound (6 x 965.7 = 5,794 B.T.UJ. 886 23. Percentage of moisture in coal * ... per cent. 24. Total weight of dry coal consumed ... lbs. 25. Total ash and refuse lbs. 2(3. Total combustible consumed 27. Percentage of ash and refuse in dry coal per cent Proximate Analysis of Coal. Of Coal. Of Combustible. 28. Fixed carbon per cent. per cent. 29. Volatile matter " " 30. Moisture " 31. Ash " 100 per cent 100 per cent. 32. Sulphur, separately determined " " Ultimate Analysis of Dry Coal. 33. Carbon (C) per cent. 34. Hydrogen (R) " 35. Oxygen (O) 36. Nitrogen (N) " 37. Sulphur (S) 100 per cent. 38. Moisture in sample of coal as received " Analysis of Ash and Refuse. 39. Carbon per cent. 40. Earthy matter ... " Fuel per Hour. 41. Dry coal consumed per hour lbs. 42. Combustible consumed per hour " 43. Dry coal per square foot of grate surface per hour ... " 44. Combustible per square foot of water-heating surface per hour " Calorific Value of Fuel. 45. Calorific value by oxygen calorimeter, per lb. of dry coal . B. T. U. 46. Calorific value by oxygen calorimeter, per lb. of combustible " 47. Calorific value by analysis, per lb. of dry coalt " 48. Calorific value by analysis, per lb. of combustible .... " Quality of Steam. 49. Percentage of moisture in steam per cent, 50. Number of degrees of superheating deg. 51. Quality of steam (dry steam = unity) Water. 52. Total weight of water fed to boiler t lbs. 53. Equivalent water fed to boiler from and at 212° .... 54. Water actually evaporated, corrected for quality of steam 55. Factor of evaporation § 56. Equivalent water evaporated into dry steam from and at 212°. (Item 54 -f Item 55) " * This is the total moisture in the coal as found by drying it artificially. t See formula for calorific value under Article XVI. of Code. % Corrected for inequality of water level and of steam pressure at begin- ging and end of test. § Factor of evaporation = ~ ' in which H and h are respectively the total heat in steam of the average observed pressure, and in water of the aver- ! age observed temperature of the feed. RULES FOR CONDUCTING BOILER TESTS. 887 Water per Hour 57. Water evaporated per hour, corrected for quality of steam lbs. 58. Equivalent evaporation per hour from and at 212° .... " 59. Equivalent evaporation per hour from and at 212° per square foot of water-heating surface " Horse-Power. 60. Horse-power developed. (34£ lbs. of water evaporated per hour into dry steam from and at 212° equals one horse- power) * H.P. 61. Builders' rated horse-power " 62. Percentage of builders' rated horse-power developed . . . per cent. Economic Results. 63. Water apparently evaporated per lb. of coal under actual conditions. (Item 53 -j- Item 22) lbs. 64. Equivalent evaporation from and at 212° per lb. of coal (including moisture). (Item 56 -f- Item 22) " 65. Equivalent evaporation from and at 212° per lb. of dry coal. (Item 56 -f- Item 24) " 66. Equivalent evaporation from and at 212° per lb. of combus- tible. (Item 56 -± Item 26) " (If the equivalent evaporation, Items 64, 65, and 66, is not corrected for the quality of steam, the fact should be stated.) Efficiency. 67. Efficiency of the boiler ; heat absorbed by the boiler per lb. of combustible divided by the beat value of one lb. of combustible t per cent. 68. Efficiency of boiler, including the grate ; heat absorbed by the boiler, per lb. of dry coal fired, divided by the heat value of one lb. of dry coal % Cost of Evaporation. 69. Cost of coal per ton of 2,240 lbs. delivered in boiler room . $ 70. Cost of fuel for evaporating 1,000 lbs. of water under ob- served conditions $ 71. Cost of fuel used for evaporating 1,000 lbs. of water from and at 212° $ Smoke Observations. 72. Percentage of smoke as observed 73. Weight of soot j>er hour obtained from smoke meter . . . 74. Volume of soot obtained from smoke meter per hour . . Table Ufo. 3. Data and Results of Evaporative Test. Arranged in accordance with the Short Form advised by the Boiler Test Committee of the American Society of Mechanical Engineers. Made by on determine * Held to be the equivalent of 30 lbs. of water per hour evaporated from 100° Fahr. into dry steam at 70 lbs. gauge pressure. t In all cases where the word " combustible " is used, it means the coal with- out moisture and ash, but including all other constituents. It is the same as what is called in Europe " coal dry and free from ash." t The heat value of the coal is to be determined either by an oxygen calorim- eter or by calculation from ultimate analysis. When both methods are used the mean value is to be taken. Grate surface sq.ft. Water-heating surface " Superheating surface " Kind of fuel " Kind of furnace " Total Quantities. 1. Date of trial 2. Duration of trial hours. 3. Weight of coal as fired lbs. 4. Percentage of moisture in coal per cent. 5. Total weight of dry coal consumed lbs. 6. Total ash and refuse " 7. Percentage of ash and refuse in dry coal per cent. 8. Total weight of water fed to the boiler lbs. 9. Water actually evaporated, corrected for moisture or super- heat in steam " Hourly Quantities. 10. Dry coal consumed per hour lbs. 11. Dry coal per hour per square foot of grate surface ... " 12. Water fed per hour " 13. Equivalent water evaporated per hour from and at 212° corrected for quality of steam " 14. Equivalent water evaporated per square foot of water- heating hour " Average Pressures, Temperatures, etc. 15. Average boiler pressure lbs. per sq. ii\ 16. Average temperature of feed-water deg. 17. Average temperature of escaping gases " 18. Average force of draft between damper and boiler . . . ins. of watei1 19. Percentage of moisture in steam, or number of degrees of superheating Horse-Power. 20. Horse-power developed (Item 13 -^ 3U) H.P. 21. Builders' rated horse-power " 22. Percentage of builders' rated horse-power per cent. Economic Besults. 23. Water apparently evaporated per pound of coal under actual conditions. (Item 8 -=- Item 3) lbs. 24. Equivalent water actually evaporated from and at 212° per pound of coal as fired. (Item 9 -f- Item 3) " 25. Equivalent evaporation from and at 212° per pound of dry coal. (Item 9 -f- Item 5) " 26. Equivalent evaporation from and at 212° per pound of combustible. [Item 9 -)- (Item 5 — Item 6)] " (If Items 23, 24, and 25 are not corrected for quality of steam, the fact should be stated.) Efficiency . 27. Heating value of the coal per pound . B. T. U. 28. Efficiency of boiler (based on combustible) " 29. Efficiency of boiler, including grate (based on coal) ... " Cost of Evaporation. 30. Cost of coal per ton of 2,240 pounds delivered in boiler-room $ 31. Cost of coal required for evaporation of 1,000 pounds of water from and at 212° $ DETERMINATION OF MOISTURE. 889 DETEIlMiarATIOar OP THE MOISTURE I1V SKE1A The determination of the quality of steam supplied by a boiler is one of the most important items in a boiler test. The three conditions to be de- termined are : a. If the steam is saturated, i.e., contains the quantity of heat due to the pressure. b. If the steam is wet, i.e., contains less than the amount of heat due to the pressure. c. If the steam is superheated, i.e., contains more than the amount of heat due to the pressure. There are several methods of determining the quality of steam ; one being to condense all the steam evaporated by a boiler in a surface condenser, and weigh the condensing water, taking the temperature at its entrance to and exit from the condenser. Another is by use of a barrel calorimeter, in which a sample of the steam is condensed directly in a barrel partly filled with cold water, the added weight and temperature taken, and by use of a formula the quality of steam can be determined. Both the above-named methods are now practically obsolete, as their place has been taken by the throttling calorimeter, used for steam in which the moisture does not exceed 3 per cent, and the separating calorimeter, for steam containing a greater amount of moisture. Throttling- Calorimeter. In its simplest form this instrument can be made up from pipe fittings, the only special parts necessary being the throttling nozzle, which is readily made by boring out a piece of brass rod that is the same diameter as a half- inch steam pipe, leaving a small hole in one end, say Jg inch diameter. The inside end of the small hole should be tapered with the end of a drill so as not to cause eddies ; and the thermometer well, which is a small piece of brass pipe, plugged at one end, and fitted into a half-inch brushing to fit into place. The following cut hows the instrument as made up from fittings. The whole must be carefully covered with some non-conductor, as hair felt. Fig. 6. For more accurate work the instruments designed by George H. Barrus, M.E., and Prof. R. C. Carpenter, are to be preferred. Professor Carpenter's instrument is shown in the following cut, and differs from the primitive instrument previously described only by the addition of the manometer, 890 which determines the pressure of the steam above the atmosphere in the body of the calorimeter. With a free exit to the air the pressure in the calorimeter may be taken as that of the atmosphere. Carpenter's Throttling- Calorimeter. (\ size. Schaeffer & Budenberg.) Fig. 7. The perforated pipe for obtaining the sample of steam to be tested should preferably be inserted in a vertical pipe, and should reach nearly across its diameter. IHrections for "Use. — Connect as shown in the preceding cuts, till the thermometer cup with cylinder oil and insert the thermometer. Turn on the Globe valve for ten minutes or more in order to bring the tempera- ture of the instrument to full heat, after which note the reading of the ther- mometer in the calorimeter, and of the attached manometer or of a barometer. The steam gauge should be carefully calebrated to see that it is correct. A barometer reading taken at the time the calorimeter is in use, gives greater accuracy in working up the results than taking the average atmospheric pressure as 14.65 pounds. Pressure in pounds may be deter- mined from the mercury column of the barometer and manometer by divid- ing the inches rise by 2.03, or taking one pound for each two inches of mercury. Following is the formula for determining the quality of steam by use of the throttlihy calorimeter. ■ U= total heat in a pound of steam at the pressure in the pipe. h = total heat in a pound of steam at the pressure in the calorimeter. L = latent heat in a pound of steam at the pressure in the pipe. t = temperature in the calorimeter. b = temperature of boiling point at calorimeter pressure (taken as 212° with the " fittings" instrument). 0.48 = specific heat of superheated steam. x = quality of the steam. y z= percentage of moisture in the steam. „ = "-"- f«-»y.m. x — 100 — y. DETERMINATION OF MOISTURE. 891 If h be taken as 212°, as it can be with but slight error, then II — 1146.6 — .48 (t — 212) Following are tables calculated from the above formula. X 100. Moisture in Steam. Determinations by Throttling Calorimeter. Gauge-pressures. 5 10 20 30 40 50 60 70 75 80 85 90 Per Cent of Moisture in Steam 0° 0.51 0.90 1.54 2.06 2.50 2.90 3.24 3.56 3.71 3.86 3.99 4.13 UF 0.01 0.39 1.02 1.54 1.97 2.36 2.71 3.02 3.17 3.32 3.45 3.58 20° .51 1.02 1.45 1.83 2.17 2.48 2.63 2.77 2.90 3.03 30° .00 .50 .92 1.30 1.64 1.94 2.09 2.23 2.35 2.49 40° .39 .77 .24 1.10 .57 .03 1.40 .87 .33 1.55 1.01 .47 l!l5 .60 .06 1.80 1.26 .72 .17 1.94 50° 1.40 fin0 .85 70° .31 Gauge-pressure. I 100 110 120 130 140 150 160 170 180 190 200 250 Per Cent of Moisture in Steam 0° 4.39 4.63 4.85 5.08 5.29 5.49 5.68 5.87 6.05 6.22 6.39 7.16 10u 3.84 4.08 4.29 4.52 4.73 4.93 5.12 5.30 5.48 5.65 5. 8L 6.58 20u 3.29 3.52 3.74 3.96 4.17 4.37 4.56 4.74 4.91 5.08 5.25 6.00 30° 2.74 2.97 3.18 3.41 3.61 3.80 3.99 4.17 4.34 4.51 4.67 5.41 40° 2.19 2.42 2.63 2.85 3.05 3.24 3.43 3.61 3.78 3.94 4.10 4.83 50° 1.64 1.87 2.08 2.29 2.49 2.68 2.87 3.04 3.21 3.37 3.5o 4.25 60° 1.09 1.32 1.52 1.74 1.93 2.12 2.30 2.48 2.64 2.8( 2.96 3.67 70u .55 .77 .97 1.18 1.38 1.56 1.74 1.91 2.07 2.23 2.3f- 3.09 80-' .00 .22 .42 .63 .82 1.00 1.18 1.34 1.50 1.66 1.8' 2.51 90° .07 .26 .44 .61 .78 .94 1.09 1.2- 1.93 Kill' .05 .21 .37 .52 .67 .1C 1.34 110° .76 The easiest method of making the determinations from the observations is by use of the following diagram, prepared by Professor Carpenter. Find in the vertical column at the left the pressure observed in the main pipe -4- atmospheric pressure (the absolute pressure), then move hori- zontally to the right until over the line giving the degree of superheat (t — b), and the quality of steam will be found in a curve corresponding to one of those shown, and which may be interpolated where results do not come on one of the lines laid down. «y 892 180 170 160 150 140 / / * / / l / / / / / / ) / / 7 1 / i / / i / 1 / I / / / / 1 / / / / / / / / 1 1 , / ' / I / / / A> / / / / / / 1 / *l / /' / 1 / / j ' / / / / 1 j I 1 ' 1 ! / / 1 / 1 / / k// ; / / / i / / 1 / /' / / / / j / j / «; l / ' i / / i i ~l / / / '-/ / i 1 / / / / / 1 7 7 »\7 / / / / > */ / / / g/ f 1 / / / / / / V / 1 / / t / f .it / / 1 / / / N. / / I / / / / i / L"4/ / / / / / / / ' ? ' j/ / / f / / ft / / 4/ / / / / / £/ ?/ / / t l_ / / £ / / / / / / / 4* 7 10 / / / / / ' / / / / / w ,10 MP TIN 2 ER Q F >o AT ^E£ 2 UR UL Fj 0 E TS G. 2 N W 30 CA TH 2 LO T J0 RIIV HR 3 El OT )0 "ER TL 3 NG L0 C/ ■i 0 )RI 3 ME $0 TE 340 R. DETERMINATION OF MOISTURE. 895 duality of Steam Shown l*y Color of Issuing- Jet. Prof. J. E. Denton (Trans. A. S. M. E., vol. x.,p. 349) lias demon- strated that jets of steam escaping from an orifice in a boiler or steam reservoir snow unmistakable change of appearance to the eye ■when the steam varies less than one per cent from the condition of saturation either in the direction of "wetness or superheating. Conse- quently if a jet of steam flow from a boiler into the atmosphere under circumstances such that very lit- tle loss of heat occurs through radiation, etc., and the jet be transparent close to the orifice, of be even a grayish white color, the steam may be assumed to be so nearly dry that no portable condensing calorimeter will be capable of measuring the amount of water therein. If the jet be strongly white, the amount of water maybe roughly judged up to about 2 per cent, but beyond this a calorimeter only can deter- mine tbe exact amount of moist- ure. With a little experience any one may determine by this meth- od the conditions of steam within the above limits. A common brass pet cock may be used as an orifice, but it should, if possible, be set into the steam drum of the boiler and never be placed farther away from the latter than four feet, and then only when the in- termediate reservoir or pipe is well covered, for a very short travel of dry steam through a naked pipe will cause it to become perceptibly moist. FACTORS &W EVAPO- Fig. 10. Carpenter's New Evaporat- RA1IOI. ing Calorimeter. (Schaeffer & Bu- denberg.) In order to facilitate the calcu- lation of reducing the actual rate of evaporation of water from a certain temperature into steam of a cer- tain pressure, into the rate from water at 212° F. into steam of 212° a table of factors of evaporation is made up from the formula where 965.7 His the total heat of steam at the observed pressure, and h the total heat of feed-water of the observed temperature. 896 Table of factor* of Evaporation. (Compiled by W. Wallace Christie.) Gauge Pressure. 0 10 20 30 40 45 50 52 54 Temp, of lbs. lbs. lbs. lbs. lbs. lbs. lbs. lbs. lbs. Feed. 212° F. 1.0003 1.0088 1.0149 1.0197 1.0237 1.0254 1.0271 1.0277 1.0283 209 1.0035 1.0120 1.0180 1.0228 1.026X 1.0286 1.0301' 1.0309 1.0315 206 1.0066 1.0151 1.0212 1.0260 1.0299 1.0317 1.0334 1.0340 1.0346 203 1.0098 1.0183 1.0243 1.0291 1.0331 1.0349 1.0365 1.0372 1.0378 200 1.0129 1.0214 1.0275 1.0323 1.0362 1.0380 1.03! 17 1.0403 1.0409 197 1.0160 1.0246 1.0306 1.0354 1.0394 1.0412 1.0-128 1.0434 1.0441 194 1.0192 1.0277 1.0338 1.0385 1.0425 1.0443 1.0460 1.0466 1.0472 191 1.0223 1.0308 1.0369 1.0417 1.0457 1.0474 1.0491 1.0497 1.0503 188 1.0255 1.0340 1/M00 1.0448 1.0488 1.0506 1.0522 1.0528 1.0535 185 1.0286 1.0371 1.0432 1.0480 1.0519 1.0537 1.0554 1.0560 1.0566 182 1.0317 1.0403 1.0463 1.0511 1.0551 1.0568 1.05X5 1.0591 1.05! IS 179 1.0349 1.0434 1.0495 1.0542 1.0582 1.0600 1.0616 1.0623 1.0629 17G 1.03S0 1.0465 1.0526 1.0574 1.0613 1.0631 1.0648 1.0654 1.0660 173 1.0411 1.0497 1.0557 1.0605 1.0645 1.0663 1.0679 1.0685 1.0692 170 1.0443 1.0528 1.0589 1.0636 1.0676 1.0694 1.0710 1.0717 1.0723 167 1.0474 1.0559 1.0620 1.0668 1.0707 1.0725 1.0742 1.0748 1.0754 164 1.0505 1.0591 1.0651 1.0699 1.0739 1.0756 1.0773 1.0780 1.0786 161 1.0537 1.0622 1.0682 1.0730 1.0770 1.0788 1.0804 1.0811 1.0817 158 1.0568 1.0653 1.0714 1.0762 1.0801 1.0819 1.0S36 2.0842 1.0848 155 1.0599 1.0684 1.0745 1.0793 1.0833 1.0850 1.0867 1.0873 1.0880 152 1.0631 1.0716 1.0776 1.0824 1.0864 1.0882 1.0898 1.0905 1.0911 149 1.0662 1.0747 1.0808 1.0855 1.0*95 1.0913 1.0930 1.0936 1.0942 146 1.0693 1.0778 1.0839 1.0887 1.0926 1.0944 1.0961 1.0967 1.0973 143 1.0724 1.0810 1.0870 1.0918 1.0958 1.0975 1.0992 1.0998 1.1005 140 1.0756 1.0841 1.0901 1.0949 1.0989 1.1007 1.1023 1.1030 1.1036 137 1.0787 1.0872 1.0933 1.0980 1.1020 1.1038 1.1055 1.1061 1.1067 134 1.0818 1.0903 1.0964 1.1012 1.1051 1.1069 1.1086 1.1092 1.1098 131 1.0849 1.0934 1.0995 1.1043 1.1083 1.1100 1.1117 1.1123 1.1130 128 1.0881 1.0966 1.1026 1.1074 1.1114 1.1132 1.1148 1.1155 1.1161 125 1.0912 1.0997 1.10o7 1.1105 1.1145 1.1163 1.1179 1.1186 1.1192 122 1.0943 1.1028 1.1089 1.1136 1.1176 1.1194 1.1211 1.1217 1.1223 119 1.0974 1.1059 1.1120 1.1168 1.1207 1.1225 1.1242 1.1248 1.1254 116 1.1005 1.1090 1.1151 1.1199 1.1239 1.1256 1.1273 1.1279 1.1286 113 1.1036 1.1122 1.1182 1.1230 1.1270 1.1288 1.1304 1.1310 1.1317 110 1.1068 1.1153 1.1213 1.1261 1.1301 1.1319 1.1335 1.1342 1.1348 107 1.1099 1.1184 1.1245 1.1292 1.1332 1.1350 1.1366 1.1373 1.1379 104 1.1130 1.1215 1.1276 1.1323 1.1 30:; 1.1381 1.1398 1.1404 1.1410 101 1.1161 1.1246 1.1307 1.1355 1.1394 1.1412 1.1429 1.1435 1.1441 98 1.1192 1.1277 1.1338 1.1386 1.1426 1.1443 1.1460 1.1466 1.1473 95 1.1223 1.1309 1.1369 1.1417 1.1457 1.1475 1.1491 1.1497 1 1504 92 1.1255 1.1340 1.1400 1.1448 1.1488 1.1506 1.1522 1.1529 1.1535 89 1.1286 1.1371 1.1431 1.1479 1.1519 1.1537 1.1553 1.1560 1.1506 86 1.1317 1.1402 1.1463 1.1510 1.1550 1.1568 1.1584 1.1591 1.1597 83 1.1348 11433 1.1494 1.1541 1.1581 1.1599 1.1616 1.1622 1.1628 80 1.1379 1.1464 1.1525 1.1573 1.1612 1.1630 1.1647 1.1653 1.1659 77 1.1410 1.1495 1.1556 1.1604 1.1(344 1.1661 1.1678 1.1684 1.1690 74 1.1441 1.1526 1.1587 1.1635 1.1675 1.1692 1.1709 1.1715 1.1722 71 1.1472 1.1558 1.1618 1.1666 1.1706 1.1723 1.1740 1.1746 1.1753 68 1.1504 1.1589 1.1649 1.1697 1.1737 1.1755 1.1771 1.1778 1.1784 65 1.1535 1.1620 1.1680 1.1728 1.1768 1.1786 1.1802 1.1809 1.1815 62 1.1566 1.1651 1.1711 T 1759 1.1799 1.1817 1.1833 1.1840 1.1846 59 1.1597 1.1682 1.1743 1.1790 1.1830 1.1848 1.1864 1.1871 1.1877 56 1.1628 1.1713 1.1774 1.1821 1.1861 1.1879 1.1896 1.1902 1.1908 53 1.1659 1.1744 1.1805 1.1852 1.1892 1.1910 1.1927 1.1933 1.1939 50 1.1690 1.1775 1.1836 1.18S4 1.1923 1.1941 1.1958 1.1964 1.1970 47 1.1721 1.1806 1.1867 1.1915 1.1954 1.1972 1.1989 1.1995 1.2001 44 1.1752 1.1837 1.1898 1.1946 1.1986 1.2003 1.2020 1.2026 1.2032 41 1.1783 1.1868 1.1929 1.1977 1.2017 1.2034 1.2051 1.2057 1.2064 38 1.1814 1.1900 1.1960 1.2008 1.204s 1.2065 1.2082 1.2088 1.2095 35 1.1845 1.1931 1.1991 1.2039 1.2079 1.20!)6 1.2113 1.2119 1.2126 ' 32 1.1876 1.1962 1.2022 1.2070 1.2110 1.2128 1.2144 1.2151 1.2157 FACTORS OF EVAPORATION. 897 Table of factors of Evaporation. Gauge Pressure. 56 58 60 65 70 75 80 85 90 95 Temp, of Feed. lbs. lbs. lbs. lbs. 11)315 lbs. lbs. lbs. 1~X)353 lbs. lbs. lbs. 212° F. 1.0290 1.0295 1.0301 1.0329 1.0341 1.0365 1.0376 1.0387 209 1.0321 1.0327 1.0333 1.0346 1.0360 1.0372 1.0385 1.0397 1.0408 1.0419 206 1.0352 1.0358 1.0364 1.0378 1.0391 1.0403 1.C416 1.0428 1.0439 1.0450 203 1.0384 1.0390 1.0396 1.0464 1.0423 1.0435 1.0448 1.0460 1.0471 1.0482 200 1.0415 1.0421 1.0427 1.0441 1.0454 1.0466 1.0479 1.0491 1.0502 1.0513 197 1.0447 1.0453 1.0458 1.0477 1.0486 1.0498 1.0511 1 .0522 1.0533 1.0544 194 1.0478 1.0484 1.0490 1.0504 1.0517 1.0529 1.0542 1.0553 1.0565 1.0576 191 1.0510 1.0515 1.0521 1.0535 1.0549 1.0561 1.0573 1.0585 1.0596 1.0607 188 1.0541 1.0547 1.0553 1.0566 1.0580 1.0592 1.0605 1.0616 1.0628 1.0639 185 1.0572 1.0578 1.0584 1.0598 1.0611 1.0623 1.0636 1.0648 1.0659 1.0670 182 1.0604 1.0610 1.0615 1.0629 1.0643 1.0655 1.0668 1.0679 1.0090 1.0701 179 1.0635 1.0641 1.0647 1.0660 1.0674 1.0686 1.0699 1.0710 1.0722 1.0733 176 1.0666 1.0672 1.0678 1.0092 1.0705 1.0717 1.0730 1.0742 1.0753 1.0764 173 1.0698 1.0704 1.0709 1.0723 1.0737 1.0749 1.0762 1.0773 1.0784 1.0795 170 1.0729 1.0735 1.0741 1.0754 1.0768 1.0780 1.0793 1.0804 1.0816 1.0827 167 1.0760 1.0766 1.0772 1.0786 1.0799 1.0811 1.0824 1.0836 1.0847 1.0858 164 1.0792 1.0798 1.0803 1.0817 1.0831 1.0843 1.0S5C 1.0867 1.0878 1.0889 161 1.0823 1.0829 1.0835 1.0848 1.0862 1.0874 1.0887 1.089S 1.0910 1.0921 158 1.0854 1.0860 1.0866 1.0880 1.0893 1.0905 1.0916 1.0929 1.0941 1.0952 155 1.0886 1.0892 1.0897 1.0911 1.0925 1.0937 1.0941 1.0961 1.0972 1.0983 152 1.0917 1.0923 1.0929 1.0942 1.0956 1.0968 1.0981 1.0992 1.1004 1.1015 149 1.0948 1.0954 1.0960 1.0974 1.0987 1.0999 1.1011 1.102; 1.1035 1.1046 146 1.0979 1.0985 1.0991 1.1005 1.1018 1.1030 1.1043 1.1055 1.1066 1.1077 143 1.1011 1.1017 1.1022 1.1036 1.1050 1.1062 1.1074 1.1086 1.1097 1.1108 140 1.1042 1.1048 1.1054 1.1067 1.1081 1.1093 1.1HK 1.1117 1.1129 1.1140 137 1.1073 1.1079 1.1085 1.1099 1.1112 1.1124 1.1137 1.1148 1.1160 1.1171 134 1.1104 1.1110 1.1116 1.1130 1.1143 1.1155 1.1168 1.1180 1.1191 1.1202 131 1.1136 1.1142 1.1147 1.1161 1.1175 1.1187 1.119S 1.1210 1.1222 1.1233 128 1.1167 1.1173 1.1179 1.1192 1.1206 1.1218 1.1231 1.1242 1.1253 1.1264 125 1.1198 1.1204 1.1210 1.1223 1.1237 1.1249 1.1261 1.1273 1.1285 1.1296 122 1.1229 1.1235 1.1241 1.1255 1.1268 1.1280 1.1293 1.1294 1.1316 1.1327 119 1.1260 1.1266 1.1272 1.1286 1.1299 1.1311 1.1324 1.1336 1.1347 1.1358 116 1.1292 1.1298 1.1303 1.1317 1.1331 1.1343 1.135E 1.1366 1.1378 1.1389 113 1.1323 1.1329 1.1334 1.1348 1.1362 1.1374 1.1387 1.1398 1.1409 1.1420 110 1.1354 1.1360 1.1366 1.1374 1.1393 1.1405 1.1418 1.1429 1.1441 1.1452 107 1.1385 1.1391 11397 1.1411 1.1424 1.1436 1.1448 1.1460 1.1472 1.1483 104 1.1416 1.1422 1.1428 1.1442 1.1455 1.1467 1.148C 1.1491 1.1503 1.1514 101 1.1447 1.1453 1.1459 1.1473 1.1486 1.1498 1.1511 1.1523 1.1534 1.1545 98 1.1479 1.1485 1.1490 1.1504 1.1518 1.1530 1.1541 1.1554 1.1565 1.1576 95 1.1510 1.1516 1.1521 1.1535 1.1549 1.1561 1.1574 1.1583 1.1596 1.1607 92 1.1541 1.1547 1.1553 1.1566 1.1580 1.1592 1.1605 1.1616 1.1628 1.1639 89 1.1572 1.1578 1.1584 1.1598 1.1611 1,1623 1.163C 1.1647 1.1659 1.1670 86 1.1603 1.1609 1.1615 1.1629 1.1642 1.1654 1.1667 1.1678 1.1690 1.1701 83 1.1634 1.1640 1.1646 1.1660 1.1673 1.1685 1.1698 1.1709 1.1721 1.1732 80 1.1665 1.1671 1.1677 1.1691 1.1704 1.1716 1.1729 1.1741 1.1752 1.1763 77 1.1696 1.1702 1.1708 1.1722 1.1735 1.1747 1.176C 1.1772 1.1783 1.1794 74 1.1728 1.1734 1.1739 1.1753 1.1767 1.1779 1.1791 1.1803 1.1814 1.1825 71 1.1759 1.1765 1.1770 1.1784 1.1798 1.1810 1.1823 1.1834 1.1845 1.1856 68 1.1790 1.1796 1.1802 1.1815 1.1829 1.1841 1.1854 1.1865 1.1877 1.1888 65 1.1821 1.1827 1.1833 1.1846 1.1860 1.1872 1.1885 1.1S96 1.1908 1.1919 62 1.1852 1.1858 1.1864 1.1877 1.1891 1.1903 1.1916 1.1927 1.1939 1.1950 59 1.1883 1.1889 1.1895 1.1909 1.1922 1.1934 1.1947 1.1958 1.1970 1,2981 56 1.1914 1.1920 1.1926 1.1940 1.1953 1.1965 1.1978 1.1989 1.2001 1.2012 53 1.1945 1.1951 1.1957 1.1971 1.1984 1.1996 1.2009 1.2020 1.2032 1.2043 50 1.1976 1.1982 1.1988 1.2002 1.2015 1.2027 1.2040 1.2052 1.2063 1.2074 47 1.2007 1.2013 1.2019 1.2033 1.2046 1.2058 1.2071 1.2083 1.2094 1.2105 44 1.2039 1.2044 1.2050 1.2064 1.2078 1.2090 1.2102 1.2114 1.2125 1.2136 41 1.2070 1.2076 1.2081 1.2095 1.2109 1.2121 1.2133 1.2145 1.2156 1.2167 38 1.2101 1.2107 1.2112 1.2126 1.2140 1.2162 1.2164 1.2176 1.2187 1.2198 35 1.2132 1.2138 1.2143 1.2157 1.2171 1.21831 1.2196 1.2207 1.2218 1.2229 32 1.2163 1.2169 1.2175 1.2188 1.221)2 1.2214| 1.2227 1.2239 1.2249 1.2260 898 Table of factors of Evaporation. Gauge Pressure. 100 105 115 125 135 145 155 165 185 Temp, of Feed. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. 212° F, 1.0397 1.0407 1.0427 1.0445 1 0462 1.0478 1.0493 1.0509 1.0536 209 1.0420 1.0438 1.0458 1.0476 1.0403 1.0509 1.0524 1.0540 1.0567 20G 1.0460 1.0470 1.0489 1.0510 1.0527 1.0543 1.0558 1.0574 1.0601 203 1.0192 1.0502 1.0521 1.0540 1 0557 1.0573 1.0588 1.0604 1.0631 200 1.0523 1.0533 1.0552 1.0571 1.05X8 1.0604 1.0619 1.0035 1.0062 197 1.0555 1.0565 1.0584 1.0602 1.0619 1.0635 1.0650 1.0666 1 .0093 194 1.0586 1.0596 1.0615 1.0635 1.0052 1.066S 1.0683 1.6699 1.0726 191 1.0617 1.0627 1.0647 1.0665 1.0082 1.069S 1.0713 1.0729 1.0756 188 1.0649 1.0659 1.0678 1.0696 1.0713 1.0729 1.0744 1.0760 1.0787 185 1.0680 1.0690 1.0709 1.0728 1.0745 1.0761 1.0776 1.0792 1.0819 182 1.0712 1.0722 1.0741 1.0759 1.0776 1.0792 1-0807 1.0823 1.0850 170 1.0743 1.0753 1.0772 1-0790 1.1 IS07 1.0823 1.0838 1.0854 1.0881 176 1.0774 1.0784 1.0803 1.0822 1.0839 1.0855 1-0870 1.0886 1.0913 173 1.0806 1.0816 1.0835 1.0S53 1.0870 1.0886 1.0901 1.0917 1.0944 170 1.0837 1.0847 1.0866 1.0884 1.0901 1.0917 1-0932 1.0948 1.0980 1.0975 167 1.0868 1.0S78 1.0897 10916 1.0933 1.0949 1.0964 1.1007 164 1.0900 1.0910 1.0929 1.0946 l .()!)(;:; 1.0979 1.0994 1.1010 1.1037 161 1.0031 1.0941 1.0960 1.0979 1.0996 1.1012 1.1027 1.1043 1.1070 158 1.0962 1.0972 1.0991 1.1010 1.1027 1.1043 1-1058 1.1074 1.1101 155 1.0993 1.1003 1.1023 1.1041 1.1058 1.1074 1.1089 1.1105 1.1132 152 1.1025 1.1035 1.1054 1.1073 1.1090 1.1107 1.1122 1.1138 1.1165 149 1.1056 1.1066 1.1085 1.1103 1.1120 1.1136 1.1151 1.1167 1.1194 146 1.1087 1.1097 1.1116 1.1135 1.1152 1.1168 1.1183 1.1199 1.1226 143 1.1118 1.1129 1.1148 1.1166 1.1183 1.1199 1.1214 1.1230 1.1257 140 1.1150 1.1160 1.1179 1.1197 1.1214 1.1230 1.1245 1.1261 1.1288 137 1.1181 1.1191 1.1210 1.1228 1.1245 11262 1.1277 1.1293 1.1320 134 1.1212 1.1222 1.1241 1.1260 1.1277 1.1293 1.1308 1.1324 1.1351 131 1.1243 L1253 1.1273 1.1291 1.1303 1.1324 1.1339 1.1355 1.1382 128 1.1275 1.1285 1.1304 1.1322 1,1339 1.1355 1.1370 1.1386 1.1413 125 1.1306 1.1316 1.1335 1.1353 1.1370 1.1386 1.1401 1.1417 1.1444 122 11337 1.1347 1.1366 1.1384 1.1401 1.1417 1.1438 1.1448 1.1475 119 1.1368 1.1378 1.1397 1.1415 1.1432 1.1449 1.1464 1.1480 1.1507 116 1.1399 1.1409 1.1429 1.1447 1.1464 1.1480 1.1495 1.1511 1.1538 113 1.1431 1.1441 1.1460 1.1478 1.1495 1.1511 1.1526 1.1542 1.1569 110 1.1462 1.1472 1.1491 1.1509 1.1516 1.1542 1.1557 1.1573 1.1600 107 1.1493 1.1503 1.1522 1.1540 1.1557 1.1573 1.1588 1.1604 1.1631 104 1.1524 1.1534 1.1553 1.1571 1.1588 1.1605 1.161! 1.1635 1.1062 101 1.1555 1.1565 1.1584 1.1602 1.1620 1.1636 1.1652 1.1668 1.1095 98 1.1586 1.1596 1.1616 1.1634 1.1651 1.1667 1.168: 1.1699 1.1726 95 1.1618 1.1628 1.1647 1.1665 1.1682 1.1698 1.1713 1.1729 1.1756 92 1.1640 1.1660 1.1678 1.1696 1.1713 1.1729 1.1744 1.1760 1.1787 89 1.1680 1.1690 1.1709 1.1727 1.1744 1.1760 1.1775 1.1791 1.1818 86 1.1711 1.1721 1.1740 1.1758 1.1775 1.1791 1.18(11 1.1822 1.1849 83 1.1742 1.1752 1.1771 1.1789 1.1800 1.1823 1.1837 1.1853 1.1880 80 1.1773 1.1783 1.1802 1.1820 1.1837 1.1854 1.1869 1.1885 1.1912 77 1.1804 1.1814 1.1834 1.1852 1.1869 1.1885 1.1900 1.1916 1.1913 74 1.1835 1.1845 1.1865 1.1883 1.1900 1.1916 1.1932 1.1948 1.1975 71 1.1867 1.1877 1.1896 1.1914 1.1931 1.1947 1.1961 1.1977 1.2004 68 1.1898 1.1908 1.1927 1.1945 1.1962 1.1978 1.1993 1.200! 1.2036 65 1.1929 1.1939 1.1958 1.1976 1.1993 1.2009 1.2024 1.2040 1.2067 62 1.1960 1.1970 1.1989 1.2007 1.2024 1.2040 1.2055 1.2071 1.2098 59 1.1991 1.2001 1.2020 1.2038 1.2055 1.2071 1.20X1 1.2102 1.2129 56 1.2022 1.2032 1.2051 1.2069 1.2086 1.2102 1.2117 1.2133 1.2160 53 1.2053 1.2063 1.2082 1.2100 1.2117 1.2134 1.2148 1.2164 1.2191 50 1.2084 1.2094 1.2113 1.2131 1.2148 1.2165 1.2180 1.2196 1.2223 47 1.2115 1.2125 1.2144 1.2163 1.2180 1.2196 1.2211 1.2227 1.2254 44 1.2146 1.2156 1.2176 1.2194 1.2211 1.2227 1.2242 1 .2258 1.2285 41 1.2177 1.2187 1.2207 1.2225 1.2242 1.2258 1.2273 1 .228! 1.2316 38 1.2208 1.2219 1 2238 1.2256 1.2273 1.2289 1.2304 1.2521 1.2347 35 1.2240 1.2250 1 2269 1.2287 1.2304 1,2320 1.2335 1.2351 1.2378 32 1.2271 1.2281 1.2300 1.2318 1.2335 1.2351 1.2566 1 .2382 1.2409 PROPERTIES OP SATURATED STEAM. 899 PKOPEHTIES OF SATTJUATKI* STEAM. (Compiled by W. W. Christie.) Pounds per Square Inch. , • a3 Heat Units in one Pound above 32° F. Volume. jo 6 6 'c 2 Cu. Ft. in 1 Cu. Ft. of Water. Cu. Ft, in one Lb. of Steam. 1 2 4 102. 12G.2 141.6 153.0 70.1 94.4 109.8 121.4 1042.9 102G.0 1015.2 1007.2 1113.0 1120.4 1125.1 1128. G 20G20 10720 732G 5G00 319.600 172.417 117.723 89.799 .0030 .0058 .0085 .0112 5 6 7 8 1G2.3 170.1 176.9 182.9 130.7 138.5 145.4 151.4 1000.7 995.2 990.4 986.2 1131.4 1133.8 1135.8 1137.7 4535 3814 3300 2910 72.792 61.311 53.000 46.771 .0137 .0163 .0189 .0214 9 10 11 12 188.3 193.2 197.7 201.9 156.9 161.9 1G6.5 170.7 982.4 978.9 975.7 972.8 1139.3 1140.8 1142.2 1143.5 2G07 2360 2157 19S8 41.858 37.904 34.G59 31.932 .0239 .0264 .0289 .0313 ' .304 1.3 13 14 15 1G 205.8 209.5 213.0 216.3 174.7 178.4 181.9 185.2 970.0 967.4 964.9 962.6 1144.7 1145.8 114G.9 1147.9 1846 1722 1012 1514 29.593 27.G24 25.858 24.335 .0337 .0362 .0387 .0413 2.3 3.3 4.3 5.3 17 18 19 20 219.4 222.3 225.2 227.9 188.4 191.4 194.2 197.0 960.4 958.3 956.3 954.4 1148.8 1149.7 1150.6 1151.4 1427 1350.6 1282.1 1220.3 22.985 21.781 20.701 19.725 .0437 .0462 .0487 .0511 G.3 7.3 8.3 9.3 21 22 23 24 230.5 233.0 235.4 237.7 199.6 202.2 204.6 207.0 952.5 950.8 949.0 947.4 1152.2 1153.0 1153.7 1154.4 1164.4 1113.5 10GG.9 1024.1 18.839 18.033 17.293 16.615 .0536 .05G1 .0585 .0G10 10.3 11.3 12.3 13.3 25 26 27 28 240.0 242.1 244.2 246.3 209.3 211.5 213. G 215.7 945.8 944.2 942.7 941.3 1155.1 1155.8 115G.4 1157.0 984.8 948.4 914.6 883.2 15.9S8 15.409 14.871 14.371 .0034 .0G58 .0G83 .0707 14.3 15.3 1G.3 17.3 29 30 31 32 248.3 250.2 252.1 253.9 217.7 219.7 221. G 223.5 939.9 938.9 937.1 935.9 1157.6 1158.2 1158.8 1159.3 854.0 82G.8 801.2 777.2 13.904 13.4(57 13.058 12.674 .0731 .0755 .0779 .0803 18.3 19.3 20.3 21.3 33 34 35 36 255.7 257.4 259.1 260.8 225.3 227.1 228.8 230.5 934.6 933.3 932.1 931.0 1159.9 1160.4 11G0.9 1161.5 754.7 733.5 713.4 694.5 12.312 11.971 11.649 11.344 .0827 .0851 .0875 .0899 22.3 23 3 24.3 25.3 37 38 39 40 262.4 264.0 265.6 267.1 232.1 233.8 235.3 236.9 929.8 928.6 927.5 926.4 1161.9 11G2.4 11G2.9 1163.4 676.6 659.7 643.6 628.2 11.055 10.756 10.521 10.259 .0922 .0946 .0970 .0994 20.3 27.3 41 42 2G8.6 270.0 238.4 239.9 925.4 924.3 1163.8 11G4.3 G13.4 599.3 10.037 9.811 .1017 .1041 900 PROPERTIES OF SATURATED STEAM — Gmtinued. Pounds per Heat Units in one Yol line. Square Inch. Pound above 32° F. © 0 * c o 0 2 33 Sta itent sat of ipori- tion. +gi| Rela- tive Specific §D« Cu. Ft. Cu. Ft. %Z J£ ££ s> ^X> S3 HhkJ in 1 Cu. in one geS o ° 3 6 3 ? < M'i 3*1.1 7i ^ ® 3 Rela- tive Specific c2§ g.2 « b0? 0 Cu. Ft. in lCu. Ft. of Water. Cu. Ft. n lLb. of Steam. > 70.3 71.3 72.3 73.3 85 86 87 88 316.0 316.8 317.6 318.4 287.0 287.8 288.7 '289.5 891.2 890.6 890.1 889.5 1178.3 1178.5 1178.8 1179.0 310.5 307.2 304.0 300.8 5.053 4.998 4.943 4.891 .2010 .2032 .2053 .2075 74.3 75.3 76.3 77.3 89 90 91 92 319.2 320.0 320.8 321.6 290.3 291.1 291.9 292.7 888.9 888.3 887.8 887.2 1179.3 1179.5 1179.8 1180.0 297.7 294.7 291.8 288.9 4.839 4.788 4.739 4.690 .2097 .2118 .2139 .2160 78.3 70.3 80.3 81.3 93 94 95 96 322.3 323.1 323.8 324.6 293.5 294.3 295.1 295.9 886.6 886.1 885.5 885.0 1180.2 1180.4 1180.7 1180.9 286.1 283.3 280.6 278.0 4.643 4.596 4.551 4.506 .2182 .2204 .2224 .2245 82.3 83.3 81.3 85.3 97 98 99 100 325.3 326.1 326.8 327.5 296.6 297.4 298.1 298.9 884.5 883.9 883.4 882.9 1181.1 1181.4 1181.6 1181.8 275.4 272.8 270.3 267.9 4.462 4.419 4.377 4.336 .2266 .2288 .2309 .2330 80.3 87.3 8S.3 89.3 101 102 103 104 328.2 329.0 329.7 330.4 299.6 300.4 301.1 301.8 882.3 881.8 881.3 880.8 1182.0 1182.2 1182.5 1182.7 265.5 263.2 260.9 258.7 4.296 4.256 4.217 4.179 .2351 .2371 .2392 .2413 90.3 91.3 92.3 9J.3 105 106 107 108 331.1 331.8 332.4 333.1 302.5 303.3 304.0 304.7 880.3 879.8 879.3 878.8 1182.9 1183.1 1183.3 1183.5 256.5 254.3 252.2 250.1 4.142 4.105 4.069 4.033 .2434 .2454 .2475 .2496 94.3 95.3 93.3 97.3 109 110 111 112 333.8 334.5 335.1 335.8 305.4 306.1 306.8 307.4 878.3 877.8 877.3 876.9 1183.7 1183.9 1184.1 1184.3 248.0 246.0 244.0 242.0 3.998 3.964 3.931 3.897 .2516 .2537 .2558 .2578 98.3 99.3 100.3 1C1.3 113 114 115 116 336.5 337.1 337.8 338.4 308.1 308.8 309.5 310.1 876.4 875.9 875.4 875.0 1184.5 11S4.7 1184.9 1185.1 240.1 238.2 236.3 234.5 3.865 3.833 3.802 3.771 .2599 .2619 .2640 .2661 102.3 103.3 101.3 105.3 117 118 119 120 339.1 339.7 340.3 340.9 310.8 311.4 312.1 312.7 874.5 874.0 873.6 873.1 1185.3 1185.5 1185.7 1185.9 232.7 231.0 229.3 227.6 3.740 3.711 3.681 3.652 .2681 .2707 .2722 .2742 10'3.3 107.3 1013 109.3 121 122 123 124 341.6 342.2 342.8 343.4 313.4 314.0 314.7 315.3 872.7 872.5 871.8 871.3 1186.1 1186.3 1186.5 1186.6 226.0 224.4 222.8 221.2 3 624 3.596 3.568 3.541 .2762 .2782 .2802 .2822 110.3 111.3 125 126 344.0 344.6 315.9 316.6 870.9 870.4 1186.8 1187.0 219.7 218.2 3.515 3.488 .2842 .28o2 902 PHOMJIITIES OF SATURATED STEAM- Continued. Pounds per Square Inch. ? P 11 i HhWm w Rela- tive Specific beg Cu. Ft. in 1 Cu. Ft. of Water. Cu. Ft. in 1 Lb. of Steam. 154.3 155.3 156.3 157.3 169 170 171 172 367.6 368.1 368.6 369.1 340.4 340.9 341.4 341.9 853.6 853.2 852.9 852.6 1194.0 1194.2 1194 3 1194.5 169.2 168.4 167.6 166.8 2.646 2.633 2.617 2.603 158.3 159.3 160.3 161.3 173 174 175 176 369.5 370.0 370.5 370.9 342.4 342.8 343.3 343.8 852.2 S51.9 851.5 851.2 1194.6 1194.8 1194.9 1195.0 166.0 165.2 164.4 163.6 2.589 2.575 2.561 2.547 162.3 163.3 164.3 165.3 177 178 180 371.4 371.9 372.3 372.8 344.3 344.8 345.3 345.7 850.8 850.5 850.2 849.8 1195.2 1195.3 1195.5 1195.6 162.8 162.0 161.2 160.4 2.533 2.521 2.507 2.494 166.3 167.3 168.3 169.3 181 182 183 1S4 373.2 373.7 374.1 374.6 346.2 346.7 347.1 347.6 849.5 849.2 848.8 848.5 1195.7 1195.9 1196.0 1196.2 159.7 159.0 158.3 157.6 2.480 2.468 2.455 2.443 170.3 171.3 172.3 173.3 185 186 187 188 375.0 375.5 375.9 376.4 348.1 348.6 349.0 349.5 848.2 847.8 847.5 847.2 1196.3 1196.4 1196.6 1196.7 156.9 156.2 155.5 154.8 2.430 2.418 2.406 2.394 174.3 175.3 176.3 177.3 189 190 191 192 376.8 377.2 377.7 378.1 349.9 350.4 350.8 351.3 846.9 846.5 846.2 845.9 1196.8 1197.0 1197.1 1197.2 154.1 153.4 152.7 152.0 2.382 2.370 2.358 2.347 178.3 179.3 180.3 181.3 193 194 195 196 378.5 379.0 379.4 379.9 351.7 352.2 352.6 353.1 845.6 845.3 845.0 844.6 1197.4 1197.5 1197.6 1197.8 151.3 150.7 150.1 149.5 2.335 2.324 2.312 2.302 182.3 183.3 184.3 185.3 197 198 199 200 380.3 380.7 381.1 381.5 353.5 354.0 354.4 354.8 844.3 844.0 843.7 843.4 1197.9 1198.0 1198.1 1198.3 148.9 148.3 147.7 147.1 2.290 2.279 2.269 2.258 186.3 187.3 188.3 • 189.3 201 202 203 204 381.9 382.4 382.8 383.2 355.3 355.7 356.1 356.6 843.1 842.8 842.5 842.2 1198.4 1198.5 1198.7 1198.8 146.5 145.9 145.3 144.7 2.248 2.238 2.227 2.216 190.3 191.3 192.3 193.3 205 206 207 208 383.6 384.0 384.4 384.8 357.0 357.4 357.9 358.3 841.8 841.5 841.2 841.0 1198.9 1199.0 1199.2 1199.3 144.1 143.5 1*2.9 142.3 2.204 2.196 2.186 2.176 194.3 195.3 209 210 385.2 385.6 358.7 359.1 840.7 840.4 1199 4 1199.5 141.8 141.3 2.166 2.1o7 904 CO»I>EI¥!iATi:01¥ Mf STIAM.PIPES. (w. w. c.) No very satisfactory figures are found for the absolute condensation losses in steam pipes, most of reported tests being compared with hair felt. 0.012 lbs. per 24 hours per sq. ft. of pipe per degree Fahr., difference in temperature of steam and external air, which may be used in calculations, is based on the following : Lbs. of Water. d> u ^ Sq. ft. Sur- face. 0> 3 is* 2. & • 5 ate &3 u o "el** ■ p Jd So Test by. in 24 hrs. per sq. ft. in 24 hrs. Covering. Bedle & Bauer. 4130 11315 2.74 262 .0104 Asbestos. Norris. 3892 9360 2.40 234 .0103 Asbestos. Brill. 308 .0105 Magnesia sect'l. Norton. 315 .0125 Magnesia. The last test by C. L. Norton (Trans. A. S. M. E., 1898) was made with the utmost care. Mr. Norton found that a pipe boxed in with charcoal 1 inch minimum thickness Avas 20 per cent better insulated than when magnesia was used, corroborating Mr. Reinhart's statements concerning his experi- ence using flue dust to insulate pipes. Aboard Ship. — The battleship "Shikishima" carries 25 Belleville boilers capable under full steam of developing 15,000 I.H.P. in the main engines besides working the auxiliaries, each boiler supplying steam for 150 I.H.P. When at anchor, one boiler under easy steam, i.e., evaporating from 9 lb. to 10 lbs. of water from and at 212° F., per pound of coal — was just able to work one 48 K.W. steam dynamo at about half power, together with one feed pump, and the air and ciroulating pumps connected with the auxiliary condenser, into which the dynamo engine exhausted, besides working a fire and bilge pump occasionally. The dynamo was about 160 ft. of pipe length away from the boiler, the total range of steam pipe length connected being 500-600 ft. Performing the first-mentioned service with only one boiler under steam, the coal burned varied from 3h to 5 tons per day of 18 hours, for about 65 I.H.P., or about 7 lbs. per indicated horse-power at the best to 10 lbs. at the worst, an average of 8 lbs. and over, which shows that more than half the fuel must have been expended in keeping the pipes warm. All pipes were well covered and below decks, and machinery in first-class condition. (London-Engr.) Heating* JPipes. — To determine the boiler H.P. necessary for heating, it maybe assumed that each sq. ft. of radiating surface will condense about 0.3 lbs. of steam per hour as a maximum when in active service ; thus 20,000 sq. ft. times 0.3=6000 lbs. of condensation, which divided by 30 gives 200 boiler horse-power. Condensed steam in which there is no oil may be returned to the boiler with the feed-water to be re-evaporated. OUTFLOW OF STEAM. 905 OCTFIOW OF STEAM FROM A 6ITM OITI.1I PR£^lTRE IUTO VARIOUS LOWER PRESSURES. (D. K. Clark.) Absolute Outside Velocity of Actual Ve- Weight Dis- Pressure in Pressure Ratio of Outflow at locity of charged per Sq. Boiler per per Sq. Expansion. Constant Outflow In. of Orifice Sq. Inch. Inch. Density. Expanded. per Minute. Lbs. Lbs. Ratio. Ft. per Sec. Ft. per Sec. Lbs. 75 74 1.012 227.5 230 16.68 75 72 1.037 386.7 401 28.35 75 70 1.063 400 521 35.93 75 65 1.136 660 749 48.38 75 61.62 1.198 736 876 53.97 75 60 1.219 765 933 56.12 75 50 1.434 873 1252 64. 75 45 1.575 890 1401 65.24 75 43.46, 58 % 1.624 890.6 1446.5 65.3 75 15 1.624 890.6 1446.5 65.3 75 0 1.624 890.6 1446.5 65.3 When, however, steam of varying initial pressure is discharged into the atmosphere — pressures of which the atmospheric pressure is not more than 58 per cent — the velocity of outflow at constant density, that is, sup- posing the initial density to be maintained, is given by the formula — V— 3.5953 yfh, where V= the velocity of outflow in feet per minute, as for steam of the j initial density, h = tlie height in feet of a column of steam of the given absolute initial pressure of uniform density, the weight of which is equal to | the pressure on the unit of base. The following table is calculated from this formula : OUTEUOW OF STEAM INTO THE ATMOSPHERE. (D. K. Clark.) Absolute Initial Outside Ratio of Velocity of Actual Ve- Weight Dis- Pressure in Pressure Expansion Outflow at locity of charged per Boiler in in Lbs. per in Constant Outflow, Sq. Inch of Lbs. per Sq. Inch. Sq. Inch. Nozzle. Density. Expanded. Orifice per Min. Lbs. Lbs. Ratio. Ft. per Sec. Ft. per Sec. Lbs. 25.37 14.7 1.624 863 1401 22.81 30 14.7 1.624 867 1408 26.84 40 14.7 1.624 874 1419 35.18 45 14.7 1.624 877 1424 39.78 50 14.7 1.624 880 1429 44.06 60 14.7 1.624 8S5 1437 52.59 70 14.7 1.624 889 1444 61.07 75 14.7 1.624 891 1447 65.30 90 14.7 1.624 895 1454 77.94 100 14.7 1.624 898 1459 86.34 115 14.7 1.624 . 902 1466 98.76 135 14.7 1.624 906 1472 115.61 155 14.7 1.624 910 1478 132.21 165 14.7 1.624 912 1481 140.46 215 14.7 1.624 919 1493 181.58 906 STEAM PIPES. Rankine says the velocity of steam flow in pipes should not exceed 6000 feet per minute (100 feet per second). As increased size of pipe means in- creased loss by radiation, care should be taken that in order to decrease the velocity of flow, the losses by radiation do not become considerable. The quantity discharged per minute may be approximately found by JRankine's formula (" Steam Engine," p. 298;, W = 00 ap -j- 70 — Gajj-f 7, in which W = Aveight in pounds, a = area of orifice in square inches, and p ■= absolute pressure. The results must be multiplied by lc = 0.93 for a short, pipe, and by lc = 0.63 for their openings as in a safety valve. Where steam flows into a pressure greater than two-thirds the pressure in ; the boiler, W = 1.9 ak^(p — d) d, in which d = difference in pressure in pounds per square inch between the two sides, and a,p, and A; as above. Multiply the results by 2 to reduce to h.p. To determine the necessary dif- ference in pressure where a given h.p. is required to flow through a given opening, (L - 2 V 4 14 aVc Flow of Steam Tliroug-li Pipes. (G. H. Babcock in " Steam.") The approximate weight of any fluid which will flow in a minute through any given pipe with a given head or pressure may be found by the formula r=87t /: D (Pi — Pi) d5 3.6\ *+¥) in which W= weight in pounds, d = diameter in inches, D =. density or weight per cubic foot, ^ = initial pressure, p2 — pressure at the end of the pipe, and L = length in feet. The following table gives, approximately, the weight of steam per minute which will flow from various initial pressures, with one pound loss of pres- sure through straight smooth pipes, each having a length of 240 times its own diameter. For sizes below 6 inches, the flow is calculated from the actual areas of " standard " pipe of such nominal diameters. For h.p. multiply the figures in the table by two. For any other loss of pressure, multiply by the square root of the given loss. For any other length of pipe, divide 240 by the given length expressed in diameters, and multiply the figures in the table by the square root of this quotient, which will give the flow for 1 pound loss of pressure. Conversely dividing the given length by 240 will give the loss of pressure for the flow given in the table. Table of Flow of Steam Tliroug-li Pipes. Initial Pres- sure by Gauge. Lbs. per Sq. Inch. 10 20 30 40 50 60 70 Diameter of Pipe in Inches. Length of each — 240 Diameters. Weight of Steam per Min. in Lbs., with 1 Lb. Loss of Pressure. 1.16 2.07 5.7 10.27 15.45 25.38 46.85 1.44 2.57 7.1 12.72 19.15 31.45 58.05 1.70 3.02 8.3 14.94 22.49 36.94 68.20 1.91 3.40 9.4 16.84 25.35 41.63 76.84 2.10 3.74 10.3 18.51 27.87 45.77 84.49 2.27 4.04 11.2 20.01 30.13 49.48 91.34 2.43 4.32 11.9 21.38 32.19 52.87 97.60 2.57 4.58 12.6 22.65 34.10 56.00 103.37 2.71 4.82 13.3 23.82 35. S7 58.91 108.74 2.83 5.04 13.9 24.92 37.52 61.62 113.74 2.95 5.25 14.5 25.96 39.07 64.18 118.47 3.16 5.63 15.5 27.85 41.93 68.87 127.12 3.45 6.14 17.0 30.37 45.72 75.09 138.61 STEAM PIPES. 907 Table of Plow of Steam Xliroug-ti I*ipe«.— Continued. Initial Pres- sure by Gauge. Lbs.' per Sq Inch. Diameter of Pipe in Inches. Length of Each = 240 Diameters. I Weight of Steam per Min. in Lbs., with 1 Lb. Loss of Pressure. 10 20 40 50 100 120 150 95.8 112.6 126.9 139.5 150.8 161.1 170.7 179.5 187.8 195.6 209.9 115.9 211.4 341.1 143.6 262.0 422.7 168.7 307.8 496.5 190.1 346.8 559.5 209.0 381.3 615.3 226.0 412.2 665.0 241.5 440.5 710.6 255.8 466.5 752.7 269.0 490.7 791.7 281.4 513.3 828.1 293.1 534.6 862.6 314.5 573.7 925.6 343.0 625.5 1009.2 502.4 622.5 731.3 824.1 906.0 979.5 1046.7 1108.5 1166.1 1219.8 1270.1 1363.3 1486.5 804 1177 996 1458 1170 1713 131 S 1930 1450 2122 1567 2294 1675 2451 1774 2596 1866 2731 1951 2856 2032 2975 21S1 3193 2378 3481 The loss of head due to getting up the velocity, to the friction of the steam entering the pipe and passing elbows and valves, will reduce the "~-y given in the table. The resistance at the opening and that at a be valve are each about the same as that for a length of pipe equal to 114 diameters divided by a number represented by 1 -f- -j- ■ For the sizes of pipes given in the table these corresponding lengths are : 20 1 25 1 34 | 41 I 47 j 52 | 60 | 71 79 84 92 95 The resistance at an elbow is equal to § that of a globe valve. These equivalents — for opening, fur elbows, and for valves —must be added in each instance to the actual length of pipe. Thus a 4-inch pipe, 120 diame- ters M0 feet) long, with a globe valve and three elbows, would be equivalent to 120 + 60 + 60 4- (3 X 40) = 360 diameters long ; and 360 -f 240 = 1±-. It would therefore have 1£ ibs. loss of pressure at the flow given in the table, or deliver (1 ~ Vf| — 8.16), 81.6 per cent of the steam with the same (1 lb.) loss of pressure. ^Equation of Pipes (Steam). It is frequently desirable to know what number of one size of pipes will equal in capacity another given pipe for delivery of steam or water. At the same velocity of flow two pipes deliver as the squares of their internal diameters, but the same head will not produce the same velocity in pipes of different sizes or lengths, the difference being usually stated to vary as the square root of the fifth power of the diameter. The friction of a fluid within itself is very slight, and therefore the main resistance to flow is the friction upon the sides of the conduit. This extends to a limited distance, and is, of course, greater in proportion to the contents of a small pipe than ol a large. It may be approximated in a given pipe by a constant multi- plied by the diameter, or the ratio of flow found bv dividm*? some power of the diameter by the diameter increased by a constant. Careful compari- sons of a large number of experiments, bv different investigators, has de- Vf ?PJ? the followin.g as a close approximation to the relative flow in pipes of different sizes under similar conditions : W oo V d + 3.6 W being the weight of fluid delivered in a given time, and d being the internal diameter in inches. 908 The diameters of " standard " steam and gas pipe, however, vary from the nominal diameters, and in applying this rule it is necessary to take the true measurements, which are given in the following table : Tal>le of Standard Sizes Steam and GaN Pipes, s C £ CO 620 273 133 39.2 19.6 11.7 6.39 3.02 1.65 1.51 2.18 2.98 3.93 5.05 6.34 7.75 9.48 11.5 13.4 15.7 1S.2 ■ 23.9 38.2 67.6 108 159 CO cc 1Q 377 166 81.1 23.8 11.9 7.12 3.89 1.84 1.63 2.49 3.54 4.85 6.40 8.22 10.3 12.6 15.4 18.7 21.8 25.6 29.6 38.9 62.1 110 175 259 363 io c <- - 205 90.4 44.1 13.0 6.47 3.87 2.12 1.83 2.97 4.54 6.48 8.85 11.7 15.0 18.8 23.0 28.2 34.1 39.9 46.6 54.1 70.9 113 201 319 473 663 *# CO 96.9 42.5 20.9 6.13 3.06 1.83 2.21 4.03 6.56 10.0 14.3 19.5 25.8 33.1 41.6 50.7 62.2 75.3 88.0 103 119 157 250 443 705 1044 1465 - a 52.9 23.3 11.4 3.34 1.67 1.66 3.67 6.70 10.9 16.6 23.8 32.5 42.9 55.1 69.2 84.5 103 125 146 171 198 260 416 736 1172 1734 2434 5? « 31.7 14.0 6.82 1.26 1.87 3.11 6.87 12.5 20.4 31.2 44.5 60.8 80.4 103 129 158 193 234 274 320 371 487 778 1378 2193 3245 4554 CO * 15.8 6.97 3.45 2.26 4.23 7.03 15.5 28.3 46.0 70.5 101 137 181 233 293 358 438 530 619 724 840 1102 1761 3117 4961 7341 10301 * - 4.88 2.05 3.20 7.25 13.6 22.6 49.8 90.9 148 226 322 440 582 747 938 1146 1403 1098 1984 2322 2691 3532 5644 9990 15902 23531 33020 ~ - 2.27 2.90 9.30 21.0 39.4 65.4 144 263 429 656 936 1281 1688 2168 2723 3326 4070 4927 5758 6738 7810 10249 16376 2X91)0 46143 6S282 95818 * * 2.60 7.55 24.2 54.8 102 170 376 686 1116 1707 2435 3335 4393 5612 7087 8657 10600 12824 14978 17537 '20327 26676 42621 75453 120100 177724 249351 -ma rtiNsi-*^ "^Kt m'co ^iocot>.c^0"HC<'M'r*lli:5':ot:~coo~l''0':oc':iC0 •VIQ 910 ft » « 1 o ft 1 H 2 ft S si : ft 2 >, £ 2 o o H e8 3 o a S 3 o s rH ■* CiOCOlO CO rH L- -fl 10 CO •ssoq; jo opu'a; 1.000 .280 .172 .091 .056 .042 •jnou isd imfi joo^ .i8d sjuifi ui ssoq; 1077.4 301.7 185.3 98.0 60.3 45.2 .5 •jsoq; 'd "H a9d q^SuQT ui %dd£ rH 63 -t< f- C5 ■ssoq; jo opi:^ 1.000 .301 .170 .103 .063 .047 mtioh J9c[ tttl'St joojj jad s^iufi ui ssot; 729.8 219.6 128.3 46.0 34.3 _cB CO •;soq; •i ^00,3; .I8d SiJIUfl tii ssori 624.1 187.2 111.0 66.2 41.2 33.7 3 •iso-j "j 1.00 .46 .30 .20 .13 .09 muoh; .T8d mi-j-x 700j[ .I8d S'HUQ ui ssorx 219.0 100.7 65.7 43.8 28.4 19.8 •sanou J UI §lII.I8AOQ JO SS8U5[0iqX ©~ """i-H uostances. B.T.U. per Specimens. sq. ft. per min. at 200 lbs . 3.18 Box A, 1 with sand . . 2 with cork, powdered . . 1.75 3 with cork and infusorial 1.90 earth . . 4 with sawdust 2.15 5 with charcoal 2.00 6 with ashes 2.46 Brick wall 4 inches thick . . 5.18 Pine wood 1 inch thick Hair felt 1 inch thick . Cabot's seaweed quilt . Spruce 1 inch thick . . Spruce 2 inches thick . Spruce 3 inches thick . Oak 1 inch thick . . . Hard pine 1 inch thick . B.T.U. per sq. ft. per min. at 200 lbs. . 3.56 . 2.51 . 2.78 . 3.40 2.31 . 2.02 . 3.65 . 3.72 Prof. B,. C. Carpenter says that there is great difference in the flow of heat through a metal plate between different media. In discussing Professor Norton's paper he gave the values as shown in the following table as the result of experiments conducted in his laboratory. Heat transmitted, in Thermal "Units Through Clean Cast- iron T»late T7g Inch Thick. (Carpenter.) Difference of Temperature. Degrees F. Steam to Water. Lard Oil to Water. Air to Water. Per Square Foot. Per Square Foot. Per Squ are Foot. Per Deg Total per Per Deg. Total per Per Deg. Total per per hour mimne per hour minute per hour minute B. T. U B. T. U. B. T. U. B. T. IT. B. T. U. B. T. IT. 25 21 8.8 6.5 2.7 1.2 0.5 50 48 40 13 10.8 2.5 2.7 75 84 110 19.5 24.5 3.7 5.8 100 127 211 26 43.3 5.0 8.3 125 185 375 31.5 65.5 6.2 13 150 255 637 39 72.5 7.5 18.7 175 45.5 132 8.7 25.4 200 52 173 10 33 300 78 390 15 75 400 20 133 500 25 208 The above investigation indicates that the substance which surrenders the heat is of material importance, as is also the temperature of the surrounding media. 912 jo qouj jod syedjqx jo aequmsj t^COCO-^-tlT^^r-i^OOOOCOOOOOOOOOOOOOOOCOCOOOOOOOOO t s PS s H II 4 .241 .42 .559 .837 1.115 1.668 '2.244 2.678 3.609 5.739 7.536 9.001 10.665 12.34 14.502 18.762 23.271 28.177 33.701 40.065 45.95 48.985 53.921 57.893 61.77 69.66 77.57 85.47 93.37 ^"3 so . crjS o 2513. 1383.3 751.2 472.4 270. 166.9 96.25 70.66 42.91 30.1 19.5 14.57 11.31 9.02 7.2 4.98 3.72 2.88 2.29 1.82 1.456 1.27 1.04 .903 .788 .616 .495 .406 .339 o £,+= "So 53^ § ft M xi 3 14.15 10.49 7.73 6.13 4.635 3.645 2.768 2.371 1.848 1.547 1.245 1.077 .949 .848 .757 .63 .544 .478 .427 .382 .339 .319 .288 .268 .250 .221 .198 .179 .164 1 ^ 3,— i rt ^ x CD 9.44 7.075 5.657 4.547 3.637 2.904 2.301 2.01 1.608 1.328 1.091 .955 .849 .764 .687 !501 .443 .397 .355 .318 .299 .273 .255 .239 .212 .191 .174 .159 B3 *■ s s < CD l> 'S 1 .0717 .1249 .1663 .2492 .3327 .4954 .668 .797 1.074 1.708 2.243 2.679 3.174 3.674 4.316 5.584 6.926 8.386 10.03 11.924 13.696 14.579 16.051 17.23 18.407 20.764 23.12 25.477 27.832 CD X' .0573 .1041 .1917 .3048 .5333 .8626 1.496 2.038 3.356 4.784 7.388 9.887 12.73 15.961 19.99 28.888 38.738 50.04 62.73 78.839 99.402 113.098 137.887 159.485 182.655 233.706 291.04 354.657 424.558 CD = X .129 .229 .358 .554 .866 1.358 2.164 2.835 4.43 6.492 9.621 12.566 15.904 19.635 24.306 34.472 45.664 58.426 72.76 90.763 113.098 127.677 153.938 176.715 201.062 254.47 314.16 380.134 452.39 03 3 o "3 el CD CD .848 1.144 1.552 1.957 2*589 3.292 4.335 5.061 6.494 7.753 9.636 11.146 12.648 14.162 15.849 19.054 22.003 25.076 28.076 31.477 35.343 37.7 41.626 44.768 47.909 54.192 60.476 66.759 73.042 to e B N H N "eg PI CD | 1.272 1.696 2.121 2.639 3.299 4.131 5.215 5.969 7.461 9.032 10.996 12.566 14.137 15.708 17.477 20.813 23.955 27.096 30.238 33.772 37.699 40.055 43.982 47.124 50.265 56.549 62.832 69.115 75.398 •ssan5[oiqx o PI .068 .088 .091 .109 .113 .134 .14 .145 .154 .204 .217 .226 .237 .246 .259 .28 .301 .322 .344 .366 .375 .375 .375 .375 .375 .375 .375 .375 .375 0 CD s 5 !&3 i .27 .364 .494 .623 .824 1.048 1.38 1.611 2.067 2.468 3.067 3.548 4.026 4.508 5.045 6.065 7.023 7.9S2 8.937 10.019 11.25 12. 13.25 14.25 15.25 17.25 19.25 21.25 23.25 4jH " C .405 .54 .675 .84 1.05 1.315 1.66 1.9 2.375 2.875 3.5 4. 4.5 5. 5.563 6.625 7.625 8.625 9.625 10.75 12. 12.75 14. 15. 16. 18. 20. 22. 24. III J rtlXriWKII-fcw** WW-4N ~bl ~fa Pta . STEAM PIPES. 913 .i8d ^qSta^ I^uiuioj^; a o Ph .29 54 74 1.09 1.39 2.17 3. 3.63 5.02 7.67 10.25 12.47 14.97 20.54 28.58 o 33^ "So e8 G 3 F c« CP OP 18.632 12.986 9.07 7.046 5.109 4.016 3.003 2.556 1.975 1.649 1.328 1.137 1. .793 .664 £ 9.433 7.075 5.657 4.547 3.637 2.904 2.301 2.01 1.608 1.328 1.091 .955 .849 .687 .577 s a H "3 CD m .086 .161 .219 .323 .414 .648 .893 1.082 1.495 2.283 3.052 3.71 4.455 6.12 8.505 5 ^ a a1 02 .033 .068 .139 .231 .452 .71 1.271 1.753 2.935 4.209 6.569 8.856 11.449 18.193 25.967 3 ^ GO .129 .229 .358 .554 .866 1.358 2.164 2.835 4.43 6.492 9.621 12.566 15.904 24.306 34.472 a5 a ? a 5 a a a? a a .644 .924 1.323 1.703 2.312 2.988 3.996 4.694 6.073 7.273 9.085 10.549 11.995 15.120 18.064 CD O 1.272 1.696 2.121 2.639 3.299 4.131 5.215 5.969 7.461 9.032 10.996 12.566 14.137 17.477 20.813 •aStVBf) 8JT^ & !JS9J'B8_JJ 6 h* .*, * * 08 ClHOG>»t»fflOiO«rtOOOO •sson^oiqx s cot^cst-ci^fcoT-i rn y-t i-H in t> Cl CI -V l- x — o CI CC O CI -f t- CO HrlHrtHHHflCMtlMMnW-* co B s "^ G <5 1 .205 .294 .421 .542 .736 .951 1.272 1.494 1.933 2.315 2.892 3.358 3.818 4.813 5.75 s ? +s cu o .405 .54 .075 .84 1.05 1.315 1.66 1.9 2.375 2.875 3.5 4. 4.5 5.563 6.625 '3'^ si a rH ?h 7-t w -+< o - 1 i - O' o i © © i- © -r< ©' © -i oi oi ci co ro -r -f -r i o lo o; © ~ L~ 'j;.«::-tC2;:: : • ~ '" — — — © -f rH -+ -* Ci t- x ci x lo re ro -H i::L-OH-i-c-tci:i:-i: x i - x lo © x ro — io oi © l- i- -ti o i tA © ~i x «n«l>MOL-tttC0 01 t- © © X I- ttCTt CO CO CO 04 CM 01 01 01 0 1 ~l — rn +-> © 10 t- CO © X 'X O CO LO rH X LO © -f t- L0 CO -+i 01 t- CO CO Ol •* CO -P CI O O rH iHC.-fcoco:ic[-i-o-i-.-r-::-fi-:ixTHC,t- io co oi h — ■ © x CD X O O i-i © © LO CO 0! i-H © O © X I- © L0 -r -r CO CO CO 01 01 01 01 01 Ol CM rH rH IMCOCllMHHHHHHr poLOcowcooncfjMccHccnHoo-t'O'fOHOiaeint'Wot-b O © 1.0 — -+ i-i 10 L- CO i~ © L- X 01 C; ~ /. © ~r :0 i-H © c cd -r -f x -h i-h x cq 10 co Cjwqcqoqffl lo lo -h co cr> co co ei ei 01 :i:i:i- &-! t* CO Ol 01 01 rH rH rH i-i i-H rH i-i rH ' 10 1-0 CO L- X CO rH 01 rH CO CO t- 01 -H -H re CO O X -HH rH -HH CO , iTi ft* ro rr, _H r. ■* - - | .. -C O I- C O' O [- O H 1". O r ^ .,.:-*-* io co i - x i-h co r.cc-H-t<-f-j-/jcrooixi- co co x co" ^O O O O 3 O © O CO © O © © rH rH rH 01 CO ~H L0 CO L- © © 01 rH LO L- © rH -HH , . J ~ - C I - X ~ r- . ■ Co ? ■ £ tr S rr : s rH ' ^ th CN CO CO •# IO O UOH« a ih n w o » » o n co uj n ci co o n o -f n -f '»' rH rH rH rH 01 CO L0 CO L~ Ci rH CO L0 I- CO 01 >l X — - rH rH r-, rn 01 01 0 1 01 CO C- t~ t- CO lC- © ■* OlirSCOCOrHVOlrHt- CO « LO « "t W 00 C-l 00 Mm M C< ■ ~f CO — CO I - CO X 10 01 Ci X t- CO X Hl-H X -+• Ci L- CO 1.0 -f CO CO CO -« CO -H 0 . © © © rn rn oi oi co -t -t- io -o i- r- oi c- -+ •- co c. © oi lo cc •+ co i- lo -r -I lo 1 © © ~ © © © © © © © © © © © rH H 01 CO -+ -H CO L- X © -H CO -r CO X © 0 l-< LO © CO t-< CO -t< -H CO CO IO r- Lo JJ '© t Ci L0 L0 rH ,- O O I- O X t-N-* CO«H rH L- CO © rH L0 rH — CO X 01 L0 X © 01 C. © © © © L- l~ -f X X rH CO Ol 01 © L- rn . lo Ci cq © lo co r q ;«:•. -q z. — -+ 1.0 xi-ci © -i-_ t - — — i - © 01 01 © co co COT1 " ' i-i i-i oi co' -* LO) CO t-" CO ©J © -f' I- LO* -t' lo' X r^ t^ CO CO co" -+ [ - 0 i X ) o -+ -t- X rH rH rH 01 CO -H L0 L^ X © 01 -H © X rH CO © © 01 -R Ol IC- CM X CO © -H © 1 0 CO 1 0 r- CO I - X © rH 01 -H CO O © © 01 -H L0 l>- 00 © 03 CO Oi -c" oi rn ci x co , o cc o i — ro x © co © -+< © co t- — ,- c. ^ x 01 © © -r © co i- flnqtH-jHC, © x © -t o i oq i - lq - i - co © — o i -+ i o © x © — J i * lo cq cq © r^ CO CO -rJH LO CO t^ t-^ 00 Ci ©" ©' i-H Ol' -"" lo" 00 -H L0 CO rj -jH t^ © CO t^' © OCOCOrH-J_ 1060 1 11 1 2f 9| 8 ! 3 2360 7 .60 f 1120 *, 12i It 2f 10! 8 1 31 3200 S .639 1 1280 1 13* 2| ii! 8 1 3* 4190 9 .6^8 ii 1310 ft 15 li 2| 13 12 i 3* 3610 10 •713 3 1330 16 ll36 3 14i 12 £ 3f 2970 12 .79 if 1470 135 19 li 3i 16i- 12 1 3! 4280 14 .864 § 1600 T36 21 If 3* 18! 12 l H 4280 15 .904 F 1600 T36 22J If 3f 3| 20 16 l 3660 16 .946 1600 ft 23A- llB 21J 16 l 1 4210 18 1.02 1690 25 ll96 3* 22! 16 H 4540 20 1.09 1780 ¥ 27i 1*4 3f 25 20 H 5 4490 2 '2 1.18 1850 29* lit 3f 27i 20 ii 5J 4320 24 1.25 1920 l 31* 32" li 1| 3f 4 291 29* 20 ii 5* 5130 26 1.30 1980 33| 341 36 36* 1| 2 3£ 4* 311 31! 24 H 5! 5030 28 1.38 !' 2040 J Mfi-,,1n 4 4133* 34 28 H 6 5000 30 1.48 2000 38 38f 1* 21- 4 4|35* 36 28 if 61 4590 36 1.71 1* 1920 I 4U 4f>3r 51 52| 15 2^ 4J 4^42" 42f j.i r,« 4si 49A 32 if 6* 5790 42 1.87 2100 1 15 ^0 36 i* 71 5700 48 2.17 2J 2130 57* 59^ 282| 4| 5!|54| 56" 44 i* n 6090 916 STEAM. NOTES. — Sizes up to 24 inches are designed for 200 lbs. or less. Sizes from 24 to 48 inches are divided into two scales, one for 200 lbs., the other for less. The sizes of bolts given are for high pressure. For medium pressures the diameters are & inch less for pipes 2 to 20 inches diameter inclusive, and 4 inch less for larger sizes, except 48-inch pipe, for which the size of bolt is If inches. When tAvo lines of figures occur under one heading, the single columns up to 24 inches are for botb medium and high pressures. Beginning with 24 inches, the left-hand columns are for medium and the right-hand lines are for high pressures. The sudden increase in diameters at 16 inches is due to the possible inser- tion of wrought-iron pipe, making Avith a nearly constant Avidth of gasket a greater diameter desirable. When wrought-iron pipe is used, if thinner flanges than those given are sufficient, it is proposed that bosses be used to bring the bolts up to the standard lengths. This avoids the use of a reinforcement around the pipe. Figures in the third, fourth, fifth, and last columns refer only to pipe for 200 lbs. pressure. In drilling valve flanges a vertical line parallel to the spindles should be midway between tAvo poles on the upper side of the flanges. steam: EaroiwES. Steam engines are often classed according to the number of cylinders the steam passes in succession, and which are different in size, Simple expansion, Compound, Triple, Quadruple. Any one of the above classes, if run non-condensing, is called low-pres- sure, or non-condensing ; and if run with condenser is called high-pressure, or condensing. Nowadays the above classes are made in two types : high speed, including all engines running above, say, 150 revolutions per minute ; and low speed, all those running at less than 150 revolutions. This division is scarcely correct, as some of the long-stroke engines run- ning at 125 revolutions have more than 1000 feet piston speed, Avliile few of the so-called high speed machines exceed 600 feet per minute piston speed. in selecting an engine for electrical work it is necessary to see that the machine is extra heavy in all its parts ; especially so for electric railway work, as the changes in load are often great and sudden, and in case of short circuit, engines are liable to be called on for tremendous increase in output, and should have no weak parts. This especially applies to fly- wheels, of which a large number have burst on the large, slow-running engines used in railway power-houses. Bearings should all be of extra large size, especially so on the main shaft journals of large direct-connected units. The selection of size (horse-power) depends largely upon the rating of the connected electrical machinery and the number of hours it runs, much being left to the judgment of the advising engineer. For direct-connected units it is not necessary to install an engine of greater rated capacity than the rated output of the generator, as the engine Avill easily care for overload on the generator if rated at \ cut-off, as is usual. Some builders of engines rate their sizes for connections to dynamos so as to supply \\ h. p per k.AV. capacity of the dynamo. The selection of condensing or high-pressure engines has in the past de- pended largely on availability of an adequate supply of water for condens- ing purposes ; but to-day the cooling toAver Avith water enough to fill a supply-tank once, and a regular supply for boiler-feed, is a very satis- factory arrangement. STEAM ENGINES. 917 Summary of Tests of Steam Engines of Various Types. By Prof. R. C. Carpenter. %0*u f*o>- I SS9 Style of Engine. Boiler Eva p. p lb. Comb B.&A.2 Kind of Coal. Simple non- G 200 34.S 4.47 110 55 11.50 Pea A. condensing 1 405 34.5 6.54 257 63.4 9.11 Culm slide valve. 7 107r, 35.7 4.60 862 51. 9.46 Soft Pa. 11 300 37.3 4.49 90 44. 12.20 " " 11 300 34.3 4.72 95 46.7 10.20 " 111. 24 1000 31.8 5.38 717 71.7 9.15 " 31 270 41.5 5.50 126 47.5 10.60 Hard, Buck 33 270 31.6 4.61 147 54.5 10.70 Pea Average. 35.1 5.07 54.2 10.24 Simple non- 17 30( 30.1 3.09 139 46 11.45 Clearfield condensing 19 150 26.9 3.5 90 60 9.73 Hard, Buck Corliss. 22 350 28. 3.77 153 44.7 S.55 Soft, Ohio Average. 28.3 3.45 50.3 Compound 2 1000 30.5 4.22 603.5 60.3 9.03 1 Soft, 3 Hard non-con- 4 1250 36.8 4.33 674 53.8 9.92 Culm and slack densing.' 21 400 34.20 4.17 203 51. 10.23 Soft, Pa. 24 1200 30.37 4.93 754 62.7 9.01 " 111. Average of. 32.28 4.55 Compound 3a 600 29.4 4.43 174 29 10.38 1 Soft, 3 hard condensing high-speed 3 (',00 23.2 3.50 190 32 9.93 " " 8 400 20.2 3.14 154 3S 8.29 Soft, Ohio automatic. 86 400 16.7 2.40 180 45 7.75 " " 13 250 24.6 2.95 86 34.5 10.51 " Pa. 16 350 22.7 3.41 164 47 9.50 Hard pea 18 1200 25.6 3.61 904 75 10.58 " " 21 400 29.3 3.81 188 47 10.23 Soft Average. 23.96 3.41 9.64 Compound 10 825 22.7 4.06 482 58.2 8.29 Culm & Slack condensing 14 looo 21.9 2.56 277 27.7 10.96 " " Corliss, 14 1000 20. 314 31.4 10.96 " " Greene, 28 350 16.64 2.10 182 52.2 11.80 Soft Mcintosh & 27 500 16.90 2.61 290 58. 9.36 " Seymour, 30 2000 14.5 1.80 814 40.7 10.7 " etc., etc. 34 200 17.3 2.91 145 72. 11.14 « 35 1600 20.5 2.18 11.14 Average. 18.8 2.60 10.54 918 STEAM. Hoi'ie-power of Steam Engines. Xominal Horse-power. — Now very little used. D = dia. cyl. in inches. A — area of piston in sq. inches. L = length of stroke in feet. Boulton & Watt, nominal H.P. = — - • Kent gives as handy rule for estimating the h.p. of a single cylinder engine, — . This rule is correct when the product of the in.e.p. and piston speed = 21,000. The above rule also applies to compound triple and quadruple engines, and is referred to the diameter of the low-pressure cylinder, and the h.p. of such an engine then becomes (dia. low-pres. cyl.)2 TT _ , , . . i 1 ^^ = H.P. (roughly.) Indicated Horse Power : 1. 12.1*. — The power developed in the cylinder of a steam engine is correctly determined only by use of the indicator, and comparisons and steam consumption are always calculated on that basis. M.E.P. = mean pressure in pounds per square inch, as shown by the indicator card. L= stroke of piston in feet. n = number of revolutions per min. a — effective area of head side of piston. a, = effective area of crank side of piston. _ [(a x in.e.p.) + (a, X m.e.p.)] X Ln ' — 33,000 For multiple cylinder engines, compute I. H.P. for each cylinder, and add results together for total power. Brake Morse-power.— The brake horse-power (B. H.P.) of an engine is the actual or available horse-power at the engine pulley ; at any given speed and given brake-load, the B.H.P is less than the corresponding I. H.P. by the horse-power required to drive the engine itself at the given speed, and with the pressures at the bearings, guides, etc., corresponding to the given brake-load. If W= load. in lbs. on brake lever or rope, /= distance in feet of center of brake-wheel from line of action of brake-load, ]¥= revolutions per minute ; tlienB-HP = flr The mechanical efficiency of any given engine is less the greater the expansion ratio employed, and of two engines of the same type, developing the same power at the same speed, that which uses the higher degree of expansion will have the lower mechanical efficiency. The effect of this, though not usually important, is to make the best ratio of expansion in any given case somewhat less than that Avhich makes the steam consumption per I. H.P. -hour a minimum. The mechanical efficiencies on full load of modern engines range from 80 to 95 per cent. Large engines have, of course, higher mechanical efficien- cies than small ones (a very small engine may have as low a mechanical efficiency as 40 to 50 per cent, but this is generally due to bad design and insufficient care being taken of the engine), simple than compound engines, and compound than triple engines — at any rate when not very large. Prof. Thurston estimates that the total mechanical loss in non-condensing engines having balanced valves may be apportioned as follows : — main bearings 40 to 47 per cent, pistons and rods 33 per cent, crank-pins 5* per cent slide-valves and rolls 2£ per cent, and eccentric straps 5 per cent. An unbal- anced slide-valve may absorb 2G per cent, and in a condensing engine the air-pump 12 % of the total mechanical loss. STEAM ENGINES. Cylinder Ratios iu Compound Eng-ines. 919 The object of building multiple cylinder engines is, a, to use high steam pressure, b, to get the greatest number of expansions from the steam, c, to reduce the cylinder condensation. Prof. Thurston says : " Maximum expansion, as nearly adiabatic as prac- ticable, is the secret of maximum efficiency." Although the theory of determining the sizes of cylinders is perfectly understood, yet there are so many causes for varying the results that prac- tically to-day but little attention is given to calculations, the plan being to use dimensions such as have proved best practice in the past. The proportions of cylinders are supposed to be such as to equally divide the number of expansions and work among tliem, and these dimensions have to be varied somewhat to meet the experience of the engineer. Given the initial pressure (absolute) i.P. and the terminal pressure (abso- i P. lute) t.P., then the total number of expansions is E =—:— ^ , and the num- ber of expansions for each cylinder is as follows : For compound ^E, For triple expansion 3"^E, For quadruple expansion "VE. Better results are often obtained by cutting off a trifle earlier in the high- pressure cylinder ; and this fact, in connection with the extent of reheaters and receivers, changes the actual ratios from the ideal to the practical ones shown in the following table : IVumBicr of ExpanMons for Condensing* £ng-in«s. i.P. Abso- lute. Total Expan- sions. Expansions in Each Cylinder. Type. 1st. 2d. 3d. 4th. Single cylinder .... Compound ...... Triple compound . . . Quadruple compound 65 145 185 265 7 30 48 7. 4.8 3.2 2.7 4.6 3.1 2.65 3.0 2.6 2.55 For tri sizes r triple engines, Jay M. Whitham* recommends the following relative of cylinders when the piston-speed is from 750 to 1,000 ft. per minute : Boiler Pressure (above Atmosphere). High-Pressure Cylinder. Intermediate Cylinder. LoAv-Pressure Cylinder. 130 140 150 160 1 1 1 1 2.25 2.40 2.55 2.70 5.00 5.85 6.90 7.25 The following are the maximum, average, and minimum values of the relative cylinder volumes of triple-expansion condensing engines, working with boiler pressures of 150 or 160 lbs. per square inch above atmosphere, on board 65 boats launched within the last three or four years : — Maximum value Average " Minimum " High-Pressure Cylinder. Intermediate Cylinder. Low-Pressure Cylinder. 2.84 2.58 1.89 * American Society of Mechanical Engineers, 1889. 920 «*§[$ * CO © © s os 3 °5 CO os © © © CD © o os CO 8 © co © © © c^ B © © OS us OS us us OS CO OS os © os cl OS co I © *s« -3 CO c 1 a OS *# OS OS s OS OS '-0 © OS c^ OS % os H"g'«3 « CO 1 § s i % © CO oq co 00 co -# £ 1 © 00 CO 1 1 ""Oi; p c g co © 00 t-; 1 <->eo co § t- o s co 1- © co us 00 «>; (M ■* © [^ 1 s H«~ CO eo © — 00 © CI CO 1 1 00 g g g | g CD © CI CO CO co CO CD co co CO CO i © CO co § © 1 1 © g CD © •* US 1 CO us co CO cp @ CO s » ^ us © § CO « o .^ <^°* x"1*" ^^=°" --'"=" /^ .-* V i t £ ? *' ^ s^ ^t't-Z' Af- -t 7- A- y ^^' S s*s+*£* - 7^ * /v _/ /Vv^<^ - - s""===|fe?=7 ?^-^"pPP-"""" :::: :J:i?lfc^:^;$?:^z^^:±~ : ± -hi 3: €z/z^ ^gr S i fnpj? ^^^7,,Z/^+ fl ez \^%t/ Wilder ° ■, hi i i A t f J kY /^ ' A A / / /*'-/ A/m - ^-tt-h-fyl¥m- -4= i=:==|EEi-te^ = iffi?L-S|s$m== == =p== == = == - ii-hH Wft¥A fe : -^hThWf : 8 - -HiHhtffttt - Miliz-Em \WttT4-iTtttt & tttttllAttw - fr-thhtPRiFt - EttctSpr + ±jj±jj#:g s III l\ III III II WHEEL. OR PULLEY DIAMETER IN FEET 926 WATER-POWER. WATER-POWER. IN determining the feasibility of utilizing water-power to operate electri cally the industries of any particular town or city, careful consideration must be given to the following points, viz. : 1. The amount of water-power permanently available. 2. The cost of developing this power. 3. The in- terest on this amount. 4. The total demand for power. 5. The amounts and relative locations of the various kinds of power. 6. The cost of steam plants now in operation. 7. The interest on this amount. 8. Cost of fuel for plants now in operation. 9. Cost of operating present plants. Labor. 10. Cost of maintenance of present plants. 11. The amounts and kinds of electric power already in operation. 12. The distance of transmission. 13. The estimated cost of the hydraulic machinery. 14. The guaranteed efficiency and regulation of the hydraulic machinery. 15. Estimated cost of electric machinery. 16. Estimated cost of line construction. 17. Total cost of operating hydraulic and electric machinery. 18. Total cost of mainte- nance of hydraulic and electric plants. 19. The interest on the total esti- mated cost of proposed plant. 20. The estimated gross income. Charles T. Main makes the following general statements as to the value of a water-power : " The value of an undeveloped variable power is usually nothing if its variation is great, unless it is to be supplemented by a steam- plant. It is of value then only when the cost per horse-power for the doubie- plantis less than the cost of steam-power under the same conditions as mentioned for a permanent power, and its value can be represented in the same manner as the value of a permanent poAver has been represented. " The value of a developed power is as follows : If the power can be run cheaper than steam, the value is that of the power, plus the cost of plant, less depreciation. If it cannot be run as cheaply as steam, considering its cost, etc., the value of the power itself is nothing, but the value of the plant is such as could be paid for it new, which would bring the total cost of run- ning down to the cost of steam-power, less depreciation." Mr. Samuel Webber, Iron Age, Feb. and March, 1893, criticises the state- ments of Mr. Main and others who have made comparisons of costs of steam and of water-power unfavorable to the latter. He says : " They have based their calculations on the cost of steam, on large compound engines of 1000 or more h. p. and 120 pounds pressure of steam in their boilers, and by care- ful 10-hour trials succeeded in figuring down steam to a cost of about §20 per h. p., ignoring the well-known fact that its average cost in practical use, except near the coal mines, is from §40 to $50. In many instances dams, canals, and modern turbines can be all completed at a cost of $100 per h. p.; and the interest on that, and the cost of attendance and oil, will bring water-power up to but about $10 or $12 r>er annum ; and with a man compe- tent to attend the dynamo in attendance, it can probably be safely estimated at not over $15 per h. p. SYNOPSIS OF REPOBT HKQrittED OS WATEH-POWEH JPHOrJEMTY. JLocation. Geographical, etc. Sketch of river and its tributaries. Surrounding country and physical features. Sources ; lakes, springs, etc. Water's head; area drained, nature of, whether forest, swamp, snow- covered mountains, etc. Elevation of head waters and of mouth. Length from main source to mouth. Accessibility ; how and by what routes. Meports. Reports of IT. S. Coast or Geological Survey. Reports of Engineers IT. S. Army. Any other reports. Any estimate by engineers and for what purpose. When it first attracted attention and for what reason. History. REPORT ON WATER-POWER PROPERTY. 927 Rainfall. Average for several years for the drainage area. Maximum, what month. Minimum, what month. Comparison with other similar localities. Volume of Water. Gauging of river if possible. Reports by other engineers. Cubic feet per second flow. Cubic feet per second per mile of watershed = say .2 to .3 cubic feet of total rainfall and a available as water-power. Comparison with other rivers. Reservoirs. Possibility of storing water for dry time. Available Fall. Location of ; accessibility, by what routes. Can power be used locally, or would it be necessary to transmit it, and if so, where to, and distances ? Nature of country over which it would have to be carried. Volume of water in cubic feet per second. Note. — 12 cubic feet falling one foot per second = 1 h.p. (approximate). Horse-Power of River. Calculated from available fall and volume. Horse-power for each fall or dam. Location of dams, dimensions, length, and height, best method of con- struction, estimated cost. Backwater ; volume, andhow far ; what interests disturbed by it ; benefits, if any. Compare power with that of similar rivers. Probable cost of power at clams and transmitted. Applications PossiS»le. Near by ; at distance, stating when and for what. Note industries appli- cable to ; comparison with other applications. Bifew Industries Suggested, and old industries already going to Avhich power is applicable. Cost to these, and comparison with cost of other forms of power already in use. Property of the Company. Land, buildings, water rights, flowage rights, franchises, lines, rights of way. Character of deeds. Probable value. Comparison with other similar properties. Other resources. liabilities. Stocks, bonds, floating debt, other. Earning- Capacity. Probable cost of power per h. p. at power-house. Probable cost of power per h. p. delivered or transmitted. Price for which it can be sold at power-house, and price transmitted or delivered. General Features. Surrounding country, its characteristics, people, cities, and towns, indus- tries, condition of finances. Facilities for transportation, water and rail. Nearness of sources of supplies and sales of products. Horse-Power of a Waterfall. The horse-power of a waterfall is expressed in the following formula : Q = quantity of water in cubic feet flowing over the fall in 1 minute. H— total head in feet, i.e., the distance between the surface of the water at the top of the fall, and that at its foot. In a water-power the head is the distance between the surface of the water in the head-race, and that of the water in the tail-race. 928 WATER-POWER. w — weight of water per cubic foot = 62.36 lbs. at 60° F. Gross horse-power of waterfall = X,.ifJ? ™ or .00189 OH. ooOOO Loss of head at the entrance to and exit from a water-wheel, together witi the friction of the water passing through, reduces the power to the fall that can be developed to about 70 per cent of the gross power of the fall. Horse-Power of a Running- Stream. The power is calculated by the same formula as for a fall, but in this case H=. theoretical head due to the velocity of the water in the stream v2 — where v = velocity of water in feet per second. Q = the cubic feet of water actually impinging against the bucket per minute. Gross horse-power =r .00189 QH. Wheels for use in the current of a stream realize only about .4 of the gross theoretical power. Current motors are often developed to operate in strong currents, such as that of the Niagara River opposite Buffalo, but are of little use excepting for small powers. Such a small fraction of the current velocity can be made use of that a current motor is extremely inefficient. In order to realize power from a current it is necessary to reduce its velocity in taking the power, and to get the full power would necessitate the backing up of the whole stream until the actual head equaled the theoretical. Power of Water Howing- in a Pipe. v2 v2 Hdue to velocity = — - = — - where v = velocity in feet per second. f Hx due to pressure — -, where /= pressure in lbs. per square foot. and w = 62.36 lbs. = weight 1 cubic foot of water. H^ distance above datum line in feet. v2 f Total H— °—+J-+Hi. In hydraulic transmission the work or energy of a given quantity of water under pressure is the volume in cubic feet x lbs. pressure per square foot. Q = cubic feet per second. P = pressure in lbs. per square inch. 144 PO Horse-power = 55Q v = .2618 PQ. Mill-Power. It has been customary in the past to lease water-power in units larger than the horse-power, and the term mill-power has been used to designate the unit. The term has no uniform value, but is different in all localities. Emerson gives the following values for the seven more important water- power. Holyoke, Mass. — Each mill-power at the respective falls is declared to have the right during 16 hours in a day to draw 38 cubic feet of water per second at the upper fall when the head there is 20 feet, or a quantity proportionate to the height at the falls. This is equal to 86.2 horse-power as a maximum. Lowell, Mass, — The right to draw during 15 hours in the day so much water as shall give a power equal to 25 cubic feet a second at the great fall, when the fall there is 30 feet. Equal to 85 h. p. maximum. Lawrence, Mass. — The right to draw during 16 hours in a day so much water as shall give a horse-power equal to 30 cubic feet per second when the head is 25 feet. Equal to 85 h. p. maximum. Minneapolis, Minn. — 30 cubic feet of water per second with head of 22 feet. Equal to 74.8 h. p. COMPARISON OF COLUMNS. 929 Manchester, N. H. — Divide 725 by the number of feet of fall minus 1, and tbe quotient will be the number of cubic feet per second in that fall. For 20 feet fall this equals 38.1 cubic feet, equal to 86.4 h. p. maximum. Cohoes, N.Y. — "Mill-power" equivalent to the power given by 6 cubic feet per second, when the fall is 20 feet. Equal to 13.6 b. p. maximum. Passaic, A". /. — Mill-power : The right to draw 8£ cubic feet of water per second, fall of 22 feet, equal to 21.2 horse-power. Maximum rental, $700 per year for each mill-power = $33.00 per h. p. The horse-power maximum above given is that due theoretically to the weight of water and the height of the fall, assuming the water-wheel to have perfect efficiency. It should be multiplied by the efficiency of the wheel, say 75 per cent for good turbines, to obtain the h.p. delivered by the wheel. At Niagara power has in all cases been sold by the horse-power delivered to the Avheels if of water, and to the building-line if electrical. Charges for water in Manchester, Lowell, and Lawrence, are as follows : About $300 per year per mill-power for original purchases. $2 per day per mill-power for surplus. Lowell. About $300 per year per mill-power for original purchases. $2 per day per mill-power during " back-water." $4 per day per mill-power for surplus under 40 per cent. $10 per day per mill-power for surplus over 40 per cent and under 50 per cent. $20 per day per mill-power for surplus over 50 per cent. $75 per day per mill-power for any excess over limitation. Lawrence. About $300 per year per mill-power for original purchases. About $1200 per year per mill-power for new leases at present. $4 per day per mill-power fur surplus up to 20 per cent. $8 per day per mill-power for surplus over 20 and under 50 per cent. $4 per day per mill-power for surplus under 50 per cent. comparison of coifUnara of wateb ins Mercury in Inches, and Pressure in U»s., per &quai FEET, ■e Inch. Lbs. Water. 1 Merc'ry Water. Merc'ry Lbs. Merc'ry Water. Lbs. Press. 1 Press. Press. Sq. In. Feet. Inches. Feet. Inches. Sq. In. Inches. Feet. Sq. In. 1 2.311 2.046 1 0.8853 0.4327 1 1.1295 0.4S87 2 4.622 4.092 2 1.7706 0.8654 2 2.2590 0.9775 3 6.933 6.138 3 2.6560 1.2981 3 3.3885 1.4662 4 9.244 8.184 4 3.5413 1.7308 4 4,5181 1.9550 5 11.555 10.230 5 4.4266 2.1635 5 5.6476 2.4437 G 13.866 12.2276 6 5.3120 2.5962 6 6.7771 2.9325 7 16.177 14.322 7 6.1973 3.0289 7 7.9066 3.4212 8 18.488 16.368 8 7.0826 3.4616 8 9.0361 3.9100 9 20.800 18.414 9 7.9680 3.8942 9 10.165 4.3987 10 23.111 20.462 10 8.8533 4.3273 10 11.295 4.8875 11 25.422 22.508 11 9.7386 4.7600 11 12.424 5.3762 12 27.733 24.554 12 10.624 5.1927 12 13.554 5.8650 13 30.044 26.600 13 11.509 5.6255 13 14.683 6.3537 14 32.355 28.646 14 12.394 6.0582 14 15.813 6.8425 15 34.666 30.692 15 13.280 6.4909 15 16.942 7.3312 16 36.977 32.738 16 14.165 6.9236 16 18.072 7.8200 17 39.288 34.784 17 15.050 7.3563 17 19.201 8.3087 18 41.599 36.830 18 15.936 7.7890 18 20.331 8.7975 19 43.910 38.876 19 16.821 8.2217 19 21.460 9.2862 20 46.221 40.922 20 17.706 8.6544 20 22.590 9.7750 21 48.532 42.968 21 18.591 9.0871 21 23.719 10.264 22 50.843 45.014 22 19.477 9.5198 22 24,849 10.752 23 53.154 47.060 23 20.382 9.9525 23 25.978 11.241 24 55.465 49.106 24 21.247 10.385 24 27.108 11.7300 25 57.776 51.152 25 22.133 10.818 25 28.237 12.219 26 60.087 53.198 26 23.018 11.251 26 29.367 12.707 27 62.398 55.244 27 23.903 11.683 27 30.496 13.196 28 64.709 57.290 28 24.789 12 116 28 31.626 13.685 39 67.020 59.336 29 25.674 12.549 29 32.755 14.174 30 69.331 61.386 30 26.560 12.981 30 33.885 14.662 930 WATER-POWER. © & X >> o o O £ o o o 95= $15.49 23.18 20.35 29.83 48.79 58.27 105.67 O H= $14.47 22.16 19.33 28.81 47.77 57.25 104.65 $13.46 21.15 18.32 27.80 46.76 56.24 103.64 © 95 $12.45 20.14 17.31 26.79 45.75 55.23 102.63 o 95= $11.44 19.13 16.30 25.78 44.74 54.22 101.62 © m $10.42 18.il 15.28 24.76 43.72 53.20 100.60 o o O o bo o3 O •d 3 5 o o O 95= $10.15 © o 95= CO 95= o 95= 95= O 95= 95= o CO 95= O CO 95 g 95= 00 in 95= •0^9 'sgijddns 'xio ;90irepu9wy d 95= O o $4.62 12.31 9.48 18.96 37.92 47.40 94.80 S £ O 1) (a $300 per year I 2 per day . . 4 " " . . 8 " " . . 10 " " . . 20 " " . . 1 3s is PRESSURE OF WATER. 931 PRESSURE OF WATER. The pressure of water in pounds per square inch for every foot in height to 300 feet ; and then by intervals to 1000 feet head. Feet Press., Feet Press., Feet Press., Feet Press., Feet Press., He'd. Sq. In. He'd. Sq. In. He'd. Sq. In. Head. Sq. In. Head. Sq. In. 1 0.43 65 28.15 129 55.88 193 83.60 257 111.32 2 0.86 66 28.58 130 56.31 194 84.03 258 111.76 3 1.30 67 29.02 131 56.74 195 84.47 259 112.19 4 1.73 68 29.45 132 57.18 196 84.90 260 112.62 5 2.16 69 29.88 133 57.61 197 85.33 261 113.06 6 2.59 70 30.32 134 58.04 198 85.76 262 113.49 7 3.03 71 30.75 135 58.48 199 86.20 263 113.92 8 3.46 72 31.18 136 58.91 200 86.63 264 114.36 9 3.89 73 31.62 137 59.34 201 87.07 265 114.79 10 4.33 74 32.05 138 59.77 202 87.50 266 115.22 11 4.76 75 32.48 139 60.21 203 87.93 267 115.66 12 5.20 76 32.92 140 60.64 204 88.36 268 116.09 13 5.63 77 33.35 141 61.07 205 88.80 269 116.52 14 6.06 78 33.78 142 61.51 206 89.23 270 116.96 15 6.49 79 34.21 143 61.94 207 89.66 271 117.39 16 6.93 80 34.65 144 62.37 208 90.10 272 117.J2 17 7.36 81 35.08 145 62.81 209 90.53 273 118.26 18 7.79 82 35.52 146 63.24 210 90.96 274 118.69 19 8.22 83 35.95 147 63.67 211 91.39 275 119.12 20 8.66 84 36.39 148 64.10 212 91.83 276 119.56 21 9.09 85 36.82 149 64.54 213 92.26 277 119.99 22 9.53 86 37.25 150 64.97 214 92.69 278 120.42 23 9.96 87 37.68 151 65.40 215 93.13 279 120.85 24 10.39 88 38.12 152 65.84 216 93.56 2S0 121.29 25 10.82 89 38.55 153 66.27 217 93.99 281 121.72 26 11.26 90 38.98 154 66.70 218 94.43 282 122.15 27 11.69 91 39.42 155 67.14 219 94.86 283 122.59 28 12.12 92 39.85 156 67.57 220 95.30 284 123.02 29 12.55 93 40.28 157 68.00 221 £5.73 285 123.45 30 12.99 94 40.72 158 68.43 222 96.16 286 123.89 31 13.42 95 41.15 159 68.87 223 96.60 287 124.32 32 13.86 96 41.58 160 69.31 224 97.03 288 124.75 33 14.29 97 42.01 161 69.74 225 97.46 289 125.18 34 14.72 98 42.45 162 70.17 226 97.90 290 125.62 35 15.16 99 42.88 163 70.61 227 98.33 291 126.05 36 15.59 100 43.31 164 71.04 228 98.76 292 126.48 37 16.02 101 43.75 165 71.47 229 99.20 293 12692 38 16.45 102 44.18 166 71.91 230 99.63 294 127.35 39 16.89 103 44.61 167 72.34 231 160.06 295 127.78 40 17.32 104 45.05 168 72.77 232 100.49 296 128.22 41 17.75 105 45.48 169 73.20 233 100.93 297 128.65 42 18.19 106 45.91 170 73.64 234 101.36 298 129.08 43 18.62 107 46.34 171 74.07 235 101.79 299 129.51 44 19.05 108 46.78 172 74.50 236 102.23 300 129.95 45 19.49 109 47.21 173 74.94 237 102.66 310 134.28 46 19.92 110 47.64 174 75.37 238 103.09 320 138.62 47 20.35 111 48.98 175 75.80 239 103.53 330 142.95 48 20.79 112 48.51 176 76.23 240 103.80 340 147.28 49 21.22 113 48.94 177 76.67 241 104.39 350 151.61 50 21.65 114 49.38 178 77.10 242 104.83 360 155.94 51 22.09 115 49.81 179 77.53 243 105.26 370 160.27 52 22.52 116 50.24 180 77.97 244 105.69 380 164.61 53 22.95 117 50.68 181 78.40 245 106.13 390 168.94 54 23.39 118 51.11 182 78.84 246 106.56 400 173.27 55 23.82 119 51.54 183 79.27 247 1C6.99 500 216.58 56 24.26 120 51.98 184 79.70 248 107.43 600 259.90 57 24.69 121 52.41 185 80.14 249 107.86 700 303.22 58 25.12 122 52.84 186 80.57 250 108.29 800 346.54 59 25.55 123 53.28 187 81.00 251 108.73 900 389.86 60 25.99 124 53.71 188 81.43 252 109.16 1000 433.18 61 26.42 125 54.15 189 81.87 253 109.59 62 26.85 126 54.58 190 82.30 254 110.03 63 27.29 127 55.01 191 82.73 255 110.46 64 27.72 128 55.44 192 83.17 256 110.89 932 WATER-POWER. RITE1ED STEEL PIPES. Riveted sheet steel pipe is much used on the Pacific Coast for conveying water for considerable distances under high heads, say as much as 1700 feet. Corrosion of iron and steel pipe has always been an argument against its use, but for about thirty years such pipe has been in use in California; and a life of twenty-five years' is not considered the limit, when both inside and outside of the pipe are treated with a coating of asphalt. The method of covering with asphalt referred to affords perfect protec- tion against corrosion, and so long as the coating is intact, makes it practi- cally indestructible so far as all ordinary wear is concerned. The conditions which interfere with the best service are where the coating is worn off by abrasion in transportation, or Avhere the pipe is subject to severe shock by the presence of air, or by a sudden closing of tbe gates, or where the service is intermittent, causing contraction and expansion, which opens the joints and breaks the covering. With ordinary care these objections can mostly be overcome. While the primary object of coating pipe in this way is to prevent oxidization, and thus insure its durability, it is incidentally an ad- vantage in providing a smooth surface on the inside, which reduces the fric- tion of water in its passage. The Coast method of laying pipe is to take tbe shortest practicable dis- tance that the ground will permit, placing the pipe on the surface and con- necting directly from ditch, flume, or other source of supply to the wheel. Avoid short turns or acute angles, as they lessen the head and produce shock. The ordinary method of jointing is the slip joint, made up in much the same way as stove-pipe. Of course this is only adapted to comparatively low heads, special riveted-joint construction being necessary for the higher falls. In laying such pipe where the lengths come together at an angle, a lead joint should be made. This is done by putting on a sleeve, allowing a space, say three-eighths of an inch, for running in lead. With a heavy pressure, and especially on steep grades, the lengths should be wired together, lugs being put on the sections forming the joints for this purpose; and where the grade is very steep, the pipe should be securely anchored with wire cable. In laying the pipe line it is customary to commence at the wheel, and with slip joint the lower end of each length should be wrapped with cotton drill- ing or burlaps to prevent leaking ; care being taken in driving the joints together not to move the gate and nozzle from their position. Some tempo- rary bracing may be necessary to provide against this. Where several wheels are to be supplied from one pipe line, a branch from the main in the form of the letter Y is preferable to a right angle out- let. When taken from the main at a right angle, the tap-hole should be nearly as large as the main, reducing by taper joint to the size of pipe attached to the wheel gate. It is advised where practicable to lay the pipe in a trench, covering it with earth. Even in warm climates, where this is not necessary as protec- tion from frost, it is desirable to prevent contraction and expansion by variations of temperature, a? well as to afford security against accident. When laid over a rocky surface a covering of straw or manure will protect it from the sun, and generally prevent freezing ; as where kept in motion, water under pressure will stand a great degree of cold without giving trouble in this way. After connections are made, it should be tested before covering to see that the joints are tight. Care should be taken when the pipes are first filled to see that the air is entirely expelled, the use of air valves being necessary in long lines laid over undulating surfaces. Care should also be taken before starting to see that there are no obstructions in the pipe or connections to wheel, and that there are no leaks to reduce the pressure. Pipe lines of any considerable length should be graduated as to size, being larger near the top and reduced toward the lower end, the thickness of iron for various sizes being deter- mined by the pressure it is to carry. This is a saving in first cost, and facilitates transportation by admitting of length, being run inside of each other. When used near railroad stations, pipe is generally made in 27 ft. lengths for purpose of economizing freight, this being the length of a car. When transported long distances by wagon, it is usually made in about 20 ft. lengths. For pipe of large diameter, or for transportation over long: dis- tances, as also for mule packing, it is made in sections or joints of 24 to 30 inches in length, rolled and punched, with rivets furnished to put together WOODEN-STAVE PIPE. 933 on the ground where laid. Pipe of this character, being cold riveted, is easily put together with the ordinary tools for the purpose. In such case, preparation should be made for coating with asphalt before laying. In many cases much expense may be saved in pipe by conveying the water in a flume or ditch along the hillside, covering in this way a large part of the distance, then piping it down to the power station by a short line. This is more especially applicable to large plants, where the cost of the pipe is an important item. DATA JPOM WWJbM]E& AWM DITCHES. To give a general idea as to the capacity of flumes and ditches for carry- ing water, the following data is submitted : The greatest safe velocity for a wooden flume is about 7 or 8 feet per second For an earth ditch this should not exceed about 2 feet per second. In Califor- nia it is the general practice to lay a flume on a grade of about £ inch to the rod, or often 2 inches tothelOOfeet, dependingon the existing conditions. Assuming a rectangular flume 3 feet wide, running 18 inches deep, its velocity and capacity would be as shown below : — Grade. Vel. in Ft. per Sec. Quantity Cu. Ft. Min. | inch to rod 2.6 702 1 " " " 3.7 999 \ " " " 5.3 1,431 As the velocity of a flume or ditch is dependent largely on its size and character of formation, no more specific data than the above can be given. It is not safe to run either ditch or flume more than about f or | full. WOODEflfSTAVE PIPE. Although wooden-stave pipe has been in use for years on old water powers for penstocks, etc., it seems to have been given but little study until late years, when it has been used to some extent on the Pacific Coast for con- veying water long distances under heads not much exceeding 200 feet. Al- though the construction of wooden-stave pipe is quite simple, yet consider- able skill and care are necessary to make water-tight work. One of the latest pieces of work employing this type of pipe is the plant of the San Gabriel Los Angeles Transmission, California, — where several miles of wooden-stave pipe, 48 ins. diameter, are used. The pipe is laid uniformly ten feet below hydraulic grade ; and the wood is of such thickness as to be always water-soaked, and will thus outlast almost any other form of construction. The staves are placed so as to break joints, the flat sides are dressed to a true circle, and the edges to radial planes. The staves are cut off square at the ends, and the ends slotted, a tight-fitting metallic tongue being used to make the joint. The pipe depends upon steel bands for its strength, and in the case above mentioned they are of round steel rod placed ten inches apart from center to center. "Where the pressures vary along the line, bands can be spaced closer or wider apart to make the necessary strength. The preference is given round bands over flat ones, on account of their embedding themselves in the wood better as it swells. They also expose less surface to rust than would flat ones of the same strength. The ends of the bands are secured together through a malleable iron shoe, having an interior shoulder for the head of the bolt, and an exterior shoulder for the nut, the whole band thus being at right angles to the line of the pipe. Where curves are not too sharp, they can easily be made in the wooden pipe ; but for shore turns, sec- tions of steel-riveted pipe of somewhat larger internal diameter than that of the wooden pipe are introduced. The joints between wood and steel are made by a bell on the steel pipe that is larger than the outside diameter of the wooden pipe. After partly filling the space between bell and Avood with oakum packed hard, for the remainder use neat Portland cement. Advantages claimed for this type are that it costs less than any other form, and especially so where transportation is over the rugged country where it is most liable to be used ; great length of life, and greater capacity than either cast-iron or steel-riveted. Compared with new riveted pipe, the carrying capacity of stave pipe is said to be from 10 to 40 % more, and this difference increases with age as the wooden pipe gets smoother, while the friction of the metal pipe increases to a considerable degree. As compared with open flumes, the life is so much greater and repairs so much less as to considerably more than counterbalance the first cost. For detailed information on wooden-stave pipe, see papers by A. L. Adams, September, 1898, Am. Soc. C. E. 934 WATER-POWER. TAJBI,E Ol RIVETED HYDRAUIIC JPIP£. (Pelton Water Wheels Co.) Showing weight, with safe head for various sizes of double-riveted pipe. 05 ft £ °J !£.a« 55 .S Ift ft *S2 l?-d 05 .3 ft . ft . 00 ^ g«Seo a«- P ft 00 P ""' " 5 0 W Sco73 o 2 2 a 3.2 n '- a ^ 2, "~ i - _ u — z z ■rt ft to -a ft£- EC 5 £ ? ft? ® . ® ^ "* 3.2^55 3*" P J z 'z. 2^05 r-i.a so a ~^ W5£ 5 u • c~ £ £ a? | pj ft>» ^ Oa>sa Mis "5 05 . 3 18 400 9 2 18 254 16 165 320 16* 4 12 18 350 16 5** 18 "54 14 252 320 27* 4 12 16 525 16 3 18 18 254 254 12 11 385 424 320 320 5 20 18 325 25 3* 30 5 20 20 16 14 500 675 25 25 4 5 18 254 10 505 320 34 5 20 20 314 314 16 14 148 227 400 400 18 6 28 18 296 36 4i 22* fi 28 16 487 36 5| 7* 20 314 12 346 400 30 6 28 14 743 36 20 20 314 344 11 10 380 456 400 400 32* 36* 7 38 38 18 16 254 419 50 50 1 8* 7 22 380 16 135 480 20 7 38 14 640 50 22 22 22 22 3SO 3S0 3X0 380 14 206 316 347 415 480 480 480 480 24| 32} 35} 40 8 8 8 50 50 50 16 14 12 367 560 854 63 63 63 9* 13 11 10 9 9 9 63" 63 63 78 78 78 78 78 16 14 12 16 14 12 11 10 327 499 761 80 80 80 8* lOf 14* 24 24 24 24 24 452 452 452 452 452 14 12 11 10 8 188 290 318 379 466 570 570 570 570 570 271 35* 39 43* 53 10 10 10 10 10 295 450 687 754 900 100 100 100 100 100 91 Hf 15| 17* 19i 26 26 26 26 26 530 530 530 530 530 14 12 11 10 8 175 267 294 352 432 670 670 670 670 670 29* 38* 42" 47 11 95 95 95 95 95 16 14 12 11 10 269 412 626 687 820 120 120 120 120 120 9| 57i 11 11 11 17* 181 21 28 28 28 28 615 615 615 615 14 12 11 10 102 247 273 327 775 775 775 775 31i 41J 45 501 611 12 12 12 12 12 113 113 113 113 113 132 132 132 132 132 16 14 12 11 10 16_ 14 12 11 10 246 377 574 630 753 228 348 530 583 696 142 142 142 142 142 111 2S 615 8 400 775 181 19| 22| 30 30 30 30 30 Tor; 706 706 706 70(] 12 11 10 8 7 231 254 304 375 425 890 890 890 890 890 44 48 54 65 13 13 13 13 13 170 170 170 170 170 12 74 15 20 22 24* 36 36 36 36 1017 1017 1017 017 11 10 8 7 141 155 192 210 1300 1300 1300 1300 58 67 78 88 14 14 14 14 14 153 153 153 153 153 16 14 12 11 10 211 324 494 543 648 200 200 200 200 200 13 16 2H 23i 26 ~~ 13f 17 23 24* 28 40 40 40 40 40 42 42 42 42 42 42 1256 256 256 256 1256 3N5 :;s5 3X5 ::x5 13S5 3X5 10 8 7 6 4 10 8 7 6 4 1 141 174 189 213 250 1600 1600 1600 1600 1600 71 86 97 108 126 176 176 176 J76 176 201 16 14 12 11 10 ~16~ 197 302 460 507 606 225 225 225 225 225 15 15 15 15 15 135 165 180 210 240 270 1760 1760 1660 1760 1760 1760 74* 91 102 114 133 16 185 255 14* 137 16 201 14 283 255 17| 42 3X5 3 300 1760 145 16 20.1 12 432 255 24* 42 3X5 J> 321 1760 177 16 201 11 474 2*5 2(U 42 ::^> I 363 1760 216 16 201 1 10 I 567 255 29* FLOW OF WATER. Cubic Feet of Water per Ulinute Discharged Through an Orifice 1 Square Inch in Area. For any other size oj orifice, multiply by its area in square inches. ^ © ^3 o5 © ® "■3 /h x a. The best form of aperture for giving the greatest flow of water is a coni- cal aperture whose greater base is the aperture, the height or length of the section of cone being half the diameter of aperture, and the area of the small opening to the area of the large opening as 10 to 16 ; there will be no contraction of the vein, and consequently the greatest attainable discharge will be the result. JIHAHi I' J I E 31 E2I" T OF FLOW OF STllEAJfl. WATJEM O A. The quantity of water flowing in a stream may be roughly estimated as follows : Find the mean depth of the stream by taking measurements at 10 or 12 or more equal distances across. Multiply this mean depth by the width of the stream, which will give the total cross-section of the prism. Find the velocity of the flow in feet per minute, by timing a float over a measured distance, several times to get a fair average. Use a thin float, such as a shingle, so that it may not be influenced by the wind. Fig. 13. The area or cross-section of the prism multiplied by the velocity per min- ute will give the quantity per minute in cubic feet. Owing to friction of the bed and banks the actual flow is reduced to about 83 per cent of the calculated flow as above. HORSE-POWER OF AVATER. 937 miners' Inch Measurements. (Pelton Water Wheel Co.) Miners' inch is a term much in use on the Pacific Coast and in the mining regions, and is described as the amount of water flowing through a hole 1 inch square in a 2-inch plank under a head of 6 inches to the top of the orifice. Fig. 13 shows the form of measuring-box ordinarily used ; and the follow- ing table gives the discharge in cubic feet per minute of a miners' inch of water, as measured under the various heads and different lengths and heights of apertures used in California. .5 Openings 2 Inches High. Openings 4 Inches High. Itj Head to Head to Head to Head to Head to Head to 3&a Center, Center, Center, Center, Center, Center, . 5 Ins. 6 Inches. 7 Inches. 5 Inches. 6 Inches. 7 Inches. Cu.Ft. Cu. Ft. Cu. Ft. Cu. Ft. Cu. Ft. Cu. Ft. 4 1.348 1.473 1.589 1.320 1.450 1.570 6 1.355 1.480 1.596 1.336 1.470 1.595 8 1.359 1.484 1.600 1.344 1.481 1.608 10 1.361 1.485 1.602 1.349 1.487 1.615 12 1.363 1.487 1.604 1.352 1.491 1.620 14 1.364 1.488 1.604 1.354 1.494 1.623 16 1.365 1.489 1.605 1.356 1.496 1.626 18 1.365 1.489 1.606 1.357 1.498 1.628 20 1.365 1.490 1.606 1.359 1.499 1.630 22 1.366 1.490 1.607 1.359 1.500 1.631 24 1.366 1.490 1.607 1.360 1.501 1.632 26 1.366 1.490 1.607 1.361 1.502 1.633 28 1.367 1.491 1.607 1.361 1.503 1.634 30 1.367 1.491 1.608 1.362 1.503 1.635 40 1.367 1.492 1.608 1.363 1.505 1.637 50 1.368 1.493 1.609 1.364 1.507 1.639 60 1.368 1.493 1.609 1.365 1.508 1.640 70 1.368 1.493 1.609 1.365 1.508 1.641 80 1.368 1.493 1.609 1.366 1.509 1.641 90 1.369 1.493 1.610 1.366 1.509 1.641 100 1.369 1.494 1.610 1.366 1.509 1.642 Note. — The apertures from which the above measurements were obtained were through material 1\ inches thick, and the loicer edge 2 inches above the bottom of the measuring-box, thus giving full contraction. FLOW OF WATER OVER WEIRS. IVeir Dam Measurement. (Pelton Water Wheel Co.) Place a board or plank in the stream, as shown in Fig. 14, at some point where a pond will form above. The length of the notch in the dam should be from two to four times its depth for small quantities, and longer for large quantities. The edges of the notch should be beveled toward the intake side as shown. The overfall below the notch should not be less than twice its depth, that is, 12 inches if the notch is 6 inches deep, and so on. In the pond, about 6 feet above the dam, drive a stake, and then obstruct the water until it rises precisely to the bottom of the notch, and mark the stake at this level. Then complete the dam so as to cause all the water to flow through the notch, and, after time for the water to settle, mark the stake again for this new level. If preferred, the stake can be driven with its top precisely level with the bottom of the notch, and the depth of the water be measured with a rule after the water is flowing free, but the marks 938 WATER-POWER. are preferable in most cases. The stake can then be withdrawn ; and the distance between the marks is the theoretical depth of flow corresponding to the quantities in the table. JFrancis's formulae for Weirs. As given by Francis. Weirs with both end contractions ) suppressed j Weirs with one end contraction ) suppressed J Weirs with full contraction . . Q — 3.33111* Q — 3.33(1 — Q = 3.33(1 - As modified by Smith. 3.29 (l + J-) ft1 .lft) ft* 3.29lh :llnlr 3.29 l — ^\ ft2 The greatest variation of the Francis formube from the value of c given oy Smith amounts to 3* per cent. The modified Francis formulae, says Smith, will give results sufficiently exact, when great accuracy is not required, Within the limits of ft, from .5 feet to 2 feet, I being not less than 3 ft. Q = discharge in cubic feet per second, / = length of weir in feet, h = effective head in feet, measured from the level of the crest to the level of still water above the weir. If Qf = discharge in cubic feet per minute, and V and ft' are taken in inches, the first of the above formulae reduces to Q' — QAl'h'* • The values are suf- ficiently accurate for ordinary computations of water-power for weirs without end contraction, that is, for a weir the full width of the channel of approach, and are approximate also for weirs with end contraction when I = at least 10ft, but about 6 per cent in excess of the truth when / = 4ft. Weir 'Fable. Table Showing the Quantity of Water Passing over Weirs in Cubic Feet per Minute, d _j^ © ^ . s,q «|S°1 a& -t^o © .- y T! 3 Tj fa -3 -2 U MO d"<5 d © cS cS ^> o o - d 53 3 2 © ►?> q aaW? 0 0 s 5 © ^ fa.S=V- 0 yd fa !§ . 1 1-d^-l d S | § © g 0 0 -h "B © u III H£~£# &u K ©<{-r© 6&a©3^ 1 4.85 4f 50.20 8* 120.18 12* 214.32 H 5.78 *t 52.18 8f 122.82 12| 220.76 n 6.68 5 54.22 8| 125.52 13 227.30 if 7.80 5| 56.25 82 128.14 13i 233.92 1* 8.90 5i 58.33 9 130.93 13* 240/4 if if 10.00 5| 60.42 94 91 133.65 13| 247.22 11.23 5* 62.55 136.43 14 254.03 if 12.45 5f 64.68 91 139.18 14i 260.83 2 13.72 5| 66.86 9* 141.99 14* 267.77 2* 15.02 5£ 68.98 9| 144.80 14| 274.70 21 16.36 6 71.27 9| 147.64 15 281.72 2| 17.75 6* 73.45 n 150.47 15i 15* 288.82 2* 19.17 6| 75.77 10 153.35 295.93 2f 2| 20.63 6| 78.04 10* 15G.20 15| 303.10 22.11 6* 80.36 104 159.14 16 310.36 2£ 23.63 6f 82.63 10-1 162.07 16i 317.69 3 25.20 6| 85.04 10* 164.99 16* 325.03 3! 26.78 6| 87.43 lOf 167.89 16| 332.42 28.43 7 89.82 101 169.92 17 339.91 3§ 30.06 71 92 16 10J 173.90 171 17* 347.45 &h 31.75 7| 94.67 11 176.92 355.02 3f 33.45 97.11 114 179.94 17| 362.77 3| 35.22 7? 99.50 111 182.99 18 370.34 3£ 36.98 74 102.10 111 186.03 ISi 378.12 4 38.80 7| 104.63 11* 189.13 18* 385.87 a 40.63 7| 107.13 lit 192.20 18| 393.66 42.49 8 109.74 111 195.32 19 401.63 4f 44.39 8* 112.31 112 198.47 19* 409.58 4£ 46.29 1 114.91 12 201.59 19* 417.48 4f 48.22 117.51 12i 207.94 19| 425.68 HORSE-POWER OF WATER. 939 TABLE! FOR CAICIIATIJG TH11 HORiE-POWER OJP WATE1*. (Pelton Wheel Co.) Miners' Inch T able. Cubic JFeet Table. The following table gives the horse- The following table gives the powei of one miners' inch of water hor e-power of one cubic foot of under heads from one up hundred feet. This inch to eleven water per minute under heads from equals 1J one up to eleven hundred feet. cubic feet per minute. '" _^ Horse- .5 Horse- -§ CD Horse- ■rt ^j Horse- s£ Power. "31 0>R Power. Power. !l Power. M w w w 1 .0024147 320 .772704 1 .0016098 320 .515136 20 .0482294 330 .796851 20 .032196 330 .531234 30 .072441 340 .820998 30 .048294 340 .547332 40 .096588 350 .845145 40 .064392 350 .563430 50 .120735 360 .869292 50 .080490 360 .579528 60 .144882 370 .893439 60 .096588 370 .595626 70 .169029 380 .917586 70 .112686 380 .611724 80 .193176 390 .941733 80 .128784 390 .627822 90 .217323 400 .965880 90 .144892 400 .643920 100 .241470 410 .990027 100 .160980 410 .660018 110 .265617 420 1.014174 110 .177078 420 .676116 120 .289764 430 1.038321 120 .193176 430 .692214 130 .313911 440 1.062468 130 .209274 440 .708312 140 .338058 450 1.086615 140 .225372 450 .724410 150 .362205 460 1.110762 150 .241470 460 .740508 160 .386352 470 1.134909 160 .257568 470 .756606 170 .410499 480 1.159056 170 .273666 480 .772704 180 .434646 490 1.183206 180 .289764 490 .788802 190 .458793 500 1.207350 190 .305862 500 .804900 200 .482940 520 1.255644 200 .321960 520 .837096 210 .507087 540 1.303938 210 .338058 540 .869292 220 .531234 560 1.352232 220 .354156 560 .901488 230 .555381 580 1.400526 230 .370254 580 .933684 240 .579528 60) 1.448820 240 .386352 600 .965880 250 .603675 650 1.569555 250 .402450 650 1 .046370 260 .627822 700 1.690290 260 .418548 700 1.126860 270 .651969 750 1.811025 270 .434646 750 1.207350 280 .676116 800 1.931760 280 .450744 800 1.287840 290 .700263 900 2.173230 290 .466842 900 1.448820 300 .724410 1000 2.414700 300 .482940 1000 1.609800 310 .748557 1100 2.656170 310 .499038 1100 1.770780 When the Exact Head is found in Above Table. Example.— Have 100 foot head and 50 inches of water. How many horse-power ? By reference to above table the horse-power of 1 inch under 100 feet head is .241470. The amount multiplied by the number of inches, 50, will give 12.07 horse-power. When Exact Head is not found in Table. Take the horse-power of 1 inch under 1 foot head, and multiply by the number of inches, and then by number of feet head. The product will be the required horse-power. The above formula will answer for the cubic-feet table, by substituting the equivalents therein for those of miners' inches. Note. — The above tables are based upon an efficiency of 85 percent. 940 WATER-POWER. WATER.WHEELi. Undershot TO'heels, in which the water passes under acting by im- pulse, when constructed in the old-fashioned way with flat boards as floats, have a maximum theoretical efficiency of 50 per cent ; but Avith curved floats, as in Poncelet's wheel, which are arranged so that the water enters without shock and drops from the floats into the tail-race without horizontal velo- city, the maximum efficiency is as great as for overshot wheels, and the available efficiency is found to be about (30 per cent. The velocity of the periphery should be about .5 of the theoretical velocity of the water due to the head. .Breast and Overshot Wheels. The best peripheral velocity is about 6 feet per second, and for the water supplied to it about 12 feet per second, which is the velocity due to a fall of about 2J feet ; therefore, the point at which the water strikes the wheel should be 2\ feet below the top-water level. The chief cause of loss in over- shot wheels is the velocity which the water possesses at the moment it falls from the float or bucket ; overshot wheels are good for falls of 13 feet to 20 feet ; below that breast wheels are preferable. The capacity of the buckets should be three times the volume of water held in each. The distance apart of the buckets may be 12 inches in high-breast and overshot wheels, or 18 inches in low-breast wheels, while the opening of buckets may be 6 to 8 inches in high-breast, and 9 inches to 12 inches in low-breast wheels. TUHJBEMElS. These may be divided into two main classes, viz., pressure and impulse turbines. The former may be again divided into the following : parallel- flow, outward-flow, and inward-flow turbines, according to the direction in which the water flows through the turbine in relation to its axis. IParallel-flow turbines, sometimes called downward-floAV, are best suited for low falls, not exceeding say 30 feet. Fontaine's turbine is of this class, the wheel being placed at the bottom of the water-pipe or flume, just above the level of the tail-race. The water passes through guide blades and strikes the curved floats of the wheel. Jonval's turbine is of similar type, but is arranged to work partly by suction, and may be placed above the level of the tail-race Avithout loss of power, Avhich is often more convenient for Avorking. The efficiency is from 70 to 72 per cent Avith well-designed wheels of this type. Fig. 15. Victor Wheel set in ordinary Flume. Outward-flow Turbines have a somewhat higher efficiency than the parallel-floAv — as much as 88 per cent has been realized by Boyden's tur- bine ; Fourneyron's has given a maximum of 79 per cent. Inward-flow Turbines have been designed by SAvain and others. Tests made on a SAvain turbine by J. B. Francis gave a maximum effi- ciency of 84 per cent Avith full supply, and with the gate a quarter open 61 per cent, the circumferential velocity of the Avheel ranging from 80 to 60 per cent of the theoretical velocity due to the head of Avater. In Swain's turbine the edges of the floats are vertical and opposite the guide blades, DIMENSIONS OK TURBINES. 941 the edges towards the bottom of the floats being bent into a quadrant form. The Victor turbine is claimed to give 88 per cent under favorable conditions. It receives the water upon the outside, and discharges it downward and out- ward, the lines of discharge occupying the entire diameter of the lower portion of the wheel, excepting only the space tilled by the lower end of the shaft. Impulse Turbines are suitable for very high falls. The Girard and Pelton are both of this type. It is advised that pressure turbines be used on heads of 80 feet or 100 feet, but above this an impulse turbine is best. A Girard turbine is working under a fall of 650 feet. Installing* Turbines. Particular attention must be paid to the designing and construction of water-courses. The forebay leading to the flume should be of such size that the velocity of the water never exceeds 1^ feet per second, and should be free from abrupt turns or other defects likely to cause eddies. The tail-race should have similar capacity and sufficient depth below the surface of the stream to allow at least 2 feet of dead water standing when the wheels are not in motion, and with large wheels, 3 feet to 4 feet ; after extending sev- eral feet beyond the flume, this may be gradually sloped up to the level of the stream. It is not uncommon to see 2 feet or 3 feet of head lost in defective races. When setting turbines some distance above the tail-race, the mouth of the draft-tube must be 2 inches to 4 inches below the lowest level of the stand- ing tail-water. Theoretically draft-tubes may be 30 feet long ; but 20 feet is as long as is desirable on account of the difficulty of keeping air-tight ; they should be made as short as possible by placing the turbine at the bottom of the fall. Particulars of the setting recommended for Victor turbines are given below, as an example. Table of Dimensions of Victor Turbine. A. B. C. D. E. F. K. "3 01 o Diameter of Cylinder pass- ing through Floor of Flume. 0 fl 2 2 S S S 5SS Length of Shaft from Flange Resting on Floor of Flume to Center of Coupling. Diameter of Bore of Upper Half of Coup- ling. ^ 5,2 0 |^ 6 S 55^ Depth of Pit from End of Cylinder to Bottom of Wheel-Pit. 4) 2 0 6 ^•3 2 In. In. In. Ft. In. In. In. 2&£:i Lbs. 6 10 131 20J 2 12 1 5i •cDrt.g-S 165 8 10 13£ 16 2* 3 19H 22$ 1/s lis n l£11 260 350 12 18i3s 23& 31 28i lit 9§ ga_g* 500 15 fi 28 & 4 33| Vb 11 g >>'-?S $ 830 m 31i 35* 5 351 m 12| rt-S s 2 1125 20" 30J 6 37j 3& 13i as^ 3 1475 22J 33| 35i 38£ 6i 42 h% 14*- 2s'ofl 1900 25 40f 6* 43f 3/1 15| « ^Bro 2335 27J 38i 43| 7l 48i| 4| 16i S a © « 3225 30 40! 46 8 50! a 3540 321 43! 49i 9 55f n W^^'-P 4500 35 46i 53" 9 59 H 20 3 2-d§ 5450 40 52i 60^ 10 64f 5| 22 ^ % go-S 7500 44 48 56J 60J 65J 70i 11 12 67J 74f 1 24 26 ool 0 A 9380 11700 55 68 80 14 85J 28 r" O O ° ^ N N 2 C 2 19000 69 80i 92 16 96i '~g 32 'S-So^ BUJIEjYSI©]*® ©JP TI'ItBJIlHrES. Tables of sizes of turbine wheels vary so much under different makers, and are so extensive, as not to permit their insertion here, but through the kindness of Mr. Axel Ekstrom of the General Electric Company I am per- mitted to print the following sheets of curves for the McCormick type turbine and the Pelton impulse wheel. From them may be made deter- minations of dimensions in much shorter time than is necessary by use of tables. 942 WATER-POWER. § LL c > s g 3 z •§ 5 £ r-V- o CO « g ^A H i W X Ji V co £ " ui g 3 *-£ X Hiu £ S S r\ \ UJ a. OT h 5 -VVX zif < 3 °v^\ §^g J g 0 -£-\-m sSl Z H '" S^V^ £z§ J o o ^ SvxXV s§ w 3 ° ^" , \SvV\i ^^ Ig„Q y \Niva 2 Ts'*i ^ ^ §V\SV\^T . °-':°- ±z in V N,^v^VO^\ -:s0hJ ^te/ t- o^pj «^ sX^Njv, - x s « t y _ ^ *1^ o xN^sVsss* So ^ ^Pffl raN0*.X^Sv&\ sOvJn^n^sSSS g * y cd^4^ ■ s:>^^^555$§ a ^*^2>^^§§S IS^ s-4 77- ' » ^-^-^^r^SS ^^ -' 7 ti>' ^r7r " ~~~~" — --^rlpfe^is ^ ± s-s-^-^-s-^fffl3 — || 1 — ] — | — a va h|---j 2 ° co-o . o —- -,j-~ ^ -- ■" 2 o gJ ^ j rJ-| — |— | | | y*"^=— 7S0G3"-:>!|E^- — EjE- ====-=i J-- ^£0ii i2>v7x/t =~r T\X ^ ^ ""■=» "° //L/Z/V,-0'!0'' \\V H. ■>-. "^r^-i,^ £ ■—^^-^^^^-^^'^.^'^^ ^y// = i \\,xv\ >^ ""»«*<, S ^-'Z^-'^^'^ ^^ <^fs'' /'* yy/7th^^ hX \\^ s ^s. "s Xs, ^-^^ '/iJ-LLzJLit xsW ^^ .** *" ' ^^Z^'^'l''^'' ^ s / 7/7S^T^ ^ v^ v »* ""--"".a;^' ^ ^ t^_ / / t±p°±^ NNC^ s^ ^ § ^^ ^^ ^ y^ 7 . '/ .rfz*Tl± K\^T\ ^ ^^^ ^- « ^ -" ^ y y /' A J so „LT s \ \ s v 'v s^ ^ ^7 * S-, ' ' iS d0,09T l > \ N N. "a-^ *>' *' 7-7 ^_,/ -SooU1^ ^XvN^ ^ ^^ "^ *' 7 7 7 ""N ^^ "s^ ^ ^ Z ^ ' |o 1 .3 > \\ ^ Sy "**. ^ ^ ^ 7 "7 x « ot|00i' ° v ^ >. >^ \ ^^ \^ 7 7 A > = 1 ^ ^ a v 2 ■N "s. ^ -7 7 ^ i- m 00|0c"|• j. S; ^ ^ ^7 ^ v H ^;- 'a' ^ -^ ^ v- ^ s- ^ ^ 7 K > oil g j ^^ ^N -*' ^ -V ■ 7 7 -f nP oil < ^. "n 53 7 ^ ^v ^ / a 2 =,11 ; .-. - s^ W^ -/ ""1 f 5 "n V Zt bIt r- DIMENSIONS OF TURBINES. 943 |-i g OO00T > s §1 J * i z '° 3 0<]0?l - 4- A 4 ^ V V v S S S g "/ V -X s^ 5- T1^ A y X 3 \ \ _ e> -' \ ^tX A i o e oooc °f ^ i£y V X d£ 4 m,=>\ y £y c^ *r S v 3 a S 00|0" "- d/ - "S- 'A 5 V | g 00'0i h T ~i — 5 — 5~ _g -S s 0V3H *ld ~TTT++#ff S "S "§ ^T"- ^z.^m%\^ ^.^■^ >^A*isX- v^ ^-""^^^ ^--^-few--^ L o ^-""*~ ^^ i^ SrA — — ■"" ,-»'' ^ V '!§ A ^ ^«< *>■*. y Z / B2 v s ""* ^ 7 ZHg.§. V "*•> > / ^J£° S ^s' ^ 1 Zg.guj \ ^■S >'*" X Z -h_2 g z <_ ^^ :^s^ : ^ / JuirocoS N ' ^4. /_ -v Zfc; g g s \ >^^ 7 7- £iSS£ \ yS ;/ y s g g ,_: ^ V x ^ >s ^ ~~ 'S g g 2 ° -N^s - * *=/ 1 >. s - 5 Jfc _ N^ _ / |ss° X * Za A *3 3 ^ vv Z£> IF y gg £ _X ZIfgl "N- Z "p 2 it SS < 2 Hp- = ^ d = \ 4 125 H.P. supported against^ For cold. 8/ springing. r>lled iron jyp. _ «£. d_ i / <& ^ -f- L ' - 75 ' * = $? }^^n,H.P. = ^,d = ^ L 55 T For line shafting i For iron, H.P.= — \ hangers 8 feet apart. For transmission H.P. (Foriron,H.P. = ^,d = ^^ simply, no pul- J F ld_ ^ S/z^HP, ieys- r'lld iron, H.P.= — — ; d = V 35 ' " r 35 ' Jones and Laughlin's use the same formulae, with the following excep- tions : ~ 50' For transmission and for short-counters, _ ^/50 H.r Turned iron H.P. = -c— ; d=y d3r , t3/3( Cold-rolled iron H.P. Pulleys should be placed as near to hearings as practicable, but care should be taken that oil does not drip from the box into the pulley. The diameter of a shaft safe to carry the main pulley at the center of a bay may be found by multiplying the fourth power of the diameter obtained by the formulae above given, by the length of the bay, and dividing the pro- duct by the distance between centers of bearings. The fourth root of the quotient will be the required diameter. The following table is based upon the above rule, and is substantially correct : 945 946 SHAFTING, PULLEYS, BELTING, ETC. ° § o . i Diame er of Shaft necessary to carry the Load at the Center of % ~ *• a Bay, which is from Center to Center of Bearings. S^'^ii'S 2} ft, 3 ft. 3i ft. 4 ft. 5 ft. 6 ft. 8 ft. 10 ft. S 32 .0 s w in. in. in. in. in. in. in. in. in." 2 2i 24 2f 2i 2f 2| ^ 3 2i 2i 2| 21 ■ 2| 3 3i 3f 3f 3 3 31 34 3§ 3* 3| 4 44 3* 3i 3| 3| 4 4| 4* 4| 4 4 4i 44 4i 4f 5i 5| a 4£ 4f 4f 5| 5£ a 5 5 5i 5§ *t 6 5i 6i 3 6 <% 6f 6 6 6f u 7l Should the load be placed near one end of the bay, multiply the fourth power of the diameter of shaft necessary to safely carry the load at the cen- ter of the bay (see above table) by the product of the two ends of the shaft, and divide this product by the product of the two ends of the shaft where the pulley is placed in the center. The fourth root of this quotient will be the required diameter. A shaft carrying both receiving and driving pulleys should be figured as a head-shaft. Reflection of Shafting*. (Pencoyd Iron Works.) As the deflection of steel and iron is practically alike under similar con- ditions of dimensions and loads, and as shafting is usually determined by its transverse stiffness rather than its ultimate strength, nearly the same dimensions should be used for steel as for iron. For continuous line-shafting it is considered good practice to limit the deflection to a maximum of Tiff of an inch per foot of length. The weight of bare shafting in pounds — 2.6 d2L = W, or when as fully loaded with pulleys as is customary in practice, and allowing 40 lbs. per inch of width for the vertical pull of the belts, experience shows the load in pounds to be about 13 (PL = W. Taking the modulus of transverse elasticity at 26,000,000 lbs., we derive from authoritative formulas the following : L — ^873 d2, d — V -jSL for bare shafting; L—^j 175 d2, d -. 175 , for shafting carrying pulleys, etc.; L being the maximum distance in feet between bearings for continuous shafting subjected to bending stress alone, d = diam. in inches. The torsional stress is inversely proportional to the velocity of rotation, while the bending stress will not be reduced in the same ratio. It is there- fore impossible to Avrite a formula covering the whole problem and suffi- ciently simple for practical application, but the following rules are correct within the range of velocities usual in practice. For continuous shafting so proportioned as to deflect not more than T^j of an inch per foot of length, allowance being made for the weakening effect of key-seats, ,$ 50 H. P. ~,L— ^700rf2 for bare shafts ; SHAFTING. 947 , L — %J 140rf2, for shafts carrying pulleys, etc. d =r diam. in inches, L = length in feet, r = revols. per minute. The following table (by J. B. Francis) gives the greatest admissible dis- tances between the bearings of continuous shafts subject to no transverse strain, except from their own weight. Distance between Bearings in ft. Distance between Bearings in ft. iam. of Shaft, Wrought-iron Shafts. Steel in inches. Shafts 6 22.30 22.92 7 23.48 24.13 8 24.55 25.23 9 25.53 26.24 Diam. of Shaft, Wrought-iron Steel in inches Shafts. Shafts 2 15.46 15.89 3 17.70 18.19 4 19.48 20.02 5 20.99 21.57 The writer prefers to apply a formula in all cases rather than use tables, as shafting is nearly always one-sixteenth inch less in diameter than the sizes quoted. The following tables are made up from the formulae first given in this chapter. Horse-f»o*ver Transmitted oy Turned Iron Shafting-. As Prime Mover or Head Shaft well Supported by Bearings. s= Revolutions per Minute. 5 60 80 100 125 150 175 200 225 250 275 300 Ins. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. I* 2.6 3.4 4.3 5.4 6.4 7.5 8.6 9.7 10.7 11.8 12.9 2 3.8 5.1 64 8 9.6 11.2 12.8 14.4 16 17.6 19.2 2+ 5.4 7.3 8.1 10 12 14 16 18 20 22 24 2\ 7.5 10 12.5 15 18 22 25 28 31 34 37 2* 10 13 16 20 24 28 32 36 40 44 48 3 13 17 20 25 30 35 40 45 50 55 60 34 3i 16 22 27 34 40 47 54 61 67 74 81 20 27 34 42 51 59 68 76 85 93 102 m 25 33 42 52 63 73 84 94 105 115 126 4 30 41 51 64 76 89 102 115 127 140 153 U 43 58 72 90 108 126 144 162 180 198 216 5 60 80 100 125 150 175 200 225 250 275 300 H 80 106 133 166 199 233 266 299 333 366 400 Approximate Centers of Bearing's for ^Wrought Iron Line Shafts Carrying- a fair Proportion of I*ulleys. Shaft, Diameter Inches . . 1J If 2 2| 2J 2| 3 3i 4 4 c. to c. Bearings — Feet . . 7 7J 8 84 9 9i 10 11 12 13 Shaft, Diameter Inches . . 5 5i 6 6J 7 1\ 8 9 10 c. to c. Bearings — Feet . . 13£ 14 15 15| 16 17 18 19 20 948 SHAFTING, PULLEYS, BELTING, ETC. Line-shafting, Bearings 8 ft. Apart. 5 Revolutions per Minute. 100 125 150 175 200 225 250 275 300 H.P. 325 H.P. 350 Ins. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. 1* 6 7.4 8.9 10.4 11.9 13.4 14.9 16.4 17.9 19.4 20.9 1* 7.3 9.1 10.9 12.7 14.5 16.3 18.2 20 21.8 23.6 25.4 2 8.9 11.1 13.3 15.5 17.7 20 22.2 24.4 26.6 28.8 31 2£ 10.6 13.2 15.9 18.5 21.2 23.8 26.5 29.1 31.8 34.4 37 2i 12.6 15.8 19 22 25 28 31 35 38 41 44 1 15 18 22 26 29 33 37 41 44 48 52 17 21 26 30 34 39 43 47 52 56 60 2| 23 29 34 40 46 52 58 64 69 75 81 3 30 37 45 52 60 67 75 82 90 97 105 3i 38 47 57 66 76 85 95 104 114 123 133 34 47 59 71 83 95 107 119 131 143 155 167 3* 58 73 88 102 117 132 146 162 176 190 205 4 71 89 107 125 142 160 178 196 213 231 249 POWER TRANSMISSION ONLY. a Revolutions per Minute. 3 100 125 150 175 200 233 267 300 333 367 400 Ins. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. :i 6.7 8.4 10.1 11.8 13.5 15.7 17.9 20.3 22.5 24.8 27.0 8.6 10.7 12.8 15 17.1 20 22.8 25.8 28.6 31.5 34.3 if 10.7 13.4 16 18.7 21.5 25 28 32 36 39 43 i* 13.2 16.5 19.7 23 26.4 31 35 39 44 48 52 2 16 20 24 28 32 37 42 48 53 58 64 t 19 24 29 33 38 44 51 57 63 70 76 22 28 34 39 45 52 60 68 75 83 90 2| 27 33 40 47 53 62 70 79 88 96 105 1 31 39 47 54 62 73 83 93 104 114 125 41 52 62 73 83 97 111 125 139 153 167 3 54 67 81 94 108 126 144 162 180 198 216 3i 68 86 103 120 137 160 182 205 228 250 273 3* 85 107 128 150 171 200 228 257 285 313 342 Horse-power Transmitted \ry Cold-rolled Iron Shafting-. AS PRIME MOVER OR HEAD SHAFT WELL SUPPORTED BY BEARINGS. g Revolutions pei Minute. s 60 80 100 125 150 175 200 225 250 275 300 Ins. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H 2.7 3.6 4.5 5.6 6.7 7.9 9.0 10 11 12 13 if 4.3 5.6 7.1 8.9 10.6 12.4 14.2 16 18 19 21 2 6.4 8.5 10.7 13 16 19 21 24 26 29 32 1 9 12 15 19 23 26 30 34 38 42 46 12 17 21 26 31 36 41 47 52 57 62 2f 16 22 27 35 41 48 55 62 70 76 82 3 21 29 36 45 54 63 72 81 90 98 108 3 27 36 45 57 68 80 91 103 114 126 136 34 45 57 71 86 100 114 129 142 157 172 3| 42 56 70 87 105 123 140 158 174 193 2i0 4 51 69 85 106 128 149 170 192 212 244 256 ^ 73 97 121 151 182 212 243 273 302 333 364 SHAFTIXG. 949 LINE-SHAFTING, BEARINGS 8 FT. APART. Kevolutions per Minute. & s 100 125 150 175 200 225 250 275 300 325 350 Ins. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. n 6.7 8.4 10.1 11.8 13.5 15.2 16.8 18.5 20.2 21.9 23.6 n 8.6 10.7 12.8 15 17.1 19.3 21.5 23.6 25.7 28.9 31 if 10.7 13.4 16 18.7 21.5 24.2 26.8 29.5 32.1 34.8 39 n 13.2 16.5 19.7 23 26.4 29.6 32.9 36.2 39.5 42.8 46 16 20 24 28 32 36 40 44 48 52 56 ^ 19 24 29 33 38 43 48 52 57 62 67 a 22 28 34 39 45 50 56 61 68 74 80 27 33 40 47 53 60 67 73 80 86 94 2h 31 39 47 54 62 69 78 86 93 101 109 2f 41 52 62 73 83 93 104 114 125 135 145 3 54 67 81 94 108 121 134 148 162 175 189 3k 68 86 103 120 137 154 172 188 205 222 240 '3k 85 107 128 150 171 192 214 235 257 278 300 POWER TRANSMISSION AND SHORT COUNTERS. s Kevolutions pei Minute. S 100 125 150 175 200 233 267 300 333 367 400 Ins. H.P. H.P. H H.P. H.P. H.P. H.P. H.P. H.P. H.P. H.P. H 6.5 8.1 9.7 11.3 13 15.2 17.4 19.5 21.7 23.9 26 1* 8.5 10.7 12.8 15 17 19.8 22.7 25.5 28.4 31 34 1* 11.2 14 16.8 19.6 22.5 26 30 33 37 41 45 1* 14.2 17.7 21.2 24.8 28.4 33 38 42 47 52 57 lif 18 22 27 31 35 41 47 53 59 65 71 H 22 27 33 38 44 51 58 65 72 79 87 2 26 33 40 46 53 62 71 80 88 97 106 2ft 32 40 47 55 63 73 84 95 105 116 127 2* 38 47 57 66 76 89 101 114 127 139 152 n 44 55 66 77 88 103 118 133 148 163 178 ih 52 65 78 91 104 121 138 155 172 190 207 2^ 69 84 99 113 138 161 184 207 231 254 277 3 90 112 135 157 180 210 240 270 300 330 360 Hollow Shafts. Let d be the diameter of a solid shaft, and dxd2 the external and internal diameters of a hollow shaft of the same material. Then the shafts will be of equal torsional strength when d3 = 1 - • A 10-inch hollow shaft with internal diameter of 4 inches will weigh 16% less than a solid 10-inch shaft, but its strength will be only 2.56 % less. If the hole were increased to 5 inches diameter the Aveight would be 25 % less than that of the solid shaft, and the strength 4.25 % less. Table for JLaying- Out Shafting-. The table on the following page is used by ¥m. Sellers & Co. for the lay- ing out of shafting. 950 SHAFTING, PULLEYS, BELTING, ETC. I Si t »t t 2a M •saqoiq jejaureia W "*< i5 iO SO t-t-aO©5©*-C-)c5-*i~i>C:a5 •saqouj S^ffS^Ojjgjw^aoebsSjgg^g' •ant 'xog jo Sui -jrea'g jo q^3u9T; ~u ^^ fr A Form. 1 Form. 2 Form. 3 Form. 5 Double. &" single ."£ -r: H.P. = wv 550' H.P. — wv lioo H.P. = wv 1000 H.P. = wv 733' Belt H.P. — wv Belt. ® © J) Laced. Riveted r*fc.4 3.4 4.7 6.2 7.6 9.7 11 13 16 18 25 2000 2.7 4..°. 6.1 8.2 11 13 17 20 24 28 33 44 2500 4.3 6.7 9.6 13 17 21 27 32 38 45 52 69 3000 6.2 9.7 13 18 24 30 39 45 55 64 74 100 3500 X.4 13 19 25 r34 42 53 63 75 89 102 136 4000 11 17 24 33 44 54 69 82 98 116 133 177 4500 14 22 31 42 55 69 87 103 125 146 168 223 5000 17 27 39 52 69 86 109 129 156 183 210 275 5500 21 33 47 63 83 104 132 156 189 221 254 332 6000 24 39 56 75 99 125 157 188 225 257 303 396 6500 39 45 65 88 116 145 183 217 261 307 353 462 Horse-Power of Manila. Ropes. £gs Nominal Diameter of Rope in Inches. £ o£i ft f 1 1 1 9.08 1ft U If H If 23.8 If 27.5 2 2000 2.25 3.51 5.14 6.84 11.5 14.0 17.0 20.3 36.1 2100 2.35 3.67 5.27 7.15 9.40 11.8 14.7 17.8 21.1 24.8 28.8 37.6 2200 2.45 3.82 5.48 7.45 9.80 12.3 15.3 18.5 22.0 25.9 30.0 39.2 2300 2.55 3.98 5.71 7.75 10.2 12.8 15.9 19.3 22.9 26.9 31.2 40.8 2400 2.62 4.10 5.89 7.98 10.5 13.2 16.4 19.8 23.6 27.7 32.2 42.0 2500 2.70 4.21 6.05 8.21 10.8 13.6 16.8 20.4 24.3 28.5 33.1 43.2 2600 2.78 4.33 6.21 8.43 11.1 14.0 17.3 21.0 25.0 29.3 34.0 44.4 2700 2.85 4.45 6.39 8.67 11.4 14.4 17.8 21.5 25.6 30.5 35.0 45.6 2800 2.94 4.59 6.59 8.93 11.75 14.8 18.3 22.2 26.4 31.0 36.0 47.0 2900 3.00 4.68 6.73 9.13 12.0 15.1 18.7 22.7 27.0 31.6 36.8 48.0 3000 3.06 4.78 6.87 9.32 12.3 15.4 19.1 23.2 27.6 32.3 37.6 49.1 3100 3.12 4.87 7.01 9.50 12.5 15.7 19.5 23.6 28.2 33.0 38.3 50.0 3200 3.18 4.97 7.14 9.70 12.7 16.0 19.9 24.0 28.7 33.7 39.0 51.0 3300 3.25 5.07 7.27 9.89 13.0 16.3 20.3 24.5 29.2 34.3 39.8 52.0 3400 3.30 5.15 7.39 10.0 13.2 16.6 20.6 25.0 29.7 34.8 40.4 52.8 3500 3.35 5.22 7.50 10.2 13.4 16.9 20.9 25.3 30.1 35.4 41.0 53.6 3600 3.40 5.30 7.61 10.3 13.6 17.1 21.2 25.7 30.6 35.9 41.6 54.4 3700 3.44 5.36 7.70 10.4 13.7 17.3 21.5 26.0 30.0 36.3 42.1 55.0 3800 3.46 5.40 7.76 10.5 13.8 17.4 21.6 26.2 31.1 36.6 42.4 55.4 3900 3.49 5.45 7.81 10.6 13.9 17.6 21.8 26.4 31.4 36.9 42.7 55.8 4000 3.51 5.49 7.86 10.6 14.0 17.7 21.9 26.5 31.6 37.1 43.0 56.1 4100 3.53 5.52 7.92 10.7 14.1 17.8 22.0 26.7 31.8 37.3 43.2 56.4 4200 3.55 5.54 7.95 10.8 14.2 17.9 22.1 26.8 31.9 37.5 43.4 56.8 4300 3.56 5.55 7.98 10.8 14.2 17.9 21.2 26.9 32.0 37.6 43.6 56.9 4400 3.57 5.56 7.99 10.8 14.2 18.0 22.2 27.0 32.1 37.6 43.6 57.0 4500 3.56 5.55 7.96 10.8 14.2 17.9 22.2 26.9 32.0 37.6 43.5 56.9 4600 3.55 5.54 7-95 10.8 14.2 17.9 22.1 26.S 31.9 37.5 43.4 56.8 4700 3.53 5.50 7.90 10.7 14.1 17.8 22.0 26.6 31.7 37.2 43.1 56.4 4800 3.51 5.48 7.86 10.7 14.0 17.7 21.9 26.5 31.6 37.1 43.0 56.2 4900 3.49 5.45 7.81 10.6 13.9 17.6 21.8 26.4 31.4 36.9 42.7 55.8 5000 3.45 5.38 7.73 10.5 13.8 17.4 21.5 26.1 31.0 36.4 42.2 55.2 5100 3.43 5.35 7.67 10.4 13.7 17.2 21.3 25.9 30.8 36.2 41.9 54.8 5200 3.38 5.26 7.56 10.2 13.5 17.0 21.0 25.5 30.4 35.6 41.3 54.0 5300 3.34 5.20 7.47 10.1 13.3 16.8 20.8 25.2 30.0 35.2 40.8 53.4 5400 3.28 5.11 7.34 9.95 13.1 16.5 20.4 24.8 29.4 34.6 40.1 52.5 5500 3.21 5.00 7.20 9.75 12.8 16.2 20.0 24.2 28.9 33.9 39.3 51.4 6000 2.78 4.33 6.21 8.43 11.1 14.0 17.3 21.0 25.0 29.3 34.0 44.4 6500 2.17 3.38 4.85 6.60 8.6 10.9 13.5 16.4 19.5 22.9 26.5 34.7 956 SHAFTING, PULLEYS, BELTING, ETC. HORSE POWER 1* ^^ 1 1 1 1 1 1 1 1 ROPE DRIVING HORSE POWER OF MANILLA ROPE AT VARIOUS SPEEDS Vj^|T \ V \^ "^ \ \ \ X ^ v v ^t 3 S V \ V- V V I 5 S . > ■A_ . \ r \ \_ V V . 5 \. 4 4 \ \ - 4 \ 1 i t I r 4 -4 J- i-j 7- --/ z / / _i f~ / y T / 7 y^ Z/ / ^ //_ ^ ^ // /^/^ J& - p> Mit^OJCOOKftO g§§£ CO CO rfx ROPE DRIVING. 957 Horse«Power of " Stevedore " Transmission Rope at Various Speeds. In this table the effect of the centrifugal force has been taken into con- sideration, and the strain on the fibers of the rope is the same at all speeds when transmitting the horse-power given in the table. When more than one rope is used, multiply the tabular number by tbe number of the ropes. At a speed of 8.400 per minute the centrifugal force absorbs all the allowable tension tbe rope should bear, and no power will be transmitted. Table of the Horse-Power of Transmission Rope. (Hunt's Formula.) o Speed of the Rope in Feet per Minute. a ftm- ft 1.500 2.000 2.500 3.000 3.500 4.000 4.500 5.000 6.000 7.000 8.400 ^3 i 1.45 1.9 2.3 2.7 3. 3.2 3.4 3.4 3.1 2.2 .0 .20 1 2.3 3.2 3.6 4.2 4.6 5.0 5.3 5.3 4.9 3.4 .0 .25 1 3.3 4.3 5.2 5.8 6.7 7.2 7.7 7.7 7.1 4.9 .0 .30 1 4.5 5.9 7.0 8.2 9.1 9.8 10.8 10.7 9.3 6.9 .0 .36 1 5.8 7.7 9.2 10.7 11.9 12.8 13.6 13.7 12.5 8.8 .0 .42 u 9.2 12.1 14.3 16.8 18.6 20.0 21.2 21.4 19.5 13.8 .0 .54 li 13.1 17.4 20.7 23.1 26.8 28.8 30.6 30.8 28.2 19.8 .0 .60 n 18. 23.7 28.2 32.8 36.4 39.2 41.5 41.8 37.4 27.6 .0 .72 2 23.2 30.8 36.8 42.8 47.6 51.2 54.4 54.8 50. 35.2 .0 .84 For a temporary installation when the rope is not to be long in use, it might be advisable to increase the work to double that given in the tables. Slip of Ropes and Relts. (W. W. Christie.) Some French trials, with constant resistance, the power expended and slip in several modes of transmission was as follows : Ropes, 158.54 gross h.p., Slip, 0.33 per cent. Cotton belt, 159.67 " " 0.78 " Leather " 158.84 " " 0.96 " " " 160.23 " " 0.78 " Stated in percentage value, the results were : Ropes, 100.00 gross power, Slip, 0.100. Cotton belt, 100.87 " " 0.237. Leather " 100.37 " " 0.292. " 101.07 " " 0.237. 958 SHAFTING, PULLEYS, BELTING, ETC. Manila Cordage. Tarred Hemp. Size, Cir- Size, Weight of Feet in Breaking Strain Weight of cumfer'ce. Diameter. 100 one of New Ropes. 100 Inches. Inches. Fathoms. Pound. Pounds. Fathoms. For Hopes in use 11 s 31 20 deduct J from 40 n 44 14 these figures, for 55 if rs 60 10 chafing, etc. 75 2 | 79 7i 3000 100 1 fa 99 6" 4000 125 122 5 5000 155 2| ¥ 146 4 6000 190 3 176 3| 7000 225 3i Its 207 3 8500 265 3i 240 2* 9500 300 3| li 275 2^ 11000 355 4 )f 305 2 12500 405 a 355 1X3 14000 455 395 n 16000 500 5' If 490 n 20000 630 5*. If 595 1 24000 750 6" 2 705 10 in. 27000 910 6| 2J 825 8* 31500 1050 7 2* 960 7* 37000 1235 7* 2* 1100 6| 42500 1400 8" 2| 1255 5* 4850o 1600 8* 2! 1415 5 54500 1820 9 3 1585 4* 61500 2050 Hawser laid will weigh J less. Hotes on the Uses of Wire Rope. (Roebling.) Two kinds of wire rope are manufactured. The most pliable variety con- tains 19 wires in the strand, and is generally used for hoisting and running rope. For safe working load allow | or \ of the ultimate strength, according to speed, so as to get good wear from the rope. Wire rope is as pliable as new hemp rope of the same strength ; but the greater the diameter of the sheaves the longer wire rope will last. Experience has proved that the wear increases with the speed. It is, therefore, better to increase the load than the speed. Wire rope must not be coiled or uncoiled like hemp or manila — all untwisting or kinking must be avoided. In no case should galvanized rope be used for running. One day's use scrapes off the zinc coating. Tal»le of Strains Produced Uy lioads on Inclined Planes. Elevation in 100 Ft. Strain in Lbs. on Rope from a Load Elevation in 100 Ft. Strain in Lbs. on Rope from a Load of 1 Ton. of 1 Ton. Ft. Deg. Ft. Deg. 10= 5% 212 90 = 42 1347 20= 1H 30 = 16f 404 100 = 45 1419 586 110 = 47| 1487 40 = 21| 754 120 = 50J 1544 50 = 26} 905 130 = 52* 1592 60 = 31 1040 140 = 54* 1633 70 = 35 1156 150 = 56J 1671 80 = 38| 1260 160 = 58 1703 ROPE DRIVING. 959 Table of Transmission of Power l»y Wire Hopes. Showing necessary size and speed of wheels and rope to obtain any de- sired amount of power. (Roebling.) Diam. Diam Diam. Diam. of No. of Rev- Horse- of No. of Rev- of Horse- Wheel in Ft. olutions. Rope. Power. Wheel in Ft. olutions. Rope. Power. 4 80 a 3.3 10 80 ii 58.4 100 f 4.1 100 it 73. 120 5. 120 87.6 140 1 5.8 140 16 102.2 5 80 T75 6.9 11 80 11 75.5 100 8.6 100 ii 94.4 120 10.3 120 113.3 140 17S 12.1 140 is 132.1 6 80 * 10.7 12 80 I 99.3 100 i 13.4 100 I 124.1 120 h 16.1 120 148.9 140 18.7 140 1 173.7 7 80 16.9 13 80 | 122.6 100 21.1 100 153.2 120 1% 25.3 120 1 183.9 8 80 f 22. 14 80 | 148. 100 s 27.5 100 i 185. 120 I 33. 120 8 222. 9 80 f 41.5 15 80 f 217. 100 5. 51.9 100 T 259. 120 1 62.2 120 i 300. Hoisting- Ropes (lO Wires to the Strand). (Trenton Iron Company's List.) Iron. Crucible Steel. g _5 S3 £ a ■SMS II i*s 0> ® a, ° fl §Oo2 3 ao 02 O ^ o o §*g>§.3 Circumfei of Hemp of Equal Strength Circumfei of Hemp of Equal Strength 1 -5 7 8. 74 15 151 8 164.69 32.9 9 2 2 «i 6.3 65 13 14* 7 132.37 26 5 8 3 It 5+ 5.25 54 11 13 6h 108.13 21.63 n 4 1* 5 4.1 44 9 12 5 97.17 19.44 6 5 H 4* 3.65 39 8 iii 4f 86.38 17.3 16i 5* b* IS 4f 3. 33 6.5 1 4* 61.00 12.2 15 5 6 U 4 2.5 27 5.5 4 50.17 10. 12i 5 7 U 31 2. 20 4 8 3* 38.00 7.7 11 U 8 1 »* 1.58 16 3 7 3 29.2 5.8 9 4 9 f 2§ 1.2 11.5 2.5 6 2^ 21.55 4. 8 3| 3* 10 2f .88 8.64 1.75 5 2+ 14.99 3. 6k m & 2 .7 5.13 1.25 4* 2 12.53 2.5 5| 3 \% IB 1* .44 4.27 .75 4 1* 8.81 1.75 5i 4| m 10f | 1* .35 3.48 .5 3J 7.52 1.5 2 960 SHAFTING. PULLEYS, BELTING, ETC. The drums and sheaves should be made as large as possible. The mini- mum size of drum is given in a column in table. It is better to increase the load than the speed. Wire rope is manufactured either with a wire or a hemp center. The latter is more pliable than the former, and will wear better where there is short bending. The weight of rope with wire center is about 10 per cent more than with hemp center. Power Transmission and Stranding* Ropes (9 Wires to the Strand). (Trenton Iron Company's List.) Iron. Crucible Steel. a m =4H A 00 d ig^gf 0,-, °'g'2C' 1 «m § -3 « 5 p 4» +S g § j2 -=j MISCELLANEOUS TABLES. 965 Metrical Measures Equivalent to English Measures. Meters. Inches. Feet. lm/m 0.039 0.0033 2 0.079 0.0066 3 0.118 0.0098 4 0.157 0.0131 5 0.197 0.0164 6 0.236 0.0197 7 0.276 0.0230 8 0.315 0.0262 9 0.354 0.0295 10»/m = le/m 0.394 0.033 2 0.787 0.066 3 1.181 0.098 4 1.575 0.131 5 1.969 0.164 6 2.362 0.197 7 2.756 0.230 8 3.150 0.262 9 3.543 0.295 10c/ m = .lm 3.937 0.328 .2 7.874 0.656 •3 11.811 0.984 .4 15.748 1.312 .5 19.685 1.640 .6 23.622 1.969 .7 27.560 2.297 .8 31.497 2.625 .9 35.434 2.953 lm0 39.371 3.281 Table for the Conversion of Mils. (l-lOOO Inches) into Centimeters. Centi- Centi- C enti- Centi- Mils. meters. Mils. meters. Mils. m eters. Mils. meters. 1 .00254 18 .04571 35 08888 52 .1321 2 .00508 19 .•04825 36 09142 53 .1346 3 .00762 20 .05079 37 09396 54 .1372 4 .01016 21 .05333 38 09650 55 .1397 5 .01270 22 .05587 39 09904 56 .1422 6 .01524 23 .05841 40 1016 57 .1448 7 .01778 24 .06095 41 1041 58 .1473 8 .02032 25 .06348 42 1067 59 .1499 9 .02286 26 .06602 43 1092 60 .1524 10 .02540 27 .06856 44 1118 61 .1549 11 .02793 28 .07110 45 1143 62 .1575 12 .03047 29 .07364 46 1168 63 .16f0 13 .03301 30 .07618 47 1194 64 .1626 14 .03555 31 .07872 48 1219 65 .1651 15 .03809 32 .•08126 49 1245 66 .1676 16 .04063 33 .08380 50 1270 67 .1702 17 .04317 34 .08634 51 1295 68 .1727 966 MISCELLANEOUS TABLES. Table for the Conversion of Mils. — Continued. Centi- Centi- Centi- Centi- Mils. meters. Mils. meters. Mils. meters. Mils. meters. 69 .1752 77 .1956 85 .2159 93 .-2362 70 .1778 78 .1981 86 .2184 94 .2387 71 .1803 79 .2006 87 .2209 95 .2413 72 .1829 80 .2032 8S .2235 96 .2438 73 .1854 81 .2057 89 .2260 97 .2465 74 .1879 82 .2083 90 .2286 98 .2489 75 .1905 83 .2108 91 .2311 99 .2514 76 .1930 84 .2133 92 .2336 100 .2540 English Measures Equivalent to Metrical Measures. CO w . .2 « 3 ■g £

ressure, and inversely as the diam- eter of the rolling bodies, where the cylinders or balls are of the same substances, and are pulled or pushed, as in a car or wagon. Where the load is propelled, by a crank fixed on the axle, the law it reversed. TE9IPEIIATVRE, or UTTSKSIKir OF HEAT. Standard Points — Fahrenheit. Centigrade. Reaumur, Boiling point of water under ) _ oioo inno Qno one atmosphere . . . .) ~ AVl 1UU m Melting point of ice . . . . 32° 0° 0° er of Degrees Cent. = Jf umber of Degrees fall. — (Continued.) Tenths of a Degree — Centigrade Scale Degrees Cent. .0 .1 .2 3 .4 .5 6 .7 .8 .9 Fab. Fab. Fah. Fah. Fah. Fah. Fah. Fah. Fah. Fah. 4 7.20 7.38 7.56 .7.74 7.92 8.10 8.28 8.46 8.64 8.82 5 9.00 9.18 9.36 9.54 9.72 9.90 10.08 10.26 10.44 10.62 6 10 80 10.98 11.16 11.34 11.52 11.70 11.88 12.06 12.24 12.42 7 12.G0 12.78 12.96 13.14 13.32 13.50 13.68 13.86 14.04 14.22 8 14.40 14.58 14.76 14.94 15.12 15.30 15.48 15.66 15.84 16.02 9 16.20 16.38 16.56 16.74 16.92 17.10 17.28 17.46 17.64 17.82 lYumber of Degrees Fall. _: ^imil^r of Degrees Cent. Tenths of a Degree - - Fahrenheit Scale Degrees Fah. .0 .1 .2 .3 .4 .5 .6 .7 .8 .9 Cent. Cent. Cent. Cent. Cent. Cent. Cent. Cent. Cent. Cent. 0 0.00 0.06 0.11 0.17 0.22 0.28 0.33 0.39 0.44 0.50 1 0.56 0.61 0.67 0.72 0.78 0.83 0.89 0.94 1.00 1.06 2 1.11 1.17 1.22 1.28 1.33 1.39 1.44 1.50 1.56 1.61 3 1.67 1.72 1.78 1.83 1.89 1.94 2.00 2.06 2.11 2.17 4 2.22 2.28 2.33 2.39 2.44 2.50 2.56 2.61 2.67 2.72 5 2.78 2.83 2.89 2.94 3.00 3.06 3.11 3.17 3.22 3.28 6 3.33 3.39 3.44 3.50 3.56 3.61 3.67 3.72 3.78 3.83 7 3.89 3.94 4.00 4.06 4.11 4.17 4.22 4.28 4.33 4.39 8 4.44 4.50 4.56 4.61 4.67 4.72 4.78 4.83 4.89 4.94 9 5.00 5.06 5.11 5.17 5.22 5.28 5.33 5.39 5.44 5.50 Coefficients of Expansion at Ordinary Temperatures. (Solids.) Material. Coefficient of Expansion. • °F. °C. .0000114 .0000104 .00000306 .0000100 .0000055 .0000078 .00000961 .00000399 .00000521 .00000841 .0000046 .00000587 .00000677 .0000206 .0000187 Brick Cement and ) from .000010 Concrete j • " to .000014 .0000173 .00000719 .00000938 .0000151 .0000083 .0000106 .0000122 Glass .... Gold from ' * to Tron, cast Iron, wrought MISCELLANEOUS TABLES. 971 Coefficients of ^Expansion — {Continued.) Material. Coefficient of Expansion. °F. °c. .0000158 .000004 .0000026 .0000049 .00000494 .0000020 .0000040 .0000067 .0000108 .0000056 .00000611 .00000689 .0000116 .00000276 .0000163 .00002S4 Marble (average) Masonry . . . Platinum . . . Porcelain . . . Sandstone . . . Silver .... from ' * to from • * to .0030017 .00300S8 .00000890 .0000036 .0000070 .000012 .0000194 .0000102 Steel, untempered Steel, tempered .0000110 .0000124 .0000209 .00000496 .0000293 HEAT. Specific Heat of Substances. The specific heat of a body at any temperature is the ratio of the quantity of heat required to raise the temperature of the body one degree to the quantity of heat required to raise an equal mass of water at or near to its temperature of maximum density (4°C. or 39.1°F.) through one degree. Specific Heats of Metals. (Tomlinson.) Metal. Specific Heat at 0°C. or 32°F. 50°C.orl22°F. 100°Cor212°F 0.2070 0.0901 0.0941 0.1060 0.0300 0.0320 0.0473 0.0547 0.0523 0.0901 0.2185 0.0923 0.0947 0.1130 0.0315 0.0326 0.0487 0.0569 0.0568 0.0938 0.2300 Copper German Silver Iron Lead 0.0966 0.0952 0.1200 0.0331 0.0333 Platinum Silver Silver 0.0501 0.0591 0.0595 Zinc 0.0976 Mean Specific Heat of Platinum. (Pouillet.) Between 0°C. (32°F.) and 100°C. " 300°C, (212°F.) (572°F.) , 500°C. (932°F.) 700°C. (12920F.) . 1000°C. (1832°F.) . 1200°C. (2192°F.) . 0.0335 0.0343 0.0352 0.0360 0.0373 0.0382 972 MISCELLANEOUS TABLES. Mean Specific Heat of Water. (Regnault.) Between 0°C. (32°F.) and 40°C. (104°F.) . . . . . 1.0013 " " " " 80°C. (176°r.) 1.0035 " " " " 120°C. (248°F.) 1.0067 " " " " 160°C. (320°F.) 1.0109 " " " " 200°C. (392°F.) . . . . 1.0100 " " " " 230°C. (446°F.) 1.0204 Mean Specific Heat of Glass (Kohlrausch) 0.19 Specific Heat of Gases and Vapors at Constant Pressure. Air Carbon monoxide Carbon dioxide . Hydrogen . . . Nitrogen . . . . Oxygen . . . . Steam . . . . „ Specific Heat for Equal. Volumes. "Weights, 0.2375 0.2370 0.2405 0.2989 0.2375 0.2450 0.1952 3.4090 0.2438 0.2175 0.4805 Regnault Regnault Wiedermann Regnault Regnault Regnault Total Heat of Steam. British Thermal "Unit : (B. T. U.) is the quantity of heat which will raise the temperature of one pound of water one degree Fah. at or near its temperature of maximum density 39.1°. French Calorie : is the quantity of heat that will raise the tempera- ture of one kilogramme of pure water 1°C. at or near4°C. Pound Calorie: is the quantity of heat that will raise the tempera- ture of one pound of water 1°C. 1 B. T. IT. = .252 Calories. 1 Calorie = 3.968 B. T. IT. 1 lb. Calorie = 2.2046 B. T. IT. 1 pound Calorie = § Calorie. The mechanical Equivalent of Heat. Joule gives Professor Rowland, 1 B. T. IT. = 1 B. T. U. : 772 ft. 778 ft. 1H.P, = 42.416 B. T. IT. (See Table of Energy Equivalents on p. 684.) MISCELLANEOUS TABLES. 973 Off)On«OHHHnWMMO uoq.iBO 8nl^A •3-0 sauojBQ •spunoj •*0 I J° -, -fiionoowN-fOL't-'fcionHHrii- i :. ; i ". o 'O m « l: -t o o o t ro -t i - :: c i ■; uo -*< x:i co co i- co m-c-triKc i_- o >: cc : ~ -rl-l-l- X i— i 0 HriCCH-jiMhKL-'XC CC C O O CO i- M rt 01 CO CO- 01 0 O -^ O © C ^> •Slip3.T£) •no I jo •SlUK.tf) vdvn I jo >0 CC X X CO X -f — 01 C-. HNCCC C-. CO CC — CO - l- = i- -f = c i- r. r. (M oc t x t- cc .oc?i- - i- = cc — CC -f L- — c-. CC I- r- co -t- -. \- '-i.t'-i-.-n-c x x x :i r. r. ~ c - ic v. _ Ol rH 01 CO X ICO L- •* (O C K CC ICO t- 01 01 -t C OdrirtHriricq'rlrilOriridddwrirlr •I = JJV AcJIA'B.I*) ogxoadg •g r= U9.§ -oap^H A^IAIU-E) 0ff'l09dg •noi^snq ■xaoQ jo sjotipojj •cjqSpAi -noopjj,\[ loqraAg -noaioj\[ + '-' re £f 2J. jp £ S -i r-i T-i ©' OO'rid . O i-i X' CC X CC CO CO CC t- CO X CC CC L- I- . ~ - C I- X X 00 CO • HOrtdddrtri t - :• 1 1- h r. o Ji » « -f r. i- ci ^ :* £ g§ -? Lr2 .3 « £ - ± £ y 7 •: : CO 01 ^ 1-1 0"l r^ ,- rt O © CO CO -f CO i*.0 d -f -r 10 rv 0 t~ ^h r • : "-' f: c r ""• - ~00C0t~COCli-HC5 o c: « r; c 1- c 10 ci 01 t~ 00 c c ■ o tc~ O t- - l.o 1: 1: x o t- l: j> v; •+ c: ^ c: uo 10 — x lOOioocjin-fcc^a ci o 1 oi -t t- ci x r. - t-XNOn-frt-| - r •_- co 10 cc x c qxocto.c:- ddddOr-Hi-Hi-Hr-^C- ^-1^ NNNNiH-fiHOOli-'HOlOlO^HON . . n * -t •* ci ft cm ci cc c: c cc rt -f : 1 ci r. cc cc. C5 cc cc cc cc, x c. x l- cc. cc. lo. co © ■* O © co " id t^ id t>^ cc! — id cc! t- t- co' id id 1— t-^ ■* 06 h co 1-1 01 01 01 01 -f- 10 CO L-ClCU^riHHHMCO'* W" w'wVkWWWWod0 £ £ czcz ::: Er 2? ~f 3! 5 ~ r- 2? ?' 90 co ^j ^3 rt _ oi 01 cc. -1 i id t- id 1- cc! — 10 cc! t- t^ cd 10 -^ •- 1— 01 01 01 0 1 -f LO CO t^ C . . .944 .0342 Gamboge .... 1.222 .0442 Mulberry . . . .897 .0324 Gunpowder, loose . .900 .0325 Oak, heart of, 60 old 1.170 .0423 " shaken 1.000 .0361 Orange-tree . . .705 .0255 " solid . '( 1.550 .0561 Pear-tree . . . .661 .0239 • i 1800 .0650 Pomegranate-tree 1.354 .0490 Gum Arabic . . . 1.452 .0525 Poplar .... .383 .0138 Indigo Lard ...... 1.009 .0365 "' white Spanish .529 .0191 .947 .0343 Plum-tree . . . .785 .0284 Mastic ..... 1.074 .0388 Quince-tree . . .705 .0255 Spermaceti . . . .943 .0341 Sassafras . . . .482 .0174 Sugar ..... 1.605 .0580 Spruce .... .500 .0181 Tallow, sheep . . .924 .0334 old . . . .460 .0166 " calf . . . .934 .0338 Pine, yellow . . .660 .0239 " ox . . . .923 .0334 " white . . .554 .0200 Atmospheric air . .0012 000043 Vine ..... 1.327 .0480 W'g't Walnut .... .671 .0243 Gases. Vapors. cu.ft. Yew, Dutch . . .788 .0285 gr'ns. " Spanish .807 .0292 Atmospheric air . .... 1.000 527.0 Iiiquids. Ammoniacal gas .... .500 263.7 Acid, Acetic . . 1.062 .0384 Carbonic acid ..... 1.527 805.3 " Nitric . . 1.217 .0440 Carbonic oxid ..... .972 512.7 " Sulphuric . 1.841 .0666 Carbureted hydrogen . . .972 512.7 " Muriatic . 1.200 .0434 Chlorine 2.500 1316 " Fluoric . . 1.500 .0542 Chlorocarbonous acid . . 3.472 1828 " Phosphoric 1.558 .0563 Chloroprussic acid . . . 2.152 1134 Alcohol, comraer, .833 .0301 Fluoboric acid 2 371 1250 " pure .792 .0287 Hydriodic acid ..... 4.346 2290 Ammoniac, liquid .897 .0324 Hydrogen . .069 36.33 Beer, lager . . . 1.034 .0374 Oxygen 1.104 581.8 Champagne . . .997 .0360 Sulphuretted hydrogen 1.777 9370 Cider ..... 1.018 .0361 Nitrogen .972 512.0 Ether, sulphuric .739 .0267 Vapor of alcohol .... 1.613 851.0 Naptha .... .848 " turpentine spirits 5.013 2642 Egg 1.090 .0394 " water .... .623 328.0 Honey .... 1-450 .0524 Smoke of bituminous coal .102 53.80 Human blood 1.054 .0381 " wood .90 474.0 Milk ..... 1.032 .0373 Steam at 212° ..... .488 257.3 MISCELLANEOUS TABLES. 975 TABLE Or SMJCIJFIC CJHAVITCY AXn UHflT weicwiiiw. Water at 39.1° Fahrenheit = 4° Centigrade ; 62.425 pounds to the cubic foot (authority, Kent, Haswell, and D. K. Clark). Specific Gravity. Authority Lbs. per Cubic Foot. Lbs. per Cubic Inch. Aluminum, pure cast " " rolled " " anne'ld " nickel alloy, cast " " " rolled " " " ann'ld Aluminum Bronze, 10% 5% Brass, cu. 67, zn. 33 cast " cu. 60, zn. 40 " Cobalt . . . Brass, plates . high yellow Bronze composition cu. 90, tin 10 Bronze composition cu. 84, tin 16 Lithium . • Potassium . Sodium o . Rubidium Calcium . . Magnesium . Caesium . . Boron . Glucinum Strontium . Barium . . Zirconium . Selenium „ „ Titanium . . Vanadium , Arsenic . . Columbium . Lanthanum . Niobium . . Didymium . Cerium . . Antimony » Chromium „ Zinc, cast . . " pure . " rolled . "Wolfram . . Tin, pure . . Indium . . Iron, cast " wrought " wire . . Steel, Bessemer " soft Iron, pure 2.85 2.76 2.74 7.70 8.405 8.50 8.832 0.57 0.87 0.97 1.52 1.57 1.74 1.88 2.00 2.07 2.54 3.75 4.15 4.50 5.30 5.50 5.67 6.00 6.20 6.27 6.80 6.861 7.15 7.191 7.119 7.29 7.42 7.218 7.70 7.774 7.852 7.854 7.86 Haswell. Thurston. R.-A. P. R. C. Thurston. Haswell. R.-A. Haswell. R.-A. Haswell. R.-A. Haswell. R.-A. Haswell. R.-A. Haswell. R.-A. Kent. Haswell. Kent. R.-A. 159.63 167.11 165.86 178.10 172.10 170.85 480.13 515.63 519.36 524.88 530.61 535.38 541.17 551.34 36.83 54.31 60.55 94.89 98.01 108.62 117.36 124.85 129.22 158.56 234.09 259.06 280.91 330.85 343.34 353.95 374.55 387.03 391.40 408.26 417.00 418.86 429.49 428.30 446.43 448.90 444.40 455.08 463.19 450.08 480.13 485.29 479.00 489.74 490.66 .0924 .0967 .1031 .0996 .3006 .3036 .3132 .'3191 ' .0213 .0314 .0350 .0549 .0567 .0629 .0679 .0723 .0748 .0918 .1355 .1499 .1626 .1915 .1987 .2048 .2168 .2240 .2265 .2363 .2413 .2424 .2457 .2479 .2583 .2598 .2572 .2634 .2681 .2605 .2779 .2808 .2837 .2834 .2840 976 MISCELLANEOUS TABLES. IA.BIE OF SPECIFIC «RA¥IT1 . — Continued. Specific Gravity. Authority. Lbs. per Cubic Foot. Lbs. per Cubic Inch. Kilos per Cubic Deem. Manganese .... 8.00 R.-A. 499.40 .2890 8.00 Cinnabar 8.809 Haswell. 505.52 .2925 8.098 Cadmium 8.60 R.-A. 536.85 .3107 8.60 Molybdenum . . . 8.60 " 536.85 .3107 8.60 Gun Bronze .... 8.750 Haswell. 546.22 .3161 8.750 Tobin Bronze . . . 8.379 A. C. Co. 523.06 .3021 8.379 Nickel 8.80 R.-A. 549.34 .3179 8.80 Copper, pure . . . Copperplates and sheet 8.82 " 550.59 .3186 8.82 8.93 A. of C. M. 556.83 .3222 8.93 Bismuth 9.80 R.-A. 611.76 .3540 9.80 Silver 10.53 657.33 .3805 10.53 Tantalum .... 10.80 674.19 .3902 10.80 Thorium 11.10 692.93 .4010 11.10 Lead 11.37 709.77 .4108 11.37 Palladium .... 11.50 717.88 .4154 11.50 Thalium 11.85 739.73 .4281 11.85 Rhodium 12.10 755.34 .4371 12.10 Ruthenium .... 12.26 765.33 .4429 12.26 Mercury 13.59 848.35 .4909 13.59 Uranium 18.70 1167.45 .6755 18.70 Tungsten Gold 19.10 1192.31 .6900 19.10 19.32 1206.05 .6979 19.32 Platinum 21.50 1342.13 .7767 21.50 Iridium 22.42 1399.57 .8099 22.42 Osmium ..... 22.48 1403.31 .8121 22.48 - R.-A. — Professor Roberts-Austen. Haswell — Haswell's Engineer's Pocket Book. P. R. C. — Pittsburg Reduction Co.'s tests. Kent — Kent's Mechanical Engineer's Pocket Book. Thurston — Report of Committee on Metallic Alloys of U. S. Board appointed to test iron, steel, and other metals. Thurston's Materials of Engineering. Riche — Quoted by Thurston. A. C. Co. — Ansonia Brass and Copper Co. A. of C. M. — Association of Copper Manufacturers. SPECIFIC GRAVITY AT ©2° FAHBEIHMT ahimh^im: auti* AirmwiM alloys. Aluminum Commercially Pure, Cast ........•••• ' Nickel Aluminum Alloy Ingots for rolling '• " Casting Alloy Special Casting Alloy, Cast c Aluminum Commercially Pure, as rolled, sheets and wire " " " Annealed . . Nickel Aluminum Alloy, as rolled, sheets and wire ........ " " " Sheets Annealed OF 2.56 2.72 2.85 3.00 2.68 2.66 2,76 2.74 Weig-Ht. Using these specific gravities, assuming water at 62 degrees Fahrenheit, and at Standard Barometric Height, as 62.355 lbs. per cubic foot (authority, Kent and D. K. Clark). . „„„„,,. Sheet of cast aluminum, 12 inches square and 1 inch thick, weighs 13.3024 lbs, Sheet of rolled aluminum, 12 inches square and 1 inch thick,weighs 13.9259 lbs, Bar of cast aluminum, 1 inch square and 12 inches long, weighs 1.1085 lbs, Bar of rolled aluminum, 1 inch square and 12 inches long, weighs 1.1605 lbs. Bar of aluminum, cast, 1 inch round and 12 inches long, weighs .870b lbs. Bar of rolled aluminum, 1 inch round and 12 inches long, weighs .9114 lbs. 1 1ST D EX. Aboard ship, condensation of steam, 904 Acceleration, horse-power of, 447 Accumulators, electric, 552 Aerial cable, specifications for, 171 lines, resistance of, 43 Air-pumps, 923 Air space in grates, 831 Aging of iron, 344 Alloys, relative resistance of, 181 Alternating current arc lamps, 394 current armature windings, 259 current circuits, measuring power in, 51 current conductors, 103 current dynamos, 230 current electro-magnets, 87 current motors, 273 current switchboards, 590 current wiring chart and table, 131 current wiring formula, 127 E.M.F. and current in terms of d. c, 288 wiring, 121 Alternators in parallel, 269 Aluminum conductors, 174-179 data on, 174 production of, 680 weight and specific gravity, 976 American woods, weight, coal value, 849, 850 Ammeters, 25 Ammunition hoists, electrically op- erated, 740 Amperes per car, 431 Angular velocity, 967 Annealing of armor plate, 693 Annunciator wiring, 138 Anthracite coal, properties of, 851 Anti-induction cables, 142 Arc circuits, insulation resistance of, 59 Arc lamps, continuous current, 393 lamps, alternating current, 394 lamps, candle-power of, 398 lamps, inclosed, 394 lamps, installation of, N.E.C., 770 lamps, regulation of, 395 switchboards, 582 Ardois's system of signaling, 735 Armatures, alternating current, 259 cores, 250 Armature cores, energy dissipation in, 80 reaction, 264 windings, 251 windings for converters, 291 Armatures, faults in, 329 Army, electricity in the, 711 Arresters, lightning and current, 653 Automatic telephone switches, 650 Axle speed, per car, 455 Balancing coils for arc lamps, 400 Balancing of three-phase lines, 118 Ballistic galvanometer, 24 galvanometer tests, 66 B. & S. gauge, law of, 203 Baths for plating, 678 Batteries, E.M.F. of, 53 internal resistance of, measure- ment of, 62 resistance of, 42 secondary, 552 Battery cells, arrangement of, 18 Battle order indicators, 750 Beams of uniform strength, 814 safe load on Southern pine, 822 special forms, coefficient of strength, 813 white pine, formula for, 821 Bell wiring, 137 Belt, length of, 953 length of, in a roll, 953 weight of, 953 Belting, horse-power of, 951 Bends, loss of head due to, 870 Bituminous coal, properties of, 851 Block signals, 432 Boiler flues, collapsing pressure, 915 settings, 836 settings, dimensions, 838 test report, 885, 886 tests, A. S. M. E. rules, 879 tubes, dimensions lap-welded, 914 Board of fire underwriters' rules, 762 of trade tramway regulations, 504- 508 Boat cranes, electrically operated, 742 Bonding, test of rail, 519-522 977 978 INDEX. Bonds, rail, 502 Centrifugal force, 967 Booster system, railway, 514 Characteristics of dynamos, 245 Boosters, continuous current, 285 Charging current per mile of cir- for storage batteries, 568 Boulenge chronograph, 715 Brackets for trolley poles, 441 cuit, 134 storage batteries, 582 Chemical action in storage batteries, Brake controllers, 484 553 Brakes, emergency, 465 Chimney construction, 841 Brass, composition of rolled, 825 height of, 844 table, 840, 841 Aveight of sheet and bar, 825 Bridging system, telephone, 664 thin shell brick, 842, 843 Brill cars, dimensions of, 466 Chimneys, draught power, 840 Brick chimneys, dimensions and dimensions and cost of iron cost, 845 (guyed), 846 Chloride of silver cell, 15 foundations, 794 work, 823 Choke coils, 605 Bricks, sizes, 823 Chronograph, 715 weight and bulk of, 824 Circuit breakers, 596 British thermal unit, 972 Circuits, tests of railway, 516-520 Brown's rail bond tester, 522 metallic telephone, 651 Bunsen photometer, 389 overhead, on poles, 651 Burglar-alarm wiring, 139 underground, telephone, 652 Burton electric forge, 693 Circulating pumps, 924 Bus excited dynamos, 588 Coals, heating value of, 850 Bushel, 961 proximate analyses, 852 space required to stow a ton, 853 Cable joints, 201 Coast-defense guns, manipulation testing, 220 Cables, data on, 158 of, 721 Codes, telegraphic, 642 underground, 652 Coefficient of inductance, measure- Calcium carbide, production of, 677 ment of, 48 Calorie, French, 972 of self-induction, definition of, 47 pound, 972 Coke, analysis of, 853 Calorimeter, Carpenter's, quality weight and bulk, S53 curves, 894 Collapsing pressure, boiler flues, 915 Carpenter's throttling, 890 Columns, Baker's formula, 803 diagram for throttling, 892 Gordon's formula, 802 separating, 893 Hodgkinson's formula, 802 thiottling, 889 hollow cast iron, strength, 807 Candle-power of arc lamps, 398 hollow cvlindrical, strength, 80S of incandescent lamps, 404 N. Y. City building laws, 803 Capacity effects on circuits, 105 pillars or struts, 802 measurement of, 46 solid cast iron, strength, 807 of conductors, 110 wrought iron, strength, 809 of cables, tests of, 223 Combustibles, table of, 848 Carbons, arc light, 396 Common-battery system, 658 Carrying capacity of copper wire, Commutating machines, A. I. E. E. 153 report, 295 capacity of wires, National Elec- Comparison of columns of water in trical Code, 768, 788 feet, 929 Car heating, electric, 499, 689 Compound cables, 229 heating, cost of, 690 engines, cylinder ratios, 919 lighting, electric, 547 wiring diagrams, 476-480 wiring, rules for, N. E. C, 775 Concrete foundations, 794 Condensation in steam pipes, 904 in steam pipes aboard ship, 904 Cars, dimensions of, 466-469 in heating pipes, 904 weight of, 470 Condenser, ejector, 923 Cary-Foster Bridge, 40 jet, 921 Cast iron, test, 796 surface, 922 Caustic soda, production of, 676 Condensers and pumps, 921 Cement, adhesion to bricks or rub- arrangement of electrical, 46, 223 i ble, 796 standard electrical, 28 and sand, 796 Condensing engines, number of ex- average strength of neat, 796 pansions for, 919 mortar, 795 Conducting system, calculation of, Centigrade in Fahr. equivalents, 969 510-514 Central stations, storage batteries in, Conductivity of cables, 228 560-576 of copper, 140 979 Conductivity, with millivoltmeter, measurement of, 62 Conductors for electrical distribu- tion, 97 for incandescent circuits, 101 properties of, 140 Conduit railway systems, 531-536 work, National Electric Code, 772 780 Contact plate system of General Electric Co., 543-546 Continuous current dynamos, 230 current motors, 270 Controllers, installation of, 475 dimensions of, 487 electric brake, 484 rheostatic, 483-485- series parallel, 481-486 Converter armature windings, 291 Converters, rotary, 286 Cooking, electric," 685 electric, cost of, 685 Copper bar data, 587 data, 140 electrolytic refining of, 680 Copper-plating, 67S Copper, temperature coefficient of pure, 185 weight of round bolt, 825 wire, bard-drawn, 142 wire table, 143 Avire table, National Electric Code, 769 Core losses, 72 loss, test for, 312 Cost of arc and incandescent lamps, 414 of operating mining plants, 696-700 Cowles' aluminum process, 680 Cross arms, dimensions of, 219 Crosses in cables, 225 Cubic measures, metrical equiva- lent, 964 Current consumption per car, 454 densities for various metals, 269 density in street railway conduc- tors, 445 measuring with voltmeter, 56 wave form of, 705 Curves, effort exerted on, 453 railway, 423 Cutouts, installation of, National Electric Code, 781 Cylinder ratios, compound engines, 919 Deck winches, electrically operated, 744 Deflection table, for wire spans, 209-218 Densities, average current, for vari- ous metals, 269 Depreciation on street railways, 498 Diagrams for car wiring, 476-480 Dielectrics, resistance of, 193 disruptive value of, 194-197 strength, A. I. E. E. report, 300 values of various (table), 197 Dimensions of railway cars, 466-469 Dip in span wire, 439 Direct current switchboards, 589 deflection method, 220 Discharge of water through an ori- fice, 935 Disruptive value of dielectrics, 194 Distribution and diffusion of light, 409 of electric energy, 92 of light by incandescent lamps, 412 Ditches, data for flumes and, 933 Double truck cars, power required by, 453 Draft power for combustion of fuels, 844 Draw-bar pull test, 522 Drop at end of railway line, test of, 519 in street railway conductors, 446 Dry batteries, 17 Ductility of boiler plate, 835 Duplex telegraphy, 639 telephony, 661 Dynamo and motor regulation, A. I. E. E. report, 301 Dynamos, alternating current, 230 continuous current, 230 efficiency test of, 319 rooms, N. E. C, 762 Dynamos and motors, efficiency of, 294 and motors, rating of, 303 and motor standards and testing, 293 and motors temperature rise in, 307 and motors, tests of, 306 E.M.F. of, 53 for U. S. Navy, 727 insulation of, measurement of, 60 method of exciting, 588 resistance of, 43 Dynamotors, 284 Economical distributing conditions, 93 Economizers, tests of, 874, 875 Economizers, 873 Eddy current factors, 79 current, loss in dynamo and mo- tor, 313 current loss curves, 78 currents in iron cores, 72 Edison-Lai ande cell, 16 Efficiency curves of dynamos, 247 of arc lamps, 399 of incandescent lamps, 402 test of dynamos, A.I.E.E. report, 319 test of motors, 325 test of railway motors, 523 Ejector condenser, 923 Elastic limit, 804 Electric brake controllers, 484 cooking, 685 lighting, 386 080 Electric power transmission, 99, 549 welding, 691 Electrical code, national, 762 measurements, 38 standardization, A. I. E. E. report, 293 units, 4 Electricity meters, 615 Elevated railway data, 471^74 Elevation of outer rail, 428 Electro-chemistry, application of, 676 dynamometer, 32 magnetic railway system, 536 Electrolysis, 675 of pipes, 524-529 Electrolytic refining of copper, 680 Electromagnetic units, 5 Electro-magnets, alternating-cur- rent, 87 depth of Avinding for, 87 heating of, 87 lifting power of, 83 M.M.F. of, 81 permament amp. for (table), 88 properties of, 81 relation between constants of, 86 Avinding of, 84 Electrometallurgy, 677 Electrometer method for measure- ment of E.M.F., 45 Electrometers, 30 Electromotive force of dynamos, 230 Electroplating, 677 Electrostatic units, 4 voltmeter, 31 Electrotyping, 677 Elements of the usual sections, 805, 806 Emergency brakes, 465 E.M.F., Avave form of, 705 measurement of, 45 of batteries, measurement of, 53 Energy and work, units of, 12 Engine telegraphs, U.S. navy, 750 Equation of steam pipes, 907 of steam pipes, table, 909 Equipment list for one car, 480 Exhaust injectors, 868 steam, pump, 872 Expansion, coefficients of, 970 of metals, 184 of Avater, 858 Factor of safety, 804 Factors of evaporation, 895 of evaporation, table, 896 Fahrenheit in centigrade equiva- lent, 970 Faults in incandescent lamps, 408 in Avires or cables, 226 of car motors and remedies, 523 Feeder points, location of, 512 Feeders, arrangement of railwav, 508-510 Feed-Avater heaters, 871 pipes, sizes of, 869 purification by boiling, 861 Feed-water, saAing by heating, 871 Field magnets, 265 telegraph and telephone, 726 Fire alarms, for U. S. navy, 753 Fire, temperature of, 849 Flanges, standard pipe, 915 Flat plates, safe pressure on, 834 rolled iron, Aveight of, 797 Flexure of beams, fundamental formulae, 810 FIoav of steam through pipes, 905 of Avater in pipes of various sizes, 869 of Avater over weirs, 937 Flumes and ditches, data for, 933 Flux densities, G6 Fly-Avheels and pulleys, centrifugal tension in, 925 Foot valve, 924 Forging electrically, 691 Foundations, 792 Friction, 967 loss in dynamos and motors, 312 of Avater in pipes, 870 Fuel, 846 economizers, 873 kinds and ingredients of, 846 Fuels, heat of combustion, 847 Fuller cell, 15 Furnaces for oil fuels, 855 Fuse data, 694 table, 204 Fuses, electric, for gun-firing, 722 for railway circuits, 465 installation of, N.E.C., 782 Fusion of metals, temperature de- termined by, 849 Gallon, 961 Galvanized iron Avire data, 154 Galvanometers, 20 resistance of, 42 Garton lightning arrester, 614 Gas lighting, electric, N.E.C., 786 light wiring, 139 passages and flue-area, 831 Gaseous fuels, 855 Gases, composition of, 855 and vapors, specific heat of, 972 General Electric single- phase alter- nators, 241 Electric surface contact raihvay, 543-547 Generator sets, tests of U. S. Navy, 728 German silver Avire, data on, 180 Gold-plating, 679 Grades and curves, 423, 428 horizontal effort on, 454 Grate surface, 831 surface per horse-poAver, 831 Gravity cell, 14 Greek 'letters, 967 Ground connections for lightning arresters, 607 connections, National Electric Code, 767 return drop, test of, 578 981 Guard wires, 445 Guns, manipulation of, 721 Gutta-percha, data on, 198 Guys for trolley wire, 444 Gyrostatic action on dynamos, 266 Hall's aluminum process, 680 Haulage in mines, cost of, 696 Headway of cars, 457 Heat conducting power of metals, 185 intensity of, 968 mechanical equivalent of, 972 of the electric arc, 400 transmitted through cast-iron plates, 911 units, 3, 683, 973 Heaters, feed-water, 871 electric, installation of, N.E.C., 771 Heating apparatus, efficiency of, 688 apparatus, portable, 779 apparatus, principles of, 683 cars by electricity, 499 cars electrically, 689 of armatures, 263 of bare conductors, 153 of electro-magnets, 87 pipes, condensation in, 904 surface of steam boilers, 830 surface per horse-power, 831 Helm angle indicators, 750 Hemp, tarred, weight of, 958 High potential circuit breakers, 597 potential oil switches, 595 potential systems, N. E. C, 775 volta°e transmission, 550 Hollow"shafts, 848 Hopkinson's efficiency test of dyna- mos, 321 Hopkinson's permeability test, 66 Horizontal effort of cars, 452 return tubular boiler, 829 tubular boiler height above grate, 831 Horse-power, brake, 918 boiler, to supply heating pipes, 904 indicated, 918 mill power, 928 nominal, 918 of a running stream, 928 of a waterfall, 927 of acceleration, 447 of steam boilers, 829 of traction, 449 of water, cubic feet table, 939 of water, miner's inch table, 939 per car, 450 water flowing in a pipe, 928 House circuits, resistance of, 43 Human body, resistance of, 61 Hydro-electro thermic system, 693 Hydrometers, 555 Hysteresis loss in transformers, 332 factors, 73 loss in dynamos and motors, 313 meter, 75 Hysteretic constants, 72 I-beam foundations, 795 I-beams, spacing and size, 817 Impedance coil, use of, 671 diagrams, 114 effect of, 104 table, 136 Impulse water-wheels, 944 Illuminating power, 393 Incandescent lamps, 402 lamps, candle-power of, 404 lamps, faults in, 408 lamps, life of, 411 lamps, proper use of, 403 Inches and eighths in decimal of a foot, 967 Inclined planes, strains in rope on, 958 Inclosed arc lamps, 394 Incrustation, causes and prevention of, 858 tabular view, 859 Inductance and impedance table, 136 Inductance factors, 107 of aerial lines, 50 Induction coils, connections of, 757 motors, 274 motors, current taken by, 125 motors, tests of, 324 motors, transformers for, 124 Inductive resistance of lines, 106 Injector vs. pump for feeding boil- ers, 868 Injectors, exhaust, 868 live steam, 866 live steam, deliveries for, 867 performance of, 868 Installation of street car motors, 474 of telephones, 653 Insulating joints, N.E.C., 784 Insulation of dynamos, measure- ment of, 60 of light and power circuits, meas- urement of, 58 of motors, measurement of, 61 regulations, National Electrical Code, 777 resistance, A.I.E.E. report, 300 resistance, N.E.C. 764 resistance of arc circuits, 59 resistance of cables, 220 resistance of circuits, 44 Insulators, specific resistance of, 193 Intensities of sources of light, 386 Intercommunicating telephone sys- tems, 668 Interior lighting, 393 telephone systems, 663 Internal resistance of batteries, measurement of, 62 International electrical units, 9 Iron, aging of, 344 and steel, 796 magnetic properties of, 64 plating, 679 982 INDEX. Iron, weight of, 796 Manila ropes, centrif ugal-tension,955 wire data, 154-157 ropes horse-power diagram, 956 Irons, electric, 691 ropes horse-power of, 955 Isolated plants, storage batteries ropes weight and strength, 958 for, 566 Marine boiler, 829 Avork rules, N.E.C., 787 Jet condenser, 921 Masonry, 823 Jointing gutta-percha covered wire, average ultimate crushing load, 824 199 Material for one mile of overhead Joints of cables, testing of, 222 line, 436 per mile of track, 429 Mean effective pressure, table of, 920 spherical candle-poAver, 399 Kapp efficiency test of dynamos, 315 Kelvin electric balance, 33 Measurement of E.M.F. of batteries, electric balance tables, 36 55 Krupp's resistance wires, 191 of capacity, 46 of E.M.F. , 45 Lamp specification for United States of flow of water, 936 navy, 732 of high resistances, 58 Leaded wires and cables, 166 of insulation of dynamos and mo- Leclanche cell, 15 tors, 60 Leonard's system of motor control, of internal resistance of batteries, 62 of low resistances, 57 Life of incandescent lamps, 411 Light, measurement of, 389 of mutual inductance. 49 proper use of, 411 of power in alternating current units of, 387 circuits, 51 velocity and intensity of, 386 of resistance of human body, 61 Lighting, electric, 386 of self-inductance, 48 of cars, 547 Measurements, electrical, 38 schedules, 414-422 Mechanical equivalent of heat, 972 -svstem specifications for U. S, stoking, 856 Navy, 731 Metallic circuits, requirements of, Lightning and current arresters, 653 651 arresters, function of, 601 Metals, resistance of, 141 arresters, location of, 764 specific heat of, 184 arresters, non-arcing, 602 weights and specific gravity, 975 conductors, 701 Meters, electricity, 615 rods, installation of, 702 alternating current, 620 rods, tests of, 704 Wright discount, 035 Lime mortar, 795 Metric measures in English meas- Lineal measure, metrical equiva- ures, 965 lent, 962 Metropolitan Street Railway system, Liquid fuels, 854 532-536 Load factor of railway system, 510 Miles per hour, feet per minute, Loading and training gear for guns, (Table), 457 739 Milliken repeater. 637 Long-distance lines, 660 Mill power, 928 transmission, 550 Miner's inch measurements. 937 Loop test of cables, 226 Mines, electrical land, 723 Lord Rayleigh's method for meas- Mining plants, operation of, 696 urement, E.M.F., 45 Miscellaneous materials, 825 Loss of charge method, 221 tables, 961 of head due to bends, 870 Modulus of elasticity, 804 of elasticity and elastic resis- Machine shops, horse-power in, 758 tance, 814 shops, men employed in, 758 Moisture in steam, determination tools, power used by, 758 of, 889 Magnet telephone, theory of, 645 in steam, tables, 891 Magnetic circuit, principle of, 82 Moment of inertia, 804 circuit of dynamos, 236 of inertia of compound shapes, 805 flux, formula for, 82 Monocyclic circuit connections, 126 properties of iron, 64 system wiring formula, 128 units, 4 Moonlight schedules, 414-422 Magnetization curves, 65 Mortars, cement and lime, 795 curves of dynamos, 245 Motor equipments, 425 Magneto generator, 650 -generators, 284 Manganine wire, 188 trucks, weight of, 464 983 Motors, alternating current, 273 continuous current, 270 installation of, N.E.C., 764 insulation resistance of, 61 railway, 424 rating of railway, 457 series-wound, 271 shunt-wound, 272 testing of railway, 522 tests of A. I. E. E. report, 322 types of railway, 460 weight of railway, 470 Multiphase induction motors, 274 Multiple-unit system, Spr ague, 489- 498 Multiplex telephony, 661 Mutual Inductance, meas. of, 49 inductance of aerial lines, 50 induction of circuits, 117 National Electrical Code, 762 Navy, electricity in United States, 727 standard wires, 159 Ness automatic telephone switch, 672 Niagara-Buffalo transmission line, 117 Niagara line construction, 117 Nickeline wire, 187-191 Nickel plating, 679 Non-arcing lightning arresters, 602 Northrup's galvanometer, 25 Ohm, standard, 27 values of, 141 Ohm's law, 38 Oil brake switch, 594 Output of dynamos and motors, 266 Overhead line construction data, 117, 436 railway system, 508-510 Overload capacity of dynamos and motors, A. I. E. E. Report, 303 Paper insulated wires and cables, 166 Party lines, 659 Paving, cost of, 430 Peckham trucks, 471 Permeability curves, 248 values, 66 Permeameter, 68 Permissible loads on foundation beds, 794 Petroleum,chemical composition of, 854 oils, chemical composition, 855 Photometers, 389 Piles, arrangement of, 794 safe load on, 793 Pipe flanges, standard, 915 riveted hydraulic, 934 standard dimensions, steam, gas and water, 912 wooden-stare, 933 wrought iron extra-strong, 913 Pipes, equation of steam, 907 sizes of feed-water, 869 Plate iron, weight of, 800 Plates, heat transmitted through, 911 Platinum, specific heat of, 971 Plotting of electrical waves, 708 Poles for trolley systems, 438 Polyphase induction motors, 275 Portland cement — recommenda- tions, 796 Position indicators, 749 Post-office bridge, 39 Power and induction factor table, 106 curves, 448 factor chart, 137 factor, determination of, 122 factor, formula for, A. I. E. E. Report, 305 measurement of, in alternating current circuits, 51 station, 424 station, capacity of, 459 station construction-chart, 791 stations, batteries in, 575 systems for U. S. Navy, 735 transmission, 548 Pressure loss in water pipes, 870 of water, table, 931 Primary cells, 14 Projectors, 395 for U. S. Navy, 733 Prony brake, 758 brake test of motors, 322 Properties of saturated steam-tables, 899 of timber, 818 Protected rail bonds, 501 Protection of steam heated sur- faces, 910 Pulleys, 951 centrifugal tension in fly-wheels and, 925 to find size of, 951 Pump, duplex-cylinder, direct-act- ing, 866 Pumping hot water, 863 Pumps, circulating, 924 condensers and, 921 efficiency of small direct-acting, 864 exhaust, 872 feed, 863 single-cylinder direct-acting, 865 Pure copper wire table, 151 Purification of feed water by boil- ing, 861 Quadruplex telegraphy, 640 Quality of steam by color of issuing jet, 895 Radiators, electric, 689 Radius of gyration, 805 Range indicators, 753 Rate of combustion due to chimney height, 844 984 Rates for incandescent lighting, 414 Hating dynamos and motors, A. I. E. E. report, 303 street railway motors, 457 Kail bonding, test of, 519-522 bonds, 502 bonds, protected) 501 Rail welding, 694 conductivity of, 504 Rails, sectional areas of, 504 weights of, 426 Railway circuits, tests of, 516-520 motor testing, 324, 522 motors, 424 motors, rating of, 457 motors, types of, 460 Railways, battery plants for, 575 conduit, 531-537 depreciation on, 498 surface contact, 536-546 switchboard connections, 591 third rail, 529-531 turnouts, 431 Reactance coils, 361 diagrams, 114 Reaction of armatures, 264 Receiver, Bell, 646 capacity, 921 telephone, 645 Rectifying mychines, A. I. E. E. report, 297 Regulation of dynamos and motors, A. I. E. E. report, 310 Regulations of board of trade, 504-508 Regulators, a. c. feeder, 362 Renewal of lamps, 407 Repeaters, 637 Repeating coil, 660 Report of A.I. E. E. committee on standardization, 293 on water-power property, 926 Resistance boxes, 27 boxes, location of, N. E. C, 763 increase in, 307 insulation, A. I. E. E. report, 300 internal of batteries, 62 measuring with voltmeter, 57 measurement of, 38 metals, 186 of dielectrics, 193 of human body, measurement of, 61 of metals, 141 of wires, 42 ribbon, 187 test of armature, 328 Return circuit, 499 drop, test of ,518 feeder booster, 515 Reverse current circuit breakers, 598 Revolution indicators, 750 Revolutions of car wheels, 451 Rheostatic controllers, 483-485 Riveted hydraulic pipe, 934 Riveted steel pipes, 932 Rope driving, 954 Rope, H. P. of transmission, 957 strains in, on inclined planes, 958 Ropes and belts, slip of, 957 Rotary converters, 286 Round and square wrought iron, weight of, 799 Rules for conducting boiler tests, 879 Safe carrying capacity of wires, 153 load on piles, 793 load on wooden beams, 820 Safety valves, calculations for lever, 877 valves, rules, 877 Sag of wires and cables, 205 Sand and cement, 796 recommendations, 796 Saturation test, A. I. E. E. report, 327 Scale-making materials, solubility of, 859 Schmidt chronograph, 718 Schultz chronograph, 717 Scotch boiler, 829 Searchlight data (table), 714 projectors, 395 Searchlights, 711 for TJ. S. Navy, 732 Sectional rail construction, 542 Self-induction, effect of, 104 Semaphores, 433 Separating calorimeter, 893 Separation of metals, 680 Separators, 875 tests of, 876 Series-parallel controller, 481-486 Series-wound motors, 271 Sewing-machines, power required by, 757 Shafting, centers of bearings, 947 cold rolled, horse-power of, 947 deflection of, 946 hollow, 949 horse-poAver of, 947, 949 power and size, 945 table for laying out, 949, 950 Shunt boxes, 26 Shunt-wound motors, 272 Signaling systems, National Elec- trical Code, 785 Signal lights for IT. S. navy, 734 Silicon bronze wire, 219 Silver, electrolytic refining of, 681 Simultaneous telegraphy and tele- phony, 662 Single-phase alternators, 241 Single-truck cars, power required by, 453 Size of conductors, calculation of, 99 Skin effect factors, 103 Slide-Avire bridge, 40 Slip of ropes and belts, 957 Smashing point of incandescent lamps, 403 Sockets, specifications for, N, E. C, 774, 782 985 Soldering fluid formula, 787 Solid rail bonds, 500 Spacing and size of I beams, 817 Span wire data, 440 wire dip, 439 Spans of wire and cable, table of, 205 Specific energy dissipation, 80 gravity and unit weights, 975 gravity, various substances, 974, 975. heat of gases and vapors, 972 heat of metals, 184, 971 heat of substances, defined, 971 heat of water, mean, 972 resistance table, 192 Speed and torque of motors, 271 of cars, 455 of water through pump-passages and valves, 864 recorder, 754 Spikes, 429 Sprague multiple unit system, 489- 498 Square and round bars of wrought iron, weight of, 799 Squier-Crehore Photo-Chronograph, 720 Standard cells, 11, 18 Standardization, report of A. I. E. E., 293 Static transformer, 331 Station equipment, 424 Stays, boiler head, 835 Steam, 829 and exhaust pipes, for cylinder sizes, 908 and gas pipes, standard sizes, 908 boiler braces, 836 boiler efficiency, 831 boilers, working pressure, 832, 833 boilers, types, 829 determination of the moisture in, 889 engines, classification, 916 engines, horse-power of, 918 engines, tests of various types, 917 flow of, through pipes, 906 moisture in, tables, 891 outflow of, to atmosphere, 905 pipes, 906 pipes, condensation in, 904 pipes, loss of heat from, 910 total heat of, 972 ports and passages, 921 properties of, 899 Stearns duplex, 641 Steel beams, formulae for greatest safe load, 812 Steel plate chimneys, 845 plate chimneys, brick lining, 845 plate chimneys, foundation dimen- sions, 845 weight of, 796 wire data, 154-157 Steering gear, electrically operated, 745 Stone foundations, 794 Storage batteries, 552 batteries, advantages of, 560 batteries, capacity of, 554 batteries, charging, 558 batteries, E. M. F., of, 554 batteries for surface contact rail- way, 546 batteries, in power stations, 575 batteries, installation of, 556-561 batteries, installation of, N. E. C, 765 • batteries, manufacturers of, 563 batteries, solutions for, 555 batteries, testing, 579 Strain and deflection table for wire spans, 215-218 Stranded Avire cables, 157 Stray field in dynamos, 237 Street car motors, installation of, 474 car wiring, 160 lighting by arc lamps, 401 railway batteries, 575 railway depreciation, 498 Strength of materials, 803 of riveted shell, 832 Struts, safe load for white pine, 827 Submarine cables, testing of, 228 cables, data on, 173 Substation system, 516 Suggestions, general, National Elec- trical Code, 790 Sulphate of copper, resistance of, 67C of zinc, resistance of, 676 Sulphuric acid, resistance of dilute 675 " Superior " wire, 188 Supplies for installing lamps, 760 Surface condenser, 922 contact railway system, 536-546 Suspension of trolley wire, 443 Switches, automatic telephone, 650 specifications for, N. E. C, 781 Switchboards, telephone, 655 specifications for the U. S. Navy, 729 layout of, 585 construction of, 585 for arc circuits, 592 location of, 763 Symbols, electrical engineering, 1 synopsis of (table), 6 Synchronizers, 267 Synchronous machines, A. I. E. E. lieport, 295, motors, 281 motors, tests of, 326 Tables of weights and measures, 961, 962 Tangent galvanometer, 21 Telegraph, anti-induction cables, 142 cables, specifications for, 170 codes, 642 for U. S. Army use, 724 wire data, 154 Telegraphy, American, 636 European, 636 98G Telephone, anti-induction cables, 142 cables, specifications for, 163 circuits, 651 switchboards, 655 systems, interior, 663 systems intercommunicating, 668 wire data, 154 wires, aluminum, 176 Telephones installation and main- tenance of, 653 for U. S Army use, 724 for IT. S. Navy, 753 Telephony, 645 duplex and multiplex, 661 Temperature coefficients of conduc- tors, 182 effect in wire spans, 207 or intensity of heat, 968 rise in dynamos and motors, 307 rise of, A. I. E. E. report, 298 Tensile strength of copper wire, (table), 208 Testing of cables, 220 of dynamos and motors, 293, 306 rail bonds, 519-522 railway motors, 5:22 Tests of American woods, 819 of street railway circuits, 516-520 Thermo-electric scale, 757 Thermometers, comparison of F. R., and C, 968 Third-rail systems, 529-531 Thompson-Ryan dynamo, 265 Thomson galvanometer, 22 double bridge, 41 method for measuring capacity, 46 Three-phase circuits, balancing of, 118 circuity energy in, 233 wiring formula, 130 Three-wire system, railway, 514 Throttling calorimeter, 889 calorimeter, calculation curves for, 892 Thunderstorms, safety during, 703 Ties, railway, 429 Time element for circuit breakers, 600 Tools for installing dynamos, 759 Torque and horse-power, 465 of motors, 271 Track bonding, test of, 519-522 laying, 430 return circuit, 499 Tractive coefficient, 458 effort, 458 force, 450 Transformer, air-blast, 338 cores, 331 design of, 335 heating tests (tables), 339 duties of, 332 efficiencies of, 340 equations, 334 expense of operating, 346 losses in (table), 332 regulation of, 344 Transformer, static, 331 Transformers, commercial (tables), 347 connections of, 366 constant current, 357 high potential, 356 in connection with converters, 292 testing of, 372 Transmission of electric power, 99 of power, 548 Transmitter, Edison carbon, 645 Blake, 647 solid-back, 648 Transverse strength of bars, 810 strength of beams, formulae for, 811 Trenton beams and channels, strength, 815 Trimming arc lamps, 402 Trolley poles, 437 systems, 508-520 wire, size of, 512 wire suspension, 443 Avires, specification, N. E. C, 766 Trucks, weight of, 464, 470 Tubes, sizes lap-welded boiler, 914 Tubular iron and steel poles, 438 Turbines, data, McCormick tvpe, 942 data, Pelton impulse, 943 dimensions, etc., 941-943 dimensions of Victor, 941 impulse, 941 installing, 941 parallel, outward and inward flow, 940 Turnouts on railways, 431 Turret-turning system, 737 Two-phase four- wire circuits, 120 Ultimate crushing load for masonry materials, 824 Underground cables, 652 electrical construction, 203 Units, electrical and mechanical, table of, 684 electrical engineering, 2 of light, 387 U. S. standard gauge for sheet and plate iron and steel, 801 Velocity _, angular, 967 Ventilating fans for U. S. Navy, 744 Vertical fire-tube boiler, 829 Voltage regulation for incandescent lamps, 405 Volt, determination of, 10 Voltmeter, tests with, 53 high resistance of, 54 Voltmeters, 25 Vulcanized india-rubber, 198 Walmsley's rail tester, 521 Ward Leonard turret-turning sys- tem, 737 Water analyses, table of, 862 calculations of horse-power, 939 Water column equivalents, 929 Water, cubic feet discharged, per minute, 935 expansion of, 858 flow of, over weirs, 937 flow of, through an orifice, 936 for boiler feed, 858 gas, 973 heat units per pound, 904 mean specific heat of, 972 table of pressure of, 931 theoretical velocity and discharge, tube boiler, 829 weight above 212° F., 857 weight of, per cubic foot, 904 weight per cubic foot, 856 wheels, 940 Water-power, 926 expense, yearly, 930 property, report, synopsis, 926 Water-tight door gear, United States navy, 747 Wattmeter price chart, 634 Wattmeters, 615 testing and calibrating of, 620 Westinghouse integrating, 625 reading of, 632 connections of, 617 Wave form of current and E. M. F., 705 meter, 706 Weatherproof insulation, N. E. C, 778 Weaver speed recorder, 754 Weber photometer, 391 Weights and measures, 961 metrical equivalent, 963 of cars, motors and trucks, 470 of copper and brass wire and plates, 826 of flat rolled iron, 797 of iron, 796 of motor trucks, 464 of plate iron, 800 of round bolt copper, 825 of sheet and bar brass, 825 987 Weights and measures of square and round bars, wrought iron, 799 Weight of steel, 796 and specific gravity of metals, 975 of rails, 426 Weir dam measurement, 937 table, 938 Weirs, Francis' formula?, 938 Welding by electricity, 691 Westinghouse electro-magnetic rail- way sytem, 537 Wheatstone bridge, 28, 38 White core wires and cables, 160-166 Winding of armatures, 251 of electro-magnets, 84 Wire rope, galvanized iron, 827 rope, notes on uses of, 958 rope, pliable hoisting, 828 rope, transmission by means of, 827 ropes, horse-power, etc., of, 959 table of A. I. E. E., 143-150 Wires, capacity of, N. E. C, 788 general rules for, National Elec- trical Code, 771 resistance of, 42 spaces occupied by (table), 91 Wiring formulae, 127 of cars, 475-480 interior, National Electrical Code, 768 specifications for U. S. navy, 730 specifications, N.E.C., 765 Wood as a fuel, 854 bulk, 853 properties of, 818 weight per cord, 854 Wooden beams, safe load, 820 stave pipe, 933 Woods, comparative resistance of, 219 test of American, 819 weights of various, 439 Wright discount meter, 635 Yachts, battery plants for, 571 ADVERTISEMENTS. WAGNER ELECTRIC MFG. CO., ^T- LOUIS, XX. DISCOUNT TO TRADE. WE MAKE IN ADDITION TO IT A TIME CLOCK WHICH AUTOMATICALLY THROWS CURRENT, AND ALSO TURNS IT OFF, AND A SIGN SWITCH WHICH OPERATES TO ALTERNATELY THROW ON. AND OFF THE CURRENT AT SHORT INTER- VALS OF TIME, SAY HALF MINUTE PERIODS 64 Cortlandt Street, C*3 SOKT, NEW YORK. The BallS Wood Co., BUILDERS OF p$^ram£pEED.EHI|IE£ FROM 50 TO 1000 H. P, FOR ELECTRICAL AND POWER PURPOSES. Office, 120 Liberty Street, Neiv York, Works, Elizabeth, New Jersey. It is justly considered a triumph of engineer- ing skill to build an engine for an Atlantic Liner which will operate for six days continu- ously under a practically uniform load at a pis- ton speed of eight miles per hour. A Ball and Wood Compound Condensing Engine, (cylinders 21 in. and 46^ in, x 24 in.), completed a continuous duty run at 3 P, M., August 3d, 1900, of 59 days, 22 hours, with an overload of 13 per cent, and at a piston speed of 600 feet per minute, equivalent to a distance of 9804 miles. This without a hot bearing or a moment's interruption. John D. Biggeet, Pres. and Treas. R S. Robb, Vice President. John P. Robinson, Sec. Pittsburgh Trolley Pole Co., KEYSTONE TROLLEY POLES Manufactured by Patent Process from Tubular Steel, continu- ously tapered to any size. Durable, Shapely and Strong, Sufficiently Elastic to Bend and Recover Perfect Shape and Strength. "KEYSTONE." TRADE MARK MANUFACTURERS OF KEYSTONE TROLLEY POLES PATENT TAPERED. Mechanical Engineers. Iron and Steel Workers. RIVER and INLAND, BLACKSMITH WORK, WATERS TUYERES, BLOW PIPES, WAGON SKEINS, WHIFFLETREES. Works : 1 15- 117 Water Street, \ Offices : Tradesmens Building, ) JL. D. Telephone 177. Pittsburgh, Pa, American Vitrified Conduit Company. Vitrified Salt Glazed Underground and Interior Conduits. Multiple Duct. Self Cen- tering. Single Duct. CONTRACTORS FOR Complete Installation of Conduit Systems. . General Office : 39-41 Cortlandt Street, New York. SHOW WINDOW LIGHTING. Frink's Special Patent Window Reflector. Lights by reflecting downward and inward, from lamps concealed in the top of the window, near the glass. The lamps are concealed from view. FRINK'S Jk SHADES, The Best Shades -JSRL AU sizes ancl styles of made. They reflect ^flRS^ Shades and Clusters. all the light there is. A Hg !5S^ Every Kind of Elec- Not the cheapest, but jffiwf ..trie Fixture, the most econom- ^mj, W Send for Catalogue. I. P. FRl^R, 551 Pearl St., Mew York. NOV 5 WXRr e i NOV 1 1901 H. C. ROBERTS ELECTRO SUPPLY CO. 831 Arch Street, Philadelphia, Pa. LINE MATERIAL AND CONSTRUCTION TOOLS. Electrical Appliances of Approved Manufacture always in Stock. CATALOGUE OF ELECTRICAL BOOKS, Eighty Pages, Alphabetically Arranged, Classified by Subjects and Authors, SENT GRATIS ON APPLICATION. D. Van Nostrand Co., PUBLISHERS, 23 Murray and 27 Warren Sts., NEW YORK.