UNIVERSITY Of CALIFORNIA 
 LOS ANGELES
 
 THE 
 
 ELECTRIC RAILWAY 
 
 IX 
 
 THEORY AND PRACTICE 
 
 OSCAR T. CROSBY, AND LOUIS BELL, PH.D. 
 
 SECOND EDITION, REVISED AND ENLARGED 
 
 NEW YORK 
 
 THE W. J. JOHNSTON COMPANY, LIMITED 
 
 41 PARK Row (TIMES BUILDING) 
 
 1893
 
 COPYRIGHT, 1892 AND 1893, BY 
 THE W. J. JOHNSTON COMPANY, LIMITED.
 
 TF 
 
 c 
 
 '/:<?. 3 
 
 PREFACE. 
 
 IN considering so widely ramifying a subject as electri- 
 cal traction two quite distinct methods of treatment pre- 
 sent themselves one the discussion of general principles, 
 methods, and results, the other the detailed study of motors 
 and their peculiarities, the specific parts of the general 
 equipment and their use in daily work. These two are 
 mutually exclusive in any volume of finite size. Believing 
 that the latter course would soon lead us into the mere 
 allotment of a graveyard for defunct apparatus, description 
 of which would be less interesting than a collection of epi- 
 taphs and would possess but casual value even to the moral- 
 ^.,1 ist, we have chosen the former line of treatment. 
 
 We have endeavored to present both the elementary theory 
 ** of the subject and the general features of the best practice, 
 <A\ describing in detail particular methods and forms of car 
 N^ machinery only in so far as they are of importance in illus- 
 r^trating the broad principles on which they depend. Specific 
 A instructions have been introduced, however, when, in the 
 present state of the art, they seemed necessary to a fuller 
 comprehension of the subject and a more thorough grasp of 
 modern methods. 
 
 We have not ventured to forecast the future of the trans- 
 mission of energy to moving motors, for nothing save 
 prophecy after the fact could be equal to the situation ; but 
 have contented ourselves with indicating certain paths of im- 
 provement that may lead to important results. 
 
 i 
 
 379414
 
 2 PREFACE. 
 
 In pursuing this policy we hope and believe that our sins 
 will be found to be those of omission rather than commission, 
 and that the reader will deal gently with them, as well-nigh 
 unavoidable in this first general discussion of a new and 
 important branch of applied science. 
 
 Our hearty thanks are due to the many friends who have 
 most courteously aided us in the preparation of this volume 
 by freely giving information of every sort, and particularly 
 to Prof. Elihu Thomson, Mr. Carl Hering, Mr. W. E. Baker, 
 Mr. H. I. Bettis, Mr. Thomas Pray, Jr., and Mr. Theo. Steb- 
 bins, for most valuable favors of such kind. 
 
 OSCAR T. CROSBY. 
 Louis BELL.
 
 TABLE OF CONTENTS. 
 
 PAGE 
 
 PREFACE . i 
 
 CHAPTER I. 
 GENERAL ELECTRICAL THEORY, ....... 5 
 
 CHAPTER II. 
 PRIME MOVERS, . . ... . . . . . .29 
 
 CHAPTER III. 
 MOTORS AND CAR EQUIPMENT, 64 
 
 CHAPTER IV. 
 THE LINE, 123 
 
 CHAPTER V. 
 TRACK CAR HOUSES SNOW MACHINES, 147 
 
 CHAPTER VI. 
 THE STATION, 161 
 
 CHAPTER VII. 
 THE EFFICIENCY OF ELECTRIC TRACTION 202 
 
 CHAPTER VIII. 
 STORAGE-BATTERY TRACTION 230 
 
 CHAPTER TX. 
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION, . . . 253 
 
 3
 
 4 TABLE OF CONTENTS. 
 
 CHAPTER X. 
 
 PAGE 
 
 HIGH-SPEED SERVICE, ..'... . 273 
 
 CHAPTER XL 
 COMMERCIAL CONSIDERATIONS, 39 
 
 CHAPTER XII. 
 HISTORICAL NOTES ...,.. 333 
 
 APPENDICES. 
 
 APPENDIX A. 
 ELECTRIC RAILWAY vs. TELEPHONE DECISIONS 353 
 
 APPENDIX B. 
 INSTRUCTIONS TO LINEMEN, 369 
 
 APPENDIX C. 
 ENGINEER'S LOG-BOOK, 381 
 
 APPENDIX D. 
 CLASSIFICATION OF EXPENDITURES OF ELECTRIC STREET RAILWAYS, 382 
 
 APPENDIX E. 
 CONCERNING LIGHTNING PROTECTION, BY PROF. ELIHU THOMSON, 389 
 
 APPENDIX F. 
 
 MOTORS WITH BEVELED GEAR AND SERIES-MULTIPLE CONTROL OF 
 MOTORS, 403 
 
 APPENDIX G. 
 
 METHOD FOR MEASURING INSULATION RESISTANCE OF OVERHEAD 
 
 LINES, 405
 
 THE ELECTRIC RAILWAY 
 
 THEORY AND PRACTICE. 
 
 CHAPTER I. 
 
 GENERAL ELECTRICAL THEORY. 
 
 IN common parlance men speak of electricity as " that mys- 
 terious force;" and indeed it is mysterious, but not more so 
 than the " force of gravity," " capillary attraction," " chemical 
 affinity," or such like familiar phenomena. "Scratch a fact 
 and you find a mystery" is a homely phrase the truth of 
 which is as a seal of bondage upon the human race. 
 
 Two portions of iron attract each other simply by virtue of 
 their mass (itself a word impossible to define absolutely) , and 
 the force between the two bodies diminishes as the square of 
 the separating distance increases. This phenomenon we say is 
 explained by the law of gravity. But why does the force vary 
 inversely as the square of the distance ? Why not as the cube, 
 or the fourth power? Why, indeed, does it vary at all? 
 Why is there attraction or force at any distance ? Is not the 
 whole matter, ultimately, as little understood as the fact that 
 under certain conditions the same iron bodies may exhibit the 
 action of another force, called another because its relation to 
 other phenomena its law is different from that of gravi- 
 tation? We call that other force magnetic. Its law seems to 
 be somewhat less simple, but ultimately is not more mysteri- 
 ous than that of the force of gravity. 
 
 5
 
 6 THE ELECTRIC RAILWAY. 
 
 In furtherance of efforts to simplify our conceptions of that 
 is, our working hypotheses for all the operations of nature, 
 it is convenient to consider: 
 
 First, that all bodies of appreciable magnitude are com- 
 posed of minute bodies of inappreciable magnitude. 
 
 Second, that these smaller bodies atoms or molecules are 
 perpetually in motion. 
 
 Third, that the varying states of any body as to hard- 
 ness, mechanical strain, temperature, light, etc., correspond 
 to varying modes of motion of the constituent molecules. 
 These may change their paths, their velocities, or both, or 
 their groupings. 
 
 Fourth, that the resultant force between a given body and 
 the surrounding medium will generally change when any 
 change occurs in the motion of its molecules. The change of 
 state in the body may thus be recorded by corresponding 
 change in a sentient organization, such as the human being. 
 
 It seems possible to conceive that all differences of material 
 in the universe may correspond to variations, from point to 
 point in space, of the motion of particles themselves ho- 
 mogeneous. 
 
 As the very nature of these supposed ultimate particles can- 
 not be accurately conceived, this mechanical conception of 
 the universe rises up, as do all other fundamental concep- 
 tions, from an inexplicable mystery. It does, however, serve 
 a good purpose in correlating phenomena. 
 
 Magnetism, or the property of developing magnetic force, 
 has been discovered in but few substances. Iron, nickel 
 and cobalt show this property in specially marked degree. Of 
 these, iron and its compounds steel and cast iron are by 
 far the most important. No iron can be said to be wholly 
 non-magnetic, although some of its alloys are nearly so. 
 
 To define magnetism completely would be to describe in 
 full its relations to other phenomena. These are not all 
 understood. Expressions have, however, been reached for 
 the chief characteristics of magnetic action. 
 
 To-day the most important relation of magnetism is that 
 with the flow of electricity. This latter term, like magnetism, 
 can, of course, be only approximately denned, and a complete
 
 GENERAL ELECTRICAL THEORY. 7 
 
 definition cannot be made without a lengthy recitation of its 
 characteristics. We will, however, be helped by some such 
 formula as this : There is said to be a flow or current of elec- 
 tricity in a given medium when due to an initial disturbance 
 which may be a magnetic, chemical, frictional or thermal 
 change the particles of that medium so move as to (i) pro- 
 duce the sensation called heat, or (2) light, or (3) certain 
 chemical changes, or (4) magnetic change, or (5) other elec- 
 trical flow, and (6) finally appealing again to bodily sensa- 
 tion the peculiar effect experimentally associated with the 
 phenomenon described. 
 
 This definition may seem to take much for granted, may 
 seem to be only a circular reasoning, but substantially all defi- 
 nitions found themselves on things taken for granted, other- 
 wise every definition would be a complete explanation of the 
 universe. 
 
 Such a mode of molecular motion a current of electric- 
 ity may be produced in any known substance, but it is most 
 readily produced in the metals and certain liquids. When 
 the disturbing force must be very great in order to produce 
 an appreciable electrical effect in a particular medium, that 
 medium is called an insulator or non-conductor. 
 
 The earliest method of producing a sustained electrical cur- 
 rent was that familiar even to-day in primary batteries. In 
 these it is found that if a metal be brought in contact with 
 another different metal, or with certain non-metallic bodies, 
 and both be then immersed in a fluid producing unequal 
 chemical reactions upon them, a current of electricity will 
 flow across the surface of contact between the two bodies, or 
 along, a third body (as a wire) if it be a conductor and joined 
 to the two immersed bodies. Zinc has been found to be gen- 
 erally the best metal for use in batteries, as that which is to be 
 most acted upon by the fluid usually an acid. 
 
 A primary battery is very convenient and useful for many 
 cases in which an electric current is desired. When, how- 
 ever, the supply of power (or current) is required to be con- 
 siderable the method is expensive to a prohibitory degree, as 
 zinc must constantly be consumed in proportion to the energy 
 supplied by the battery. The maximum possible supply of
 
 8 THE ELECTRIC RAILWAY. 
 
 energy from a pound of zinc is about 1,808,000 foot-pounds. 
 Zinc usually costs about six cents per pound. Hence if the 
 energy be utilized by a process having 95 per cent, efficiency, 
 the cost per foot-pound = 0.0000035 cent. 
 
 The maximum possible supply of energy from a pound of 
 good coal is about 1 1, 120,000 foot-pounds. The coal, at $5.00 
 per ton, costs then 0.25 cent per pound. The efficiency of 
 utilization in a good steam plant is about 1 3 per cent. , giving 
 to the fly-wheel about 1,445,600 foot-pounds per pound of 
 coal. The dynamos of modern make return 90 per cent, in 
 electrical energy of the mechanical energy received from the 
 fly-wheel of the engine. Hence the cost per foot-pound 
 delivered from the dynamos = .0000002 cent, or only one 
 seventeenth part of the cost in the case of the primary battery. 
 
 Ignorance of these facts has resulted in the wasteful 
 expenditure of much honest effort and money. 
 
 The possibilities in the matter of the direct production of 
 electrical energy from heat seem to be worth considering, 
 and some day may hear the proclamation of a success that will 
 revolutionize present methods in electrical work. For the 
 present, however, the familiar steam engine or water-wheel, 
 connected to drive an electro-magnetic machine, is the only 
 combination fitted to produce electrical energy for general 
 distribution to lamps or motors. 
 
 (A) Faraday observed some sixty years ago the fact that if 
 a conductor, say a loop of copper wire, be moved in certain 
 ways in the neighborhood of a magnet, an electrical current 
 is caused to flow in the loop. (B) It was further estab- 
 lished that if a current be made to pass around a magnetic 
 body as when a piece of iron is surrounded by a coil carry- 
 ing a current the magnetic action in the body would be 
 affected thereby. 
 
 These two facts or laws are involved in the operation of all 
 modern dynamos. The armature of such machines consists 
 of a number of conducting loops, the circuit of which is made 
 complete through the brushes and any wire or other con- 
 ductor to the ends of which the brushes may be connected. 
 
 For simplicity the armature is represented in Fig. i as a 
 simple loop. When rotation of this loop begins under the ac-
 
 GENERAL ELECTRICAL THEORY. 
 
 tion, say, of a steam engine, we have at once the case referred 
 to in (A). The magnets and their extensions, called the 
 pole pieces, are generally of wrought or cast iron, and these 
 
 FIG. i. DIAGRAM OF SIMPLE DYNAMO. 
 
 FIG. 2. SHUNT WOUND DYNAMO 
 
 substances are often, loosely, said to be not permanently mag- 
 netic, while steel is credited with that property. But in fact 
 these field magnets are always slightly magnetic. They thus 
 create around them a "field of force," that is, a region in 
 which magnetic force may be shown to exist, as by the col- 
 lection of iron filings around the poles. Assuming some unit 
 of force which need not here be defined, the strength of the 
 field is conveniently expressed by referring to the number of 
 lines (units) of force passing through a unit area, taken at 
 right angles to the direction of the force, as shown by the iron 
 filings. The slightly magnetized pole pieces cause a few 
 lines of force to traverse the space within which the armature 
 revolves, the result being that a very weak current flows to 
 one of the brushes, called the positive, thence through the
 
 10 THE ELECTRIC RAILWAY. 
 
 coils around the magnets and returns to the other, the nega- 
 tive brush By virtue of the flow that has taken place around 
 them, the magnets become (case B) more strongly magnetic, 
 the current generated by and in. the armature wires becomes 
 stronger, and so there goes on an interactive, cumulative 
 effect between armature current and magnetic strength, 
 until, usually in about 20 to 30 seconds, the full normal 
 strength of the magnet has been attained. If the current be 
 interrupted, either by stopping the driving engine or by 
 breaking the circuit, the wrought or cast iron magnets and 
 pole pieces will lose all the strength gained from the flow of 
 current, but will retain enough permanent magnetism to start 
 again in a similar cycle. 
 
 If all the electrical energy developed by the armature were 
 expended in the magnet coils, no useful outside work could 
 be done. Two methods will be explained by which the mag- 
 net coils receive the necessary exciting current, while a much 
 larger supply is provided for heating a carbon to incandes- 
 cence, or producing rotation in a motor armature which will 
 do mechanical work on any machine suitably connected :o it 
 by belt or gears. 
 
 In Fig. 2 it will be seen that starting from the positive 
 brush two paths are provided, one leading around the mag- 
 nets, the other through a lamp and a second armature, repre- 
 senting a motor. This second path is generally called the 
 " working circuit. " The first is called the " magnet " or " field " 
 circuit. 
 
 In any conductor there is found to be a certain resistance 
 
 to the flow of an electric current ; for any given substance the 
 
 resistance is found to increase directly as the increase of 
 
 length, and decrease directly as the increase of cross-section. 
 
 The relative resistances of circuits fed from the same source, 
 
 as in Fig. 2, can thus be easily regulated. If the resistance 
 
 of the magnet coils be ten times that of the working circuit, 
 
 the latter will receive ten times the quantity of current in the 
 
 rnner. Instead of having only a single wire for each of the 
 
 circuits, there may be any number, but they may always 
 
 magmed as forming a single wire, the resistance of which 
 
 st equal to that of the combination of wires ; and in such a
 
 GENERAL ELECTRICAL THEORY. 
 
 I I 
 
 discussion as this we need consider only this representative 
 wire. 
 
 Looking at Fig. 3 it is seen that the current has but one 
 path ; leaving the positive brush, it .passes around the mag- 
 nets, thence to the outer working circuit, and returns to the 
 negative brush. If in this case the total output of current be 
 the same as in the previous case, the number of turns in the 
 magnet coil may be very much less than before, the magnet- 
 izing effect remaining equal. It has been experimentally 
 proved, and may now be called a law, that the magnetizing 
 effects of different coils carrying varying currents may be 
 compared by comparing the product of the number of coils by 
 the strength of the current flowing through them. Thus ten 
 turns of wire carrying one unit (called an ampere) of current 
 will produce the same effect as one turn carrying ten units 
 
 FIG. 3. SERIES WOUND DYNAMO 
 
 FIG. 4. COMPOUND WOUND DYNAMO 
 
 or amperes. One hundred turns carrying half an ampere 
 (100 X 0.5 = 50) will produce one-half the effect of fifty turns 
 carrying two amperes (50X2 = 100) and thus generally, 
 multiply the number of turns in the magnet coils by the
 
 I2 THE ELECTRIC RAILWAY. 
 
 number of amperes flowing through those coils, and the prod- 
 ucts formed show the relative magnetizing effects of the 
 respective coils. This product of the amperes by the turns 
 measures what may be called the magneto-motive force. 
 
 Referring to Figs. 2 and 3, if we suppose the total current 
 to be the same, and that in Fig. 2 only one-tenth of the total 
 amount goes through the magnet coils, the number of these 
 coils, for equal magnetization, would have to be ten times as 
 great as in Fig. 3, in which there is but one circuit, contain- 
 ing the magnet coils, as well as the lamps or motors for 
 which the supply is given. In Fig. 2 the work of the exter- 
 nal circuit is done by a part of the whole current generated 
 (say 0.9) flowing under the whole of the electrical pressure 
 between brushes. In Fig. 3 the external work is done by 
 the whole of the current flowing, under a pressure less than 
 that existing between brushes by the amount required to force 
 the current through the magnet coils. The ratio between the 
 energy delivered to the dynamo and the useful work per- 
 formed by it may be the same in the two cases, and this ratio 
 is called the "efficiency " of the dynamo. 
 
 The arrangement shown in Fig. 2 constitutes what is 
 known as a "shunt winding " for a dynamo, that of Fig. 3 a 
 "series winding." A third type, Fig. 4, called a " compound 
 
 winding," results from a combination of the two part of the 
 
 required magnetizing effect (or ampere turns) being given by 
 a shunt coil, and part by a series coil. The use of the word 
 " shunt " as an electrical term results from an easy transition 
 from its use (more general in England than in America) as a 
 railway term denoting a branch or siding. 
 
 Any conductor carrying a portion, not the whole, of a cur- 
 rent flowing from a given point may be said to be in 
 1 shunt " relation to some other conductor or conductors. The 
 same relation is sometimes indicated by saying that the con- 
 ductors are "in parallel," "in parallel arc," "in multiple," or 
 Hiple arc" with respect to each other. It would be 
 best, perhaps, to use no other expression than "multiple." 
 5,6? show various relations of combined conductors, 
 i speaking of the quantity of electricity flowing in a con- 
 *or we have simply used a popular and convenient expres^
 
 GENERAL ELECTRICAL THEORY. 13 
 
 sion. No direct dimensional measurement of " quantity " can 
 be had. But by observing that in one case a current separates 
 a certain weight of water into its constituent elements, while 
 in another twice or three times the weight is thus decom- 
 posed ; that now one ounce of silver is deposited on a metal 
 
 FIG. 5. RESISTANCES IN SERIES. 
 
 surface, and again two or three ounces ; that now a lamp car- 
 bon is only dull red, while again it is brilliantly white we 
 are brought face to face with the different quantities of work 
 performed, and are forced to conclude that the molecules 
 themselves of the conductor have different quantities of the 
 
 FIG. 6. RESISTANCES IN MULTIPLE. 
 
 energy of motion, due to the original disturbing forces of 
 different magnitudes. The force producing the molecular 
 movement, which in turn presents electrical phenomena, is 
 called an " electromotive " force. If electromotive forces 
 emanating from different sources are applied to a conductor 
 
 FIG. 7. RESISTANCES IN SERIES MULTIPLE. 
 
 which itself remains unchanged, and it is learned, by measur- 
 ing their effects, that currents of the same or different 
 strengths flow from these different sources, then the equality 
 or inequality of the electromotive forces is implied. So, if 
 we consider the flow of water in a particular section of pipe, 
 we judge of the relative pressures causing the flow by noting
 
 1 4 THE ELECTRIC RAILWAY. 
 
 its quantity or measuring the work it can do, the resistance 
 to flow such as friction on the walls of the pipe being- 
 assumed constant. 
 
 Generally it will be readily understood that the quantity of 
 motion produced by a disturbing force acting in any medium 
 will be relatively great or small as the resistance to that form 
 of motion is small or great ; further, whatever the resistance 
 may be, if it be constant the quantity of motion will be greater 
 as the force becomes greater. In electrics this relation 
 may be expressed thus: The current flowing in any circuit 
 varies inversely as the resistance, and directly as the electro- 
 motive force (usually written E. M. F.). Adopting a system 
 of correlated units, if the E. M. F. be of magnitude expressed/ 
 by unity and likewise the resistance, then the current flowing 
 will be unit current. If the E. M. F. remain the same and 
 
 FIG. 8. ELECTROMOTIVE FORCES IN SERIES. 
 
 the resistance become 0.5 or 2, the current would become 
 twice its former value, or half, as the case may be. If the 
 resistance remain the same and the E. M. F. become 0.5 
 or 2, then the current becomes half its former value, or 
 twice, as the case may be. Generally, then, it appears that 
 we may write: 
 
 Current = ~~~ ; or abbreviating. C = | . 
 
 This relation is always known, from the man who demon- 
 strated it, as Ohm's law. 
 
 The specific resistances of many substances (that is, the 
 
 itances of bars of unit length and cross-section), expressed 
 
 units of resistance, have been carefully tabulated. 
 
 t would also be fair to assume equality of electromotive 
 
 s in different cases if there be the same physical arrange-
 
 GENERAL ELECTRICAL THEORY. 15 
 
 ment of parts in the generator. Thus a particular combina- 
 tion of zinc, copper and acid, if applied to any number of cir- 
 cuits, may reasonably be supposed to produce the same 
 electromotive force in all ; observed differences in the quan- 
 tity of current produced would then fairly be attributable to 
 differences of resistance in the various circuits. And, again, if 
 two similar combinations be placed in the same circuit, one 
 following the other, as in Fig. 8, it is fair to assume that we 
 have twice the electromotive force that would be given by 
 one of the combinations acting alone ; and that consequently 
 we would have twice the current if the resistance of the cir- 
 cuit were the same in the two cases. 
 
 Likewise, if we make two similar dynamos and run them 
 under similar conditions, they will produce equal electro- 
 motive forces. As to the force developed in any particular 
 case, experiment shows that it increases with the number of 
 complete changes from zero to maximum in the number of 
 lines of magnetic force passing through the looped conductor 
 of the armature in any fixed interval of time, and with the 
 magnitude of each change. It follows that in an actual 
 dynamo, increase in four separate elements will independently 
 increase the electromotive force produced in its armature, (i) 
 strength of magnetic field, (2) area of revolving loop, (3) 
 number of loops, when arranged each in series with its 
 neighbor, (4) velocity of rotation. If we can determine ex- 
 perimentally the electromotive force produced by a change of 
 one line of force through the loop in unit time, the dynamo 
 may then be proportioned to produce a given E. M. F. 
 
 Concerning the strength of field or number of lines of mag- 
 netic force per unit area it has already been explained that 
 the force called magneto-motive force, producing such effects, 
 increases with the increase of ampere-turns around the body 
 of the magnetic metal. While this is always strictly true, it 
 is also true, as in all other cases of disturbing forces, that 
 the quantity of motion, of any particular kind, depends upon 
 the resistances met, as well as the original forces acting. 
 Indeed, the relation between " lines of force," magneto-motive 
 force, and magnetic resistance may be expressed in terms as 
 simple as those of Ohm's law. Using the term " Induction " in
 
 j6 THE ELECTRIC RAILWAY. 
 
 a sense analogous to that in which " Current " is used, we 
 
 may write 
 
 M. M. F. M 
 
 Induction = 
 
 Resistance' R' 
 
 All substances save three (iron, nickel and cobalt) have 
 enormously high magnetic resistance. 
 
 Magnetic action takes place along lines connecting one pole 
 
 N 
 
 FIG. O.-BAK MAGNET. 
 
 of the magnet with the other, and the resistance to magnetic 
 action along such lines determines the " strength of field " set 
 up by a given magneto-motive force. 
 
 Thus (Fig. 9) in case of the straight magnet N S there is 
 a long air space between N and S, and air is susceptible to 
 magnetic action only in a very low degree. We may then 
 increase the strength of field by simply bending the magnet 
 into a U shape, thus shortening the air space between the 
 poles, without change in the dimensions of the magnet or in 
 the number of ampere turns exciting it. An armature revolv- 
 ing between N' S' (Fig. 10) would, at the same velocity, gen- 
 erate more than twice the E. M. F. that 
 it would if placed at either A or B (Fig. 9) . 
 Further, an armature whose wires are 
 wound on an iron core, occupying as much 
 as possible of the space between N and S, 
 will generate in each coil a greater E. M. 
 F. at same speed than would one on a 
 non-magnetic core, since the total resist- 
 ance to magnetic action is less in the 
 first case than in the second. 
 
 If, starting with the straight magnet, 
 we not only bend it to bring the ends near each other, but 
 also actually shorten the magnet, we again diminish the 
 total resistance in the magnetic circuit, and with the same 
 magneto-motive force obtain more " lines of force " passing 
 through unit area from N to S or we may again diminish 
 resistance by increasing the cross-section of the magnet, just 
 
 N 
 
 S' 
 
 FIG. 10. -HORSESHOE 
 MAGNET.
 
 GENERAL ELECTRICAL THEORY. I/ 
 
 as we would decrease resistance to an electric flow by using 
 large wires instead of small wires. 
 
 It might seem from the foregoing remarks that any desired 
 strength of field could be obtained by proportionate increase 
 of magneto-motive force or decrease of resistance. But un- 
 fortunately experiment has shown that after being magnet- 
 ized to a certain degree, even iron becomes stubborn as air or 
 other non-magnetic bodies to any increase of magnetism, 
 however high we may urge the magneto-motive force. When 
 magnetized to such a degree, iron is said to be "saturated." 
 It has been further found, as might be expected, that in 
 approaching saturation the resistance of iron to further action 
 becomes gradually greater until it becomes as great as air. 
 
 20 25 
 
 FIG. n. MAGNETIC PRO 
 
 ERTIES OF VARIOUS IRONS. 
 
 Over a certain range, the increase of magnetic strength in 
 iron seems exactly proportional to increase of the magnetiz- 
 ing ampere turns, but beyond that range this strict proportion 
 disappears. Thus let us say that ten ampere turns (often 
 written A-ts) produce a magnetic strength denoted by unity. 
 Twenty ampere turns may then produce a strength 2, but in 
 going to forty we may find the strength not 4 but 3 only, 
 and in going to 160 we find the strength barely 3.5, and so
 
 ig THE ELECTRIC RAILWAY. 
 
 i until a point is reached where no perceptible increase of 
 str'eng h follows from any increase of magneto-motive force 
 This change in the magnetic resistance of different kinds of 
 iron is best shown by the curves (Fig. 1 1). 
 
 On the horizontal axis are shown magneto-motive forces 
 increasing from left to right. On the vertical axis are shown 
 the corresponding strengths of magnetization produced in 
 various materials under like conditions. If the iron did not 
 increase in magnetic resistance the E. M. F. generated in an 
 armature would increase steadily with the M. M. F., as H 
 does very nearly, until the curves turn sharply to the right- 
 but beyond that point the increase of E. M. F. is less, rela- 
 tively, than the increase of M. M. F. 
 
 The general principles governing the design of dynamos 
 and motors have now been discussed. 
 
 To show their application, let us suppose that it is desired 
 to construct a machine of the so-called U type (see shape of 
 letter U suggested by Fig. 10) capable of delivering a current 
 of 80 amperes at a pressure of 200 volts. We will predeter- 
 mine the speed let it be 600 revolutions per minute, or 10 
 per second. Further, suppose it is desirable that a solid core 
 
 armature generally known as a Siemens or drum armature 
 
 shall be used, and that the core shall be a cylinder 10 inches 
 in diameter and 10 inches long. The area inclosed by one 
 loop would then be 100 square inches. It has been experi- 
 mentally determined that when soft iron has been magnetized 
 so highly that 120,000 of the "lines of force," above men- 
 tioned, may be said to pass through each square inch of 
 metal riormal to the axis of the magnet, then the mass has 
 practically reached the point of saturation. To carry the 
 iron to that point requires a very great M. M. F. Let us be 
 content to obtain in the armature core 80,000 lines per square 
 inch. We cannot in practice avail ourselves of the entire 
 cross-section of the armature core as an iron conductor of mag- 
 netic lines of force, because this core is not made solid, but 
 generally is built up of a great number of very thin plates of 
 iron separated by paper or other non-magnetic material. 
 This is done to prevent the flow of currents of electricity in- 
 duced by moving such a conducting mass in a field of force,
 
 GENERAL ELECTRICAL THEORY. 19 
 
 and which, in a solid iron core, necessarily of very low resist- 
 ance, would be very great. Their injurious effects would be 
 (i) heating of the armature and (2) consequent loss of energy. 
 
 The use of paper reduces the iron to a cross-section about 
 85 to 90 per cent, that of the area of the loop. Hence if we 
 obtain 80,000 lines per square inch through the armature as 
 a whole, we must obtain about 18 per cent, more through the 
 iron of the armature, or about 95,000 lines per square inch 
 of iron. 
 
 Experience has shown that if the total lines of force pass- 
 ing through a section at D T (Fig. 12) be represented by 100, 
 
 FIG. 12. MAGNETIC CIRCUIT OF DYNAMO. 
 
 then the number passing through the armature will be from 
 60 to 80, the remaining 40 to 20 passing through the air, as 
 indicated in the figure. 
 
 With reasonable attention to shaping our machine we may 
 count upon a loss not greater than 30 per cent. In order, 
 then, to have 80,000 lines per square inch through the arma- 
 ture loops, we must either have the section at D T of greater 
 area than at V W, or must force about 120,000 lines per square 
 inch through the iron of the magnets, and their connecting 
 piece, the " keeper." Supposing the case not to be one which 
 requires the weight of iron to be a minimum, we would pre- 
 fer to increase the area so that the total lines at D T shall be 
 distributed at the rate of 80,000 per square inch. The sec-
 
 20 THE ELECTRIC RAILWAY. 
 
 tion A B of the magnet core may be nearly a mean between 
 the sections V W and D T, the lines per square inch being 
 still 80,000. 
 
 To obtain absolute values for these areas we may proceed 
 as follows: It has been experimentally shown that if one 
 loop be rotated once per second in a field such that one unit 
 line of force passes through the loop, then the E. M. F. gen- 
 erated in such a loop will be 0.00000002 volt. As we have 
 chosen a speed of 600 revolutions per minute, or ten per sec- 
 ond, we may at once use the larger coefficient 0.0000002 (see 
 page 17). There must be such a number of loops and such 
 an enclosed area through which 80,000 lines per square inch 
 are passing that the number of loops, multiplied by the total 
 lines of force enclosed, multiplied by the coefficient 0.0000002, 
 shall produce the desired number of volts. It has already 
 been assumed that the machine shall show a difference of 
 potential of 200 volts between tJie brushes. It must generate 
 a somewhat greater E. M. F. since the loops themselves will 
 offer a certain resistance, requiring a certain E. M. F. to force 
 the current over them. 
 
 Let us assume that this requirement shall not exceed 4.0 
 volts. The total to be generated will then be 204. Let us 
 represent by x the area, in square inches, inclosed by the 
 armature loops; by y the number of such loops; we may 
 then write this simple equation 
 
 204 x X y X 00000002 x 80000; 
 
 or 204 = x X y X 0.016; 
 
 204 _ 12750 
 
 : y x 0.016 ~ ~~y~ 
 
 We may readily obtain a working value for y based on 
 these considerations. All the loops represented by y must 
 be connected by the commutator bars, either by making only 
 one loop out of each length of wire, and connecting its ends 
 with adjacent bars ; or by making several loops from one 
 length of wire, the number of commutatator bars being less- 
 ened to one-half, one-third, one-fourth, etc. Electrical con- 
 siderations preponderate in favor of equal numbers of loops 
 and bars, and in larger machines, of comparatively low voltage
 
 GENERAL ELECTRICAL THEORY. 21 
 
 and high speed, it is good practice to produce such equality. 
 The difference of potential between successive bars is thus 
 reduced to a minimum for the given number of loops ; this 
 reduces the tendency to spark at the brushes; injurious in- 
 ductive disturbances in the loops themselves are likewise 
 reduced by decreasing the number of loops per section. Or- 
 dinarily, however, it is difficult, with present methods of 
 commutator construction, to produce this equality. Practice 
 has shown that we should limit the difference of potential 
 between successive bars to about ten volts. As we have 
 in the present case 204 volts, we might have about 20 bars 
 in each half of the commutator, or 40 in all. But, as we also 
 know from practice, a considerably larger number of bars 
 may be made into a commutator suitable for a machine of the 
 kind, in view of which we may adopt a larger factor of safety 
 as to sparking. For convenience let us assume 64 bars in the 
 commutator. If we further assume two loops to the bar, 
 the value of y becomes 128, and the value of x practically 100 
 square inches. If the cylinder of which this figure gives the 
 area of cross-section be taken as ten inches long by ten 
 inches in diameter, we have an armature of convenient and 
 usual size for the conditions to be met. We may now readily 
 determine, from what has before been said, that the section 
 D T should be about 160 square inches; the section A B 
 about 130 square inches. If these areas be circular, the 
 "diameters are about 1 3 inches. 
 
 We have now determined the cross-section of the iron in 
 the magnetic circuit, except that of the pole pieces. While 
 passing through these the magnetic lines curve in toward 
 the armature, and those on the upper side of the shaft have 
 a path somewhat shorter than those below. To produce a 
 symmetrical distribution through the armature core, it is 
 desirable to make the lower portion of the pole pieces quite 
 full, i.e., of as large area as x practicable. No exact rule need 
 here be given for this shaping of the pole pieces, but, as just 
 indicated, the magnetic resistance along E F should be as 
 nearly as possible equal to that along E G. One other con- 
 sideration affects the shape of the pole piece and its area
 
 22 THE ELECTRIC RAILWAY. 
 
 normal to the lines of force, that is, the angle between the 
 lips, SCR. Without entering into the rather lengthy discus- 
 sion's which have been heard as to the best angle, we may 
 assume for an ordinary case, without serious error, that each 
 pole piece covers 145 in arc of the armature. The length of 
 this arc on the edge of the pole pieces will then be very nearly 
 13 inches. 
 
 The length of the line E H, which may be taken as the 
 length of the magnetic circuit through the pole piece, de- 
 pends on the distance between the axes C D and E K. This 
 in turn depends on the diameter A B (13 inches) and the 
 breadth B L of the space left for the magnet coils and the free 
 space required for hand room and ventilation. From general 
 experience we may safely make B L = 4 inches, thus making 
 K D = 10.5 inches, the resulting length of keeper M N thus 
 being about 34 inches and E H about 7 inches. 
 
 If we make proper allowance for the air space from surface 
 of pole to surface of armature core (this will be done later) 
 we may, as to length of iron circuit, consider E H and H C 
 as continuous, making a total of 12 inches as to length, and 
 without great error we may consider the area for this length 
 as 100 square inches (it is greater in pole, less in armature). 
 In the keeper we have already a denned length of 34 inches, 
 or for one-half the magnetic path in this part about 17 inches, 
 with cross-section of 150 square inches. If now the length 
 E K of the magnet and the length of the air gap were known, 
 the total resistance of the magnetic circuit would become 
 known. As E K is dependent upon the M. M. F. required 
 (and hence the space needed for winding the magnet coils), 
 and as the M. M. F. is largely dependent on the resistance 
 of the air gap, let us first consider that. Its length from 
 iron to iron is determined by the space needed for laying the 
 loops of armature conductors (of copper), insulation for same, 
 and a reasonable clearance between exterior of armature and 
 pole piece. This clearance we may fix at o. i of an inch. 
 We may also allow o. i of an inch for laying the armature 
 conductors; or, since copper is practically of equal magnetic 
 stance with air, we may say that the air space is 0.2 inch 
 m length. Its area is about 130 inches, namely R S (= 13
 
 GENERAL ELECTRICAL THEORY. 23 
 
 inches) X length of pole piece (= 10 inches = length of arma- 
 ture).* 
 
 The total " induction" or flow of magnetic lines of force, 
 through the armature is 80,000 X 100 = 8,000,000. This 
 number must be forced across the air gap, being equivalent 
 to about 62,000 per square inch. The magnetic resistance 
 of this air gap will increase with its length and decrease with 
 its cross-section : we may therefore, simply for convenience, 
 calculate the M. M. F. needed for forcing 62,000 lines across 
 one square inch; that force will be the same required for 
 total flow across the total section. From experience it is 
 known that to force one unit line through an air path one 
 inch long and one square inch in cross-section we must apply, 
 as M. M. F., 0.25 ampere turn. In our particular case the 
 length of path is only 0.2 inch, while the number of unit 
 lines is 62,000 hence for the total M. M. F. we have 3,100 
 ampere turns (0.25 X 62,000 X 0.2 inch = 3,100). 
 
 Going to the pole piece and armature, as has been said, 
 the magnetic resistance of iron varies with the " induction" 
 maintained. In this case we require 80,000 lines per square 
 inch, and at that density the resistance of air is about 1,200 
 times greater than that of soft iron, or for one unit line 
 through a path one inch long and one square inch in cross- 
 section 0.000208 ampere turn is required. Hence for total 
 length (12 inches) and total induction the ampere turns = 
 0.000208 X 80,000 x 12 = 200. 
 
 We may now go to the keeper. 
 
 The machine being symmetrical, we may take half its 
 length separately, since we are now determining the M. M. 
 F. needed to overcome resistances in one-half the machine, 
 and this force is to be supplied by the current encircling one 
 of the magnets. The induction being 80,000 lines per square 
 inch, we have 0.000208 X 80,000 X M T = about 300. 
 
 Let us now assume an approximate M. M. F. for the mag- 
 net core. We may use 900 ampere turns as an outside figure 
 
 * If the armature conductors be wound in slots, cut in the periphery of the armature 
 core, instead of being laid on the surface of this core, the length of the air gap proper 
 may be considerably diminished. It is then to be remembered that since the iron of the 
 core has, to the depth of the slot, been largely cut away, and the spaces filled with non- 
 magnetic substances, the magnetic resistance of this outer ring of the core is greater than 
 before the slotting.
 
 ^ 4 THE ELECTRIC RAILWAY. 
 
 allowing a considerable margin for joints in the magnetic 
 circuit Add this to the other figures, we have 3, 100 + 200 + 
 ,00 + 000 = 4,500. Now this number of ampere turns as 
 has been explained, may be made up of any two selected fac- 
 tors representing current and tarns respectively. 
 
 If we wish to design a series motor, the current is at once 
 determined by the original assumption to be 80 amperes. 
 
 In this case the resistance and number of turns would be 
 small. If a shunt motor is under consideration, we should 
 make the current quite small, that the constant loss of power 
 in the field may be small, the resistance and number of turns 
 being relatively large. Assume 1.5 ampere for the field 
 current in this case. 
 
 Then the number of turns = 4,500 -=- 1.5 = 3>- The 
 resistance of these turns follows from the fact that this cur- 
 rent is to be produced by an electromotive force of 100 volts, 
 being one-half the total of the normal E. M. F. for which 
 the machine is calculated. From the rule previously given, 
 namely, 
 
 C = S we have 1.5 = ^' .'. R = 66 ohms. 
 K. K 
 
 This then must be the resistance of the 3,000 turns of wire. 
 The average length of each turn, assuming the winding to 
 be two inches deep, will be equal to the circumference of a 
 circle about 13.5 inches diameter the magnet- core being 
 supposed to be of circular cross-section. This length equals 
 42 inches, or for the 3,000 turns 10,500 feet. Its resistance 
 per foot must be 66 -=- 10,500 = 0.00629 ohm. From tables 
 prepared for the purpose we find that a wire having a diam- 
 eter of 0.052 inch will serve our purpose. When covered 
 with the usual cotton insulation such a wire would have an 
 outside diameter of about .075 inch. 
 
 Approximately 300 such wires could be laid along one inch 
 of the length of the magnet, the depth being two inches. 
 The length of magnet core would then be 10 inches, in order 
 to place the 3,000 turns. 
 
 Making now calculations for the M. M. F. as determined 
 by this length of magnet core we have : ampere turns re- 
 quired = . 000208 X 80,000 X 10 = 166.
 
 GENERAL ELECTRICAL THEORY. 25 
 
 Our assumed value, namely 900, is thus found to be ample, 
 providing a large allowance for joints in the magnetic circuit, 
 the resistance of a good joint having been estimated to be 
 equal to that of 8 inches of iron having a cross-section equal 
 to the area of the joint. 
 
 . It remains now to determine the size and number of wires 
 to be wound on the armature, and also the number of sepa- 
 rate sections into which the total number shall be divided. 
 
 As will be seen by reference to Fig. 13, the current in SOW- 
 
 FIG. 13. DIAGRAM OF ARMATURE CONNECTIONS. 
 
 ing over the armature as a whole is divided into two equal 
 parts. In this case we will then have 40 amperes to be 
 forced over a circuit made up of one-half the whole length of 
 wire on the armature, and we have already designated 4.0 
 volts as the E. M. F. for this purpose. The resistance of this 
 
 A 
 
 half of the armature wire must follow, 40 = . . R = o. i . 
 
 R 
 
 The total length of a turn around the armature (see dimen- 
 sions) will be 40 inches, or with a reasonable allowance for 
 the length of the end going to commutator, and for over- 
 wrapping at the armature heads, we say 44 inches. For one- 
 half the total length we have, in feet, 44 X 64 ^12 = 235. 
 
 The resistance per foot must be o. i -f- 235 = 0.00043, given 
 by a wire having a diameter of . 144 inch (No. 7 B. & S.). 
 
 We may use this, or more conveniently perhaps two smaller 
 wires in multiple, and on investigation we find that two wires
 
 2 6 THE ELECTRIC RAILWAY. 
 
 of No. 10 B. & S. gauge will give equal resistance, while per- 
 mitting more space, measured along a radius of the armature, 
 for insulation and clearance. 
 
 All the principal features in the design have now been 
 determined. The values assumed for speed, E. M. F., num- 
 ber of commutator bars, etc., entering as "known quantities" 
 in the problem, have of course been made to " fit" more 
 accurately than is often the case on first trial, in the practice 
 of design. Certain refinements have not been mentioned, 
 being considered beyond the scope of this work. 
 
 Much might be said concerning the effect of the magneti- 
 zation of the armature itself upon the poles ; of the mutual 
 and self-induction of the armature conductors ; of the exact 
 shaping of the lips of the pole pieces; of the important 
 phenomena of hysteresis, involving losses of imparted energy 
 due to rapid changes of magnetic polarity in the armature 
 core (or field magnets if they be rotated) ; of the rise of tem- 
 perature in the various parts of the machine, etc. ; but it is 
 believed that these should be left to works more technical 
 than the present. 
 
 We have spoken just now of the principles controlling the 
 design of a dynamo that is, a machine whose function is to 
 generate electric current which may be used for lighting, 
 heating, or for producing motion in other electric machines, 
 similar in construction to the dynamo, but whose function it 
 is to transform the electric energy received from the dynamo 
 into mechanical energy. Such a secondary machine is called 
 generally a motor. The principles of design are identical, 
 and indeed a machine made for use as a dynamo may instead 
 serve as a motor and vice versa. 
 
 In the dynamo the action may be briefly described thus : 
 Given lines of magnetic force and looped conductors located 
 in the resulting field of force, a rotation of such conductors 
 produces electric currents and requires continuously supplied 
 mechanical energy to continue such rotation against the 
 attraction set up between the lines of force and the electric 
 current in the conductors. In the motor a complementary 
 action takes place and maybe thus described: Given lines 
 of force and looped conductors located in the resulting " field, "
 
 GENERAL ELECTRICAL THEORY. 2/ 
 
 a flow of current through such conductors will produce an 
 effort at rotation due to attraction set up between the electric 
 current and the lines of force, and requires continuously elec- 
 tric energy to produce rotation against the mechanical resis- 
 tance presented by the friction of parts and the useful work 
 or "load," and to overcome electrical resistances in the arma- 
 ture and magnet circuits. 
 
 The energy for magnetizing the field is a very definite 
 quantity and may be calculated as explained above. Consid- 
 ering the armature of a motor, it is apparent that when rotat- 
 ing under the influence of a current from an external source 
 it likewise fulfills in all respects the conditions required for 
 action as a dynamo ; i.e., Tor the generation of an electromotive 
 force within its own conductors. Such an electromotive force 
 is indeed set up, as shown by the fact that, save when a 
 motor armature is held mechanically against all possibility 
 of rotation, the current flowing through is not that which 
 would be produced by the given external E. M. F. acting 
 over the given resistance of the armature, but is a less cur- 
 rent, such as would flow over such a resistance, under the 
 influence of an E. M. F. equal to the difference between the 
 external E. M. F. and the E. M. F. that would be produced 
 by such a machine if driven at the given speed as a dynamo. 
 This is actually the E. M. F. generated by the motor, and is 
 generally called counter E. M. F. 
 
 Now the total energy absorbed by the motor is, of course, 
 that measured by the product of the external E. M. F. (called 
 usually the "impressed" or "applied" E. M. F.) and the cur- 
 rent that may actually flow ; or, to use expressions more 
 familiar in hydraulics and pneumatics, these two elements 
 of pressure and of quantity when multiplied together meas- 
 ure the power or "work." 
 
 The mechanical energy developed by the motor armature 
 is measured by the product of the current in the armature and 
 the counter E. M. F. The ratio between the electrical energy 
 absorbed and the mechanical energy given out by the motor 
 is the measure of its efficiency. We may express absorbed 
 energy by C X E, and developed energy by C X E' (E and E' 
 being impressed and counter E. M. F. respectively). The
 
 2 8 THE ELECTRIC RAILWAY. 
 
 C* \/ Tn^' TH*' 
 
 efficiency then becomes ^ ^ = g This efficiency is 
 
 found in many motors to be as high as 93 to 95 per cent., 
 leaving little to be desired. The same machine may be 
 caused, by overload, to work at lower efficiencies, since the 
 great currents required for the work require greater propor- 
 tions of the absorbed energy simply to force these currents 
 over the internal resistance of the motor, leaving less for 
 external useful work. This point will be more fully devel- 
 oped later.
 
 CHAPTER II. 
 
 CONCERNING PRIME MOVERS. 
 
 BY prime movers we mean such mechanisms as are fitted to 
 utilize somewhat directly natural sources of energy for the 
 production of mechanical power. So far as electrical pur- 
 poses are concerned, the number of available prime movers 
 is very limited, and when we take into consideration the par- 
 ticular demands to be met in the supply of electrical power 
 for traction purposes, we find the only generally available 
 prime movers to be steam engines and water-wheels. One or 
 two very small electric roads in England are operated by 
 gas engines, but in this respect they are quite alone, and the 
 practice, for various reasons, is seldom to be advised. It 
 does not befit our purpose here to enter into any exhaustive 
 discussion of the theory of the steam engine, or to describe 
 in detail the very numerous modifications of the few types 
 that are met in modern engineering practice. We shall 
 merely attempt to give in a very brief form, without having 
 recourse to mathematics, the general principles involved in 
 the construction and operation of modern engines, and to 
 give some account of the means by which their action and 
 particular properties may be most conveniently investigated. 
 Having done this, it will only remain to describe in a manner 
 necessarily condensed the general principles of the utiliza- 
 tion of water power. . 
 
 The steam engine is to-day our most universal means of 
 obtaining mechanical power, and although the theory of its 
 action is somewhat complicated the main facts are easily 
 understood. The fundamental principle of heat engines, of 
 which steam engines are the most familiar variety, is the 
 production of pressure in a gas by the transference of heat 
 energy to it, and the utilization of this pressure to recover, 
 in the form of mechanical motion, the energy thus given. The 
 
 29
 
 , THE ELECTRIC RAILWAY. 
 
 primary source of energy is that which exists as potential 
 chemical energy in fuel of various sorts. The gas almost 
 universally employed is the vapor of water; in other words, 
 steam. It is a singular fact in physics that all gaseous bodies 
 are remarkably alike in their dynamical properties. They 
 expand and contract, if free to do so, in almost exactly the 
 same amount when the temperature varies ; or, if the volume 
 be kept constant, the pressure varies with the temperature 
 very uniformly for all gases and all temperatures. 
 
 If, then, we are to utilize steam pressure practically in an 
 engine cylinder, the higher the temperature of the steam the 
 greater will be its pressure. The amount of work that can be 
 gotten out of a given quantity of steam at a certain pressure 
 obviously depends on the extent to which we are able to 
 utilize that pressure. If we could employ it all in pushing- 
 on a piston we should be able to get very efficient engines 
 indeed ; but since we are living in an atmosphere that has 
 a pressure, and in a world that has sometimes a rather high 
 temperature, we are in practice totally unable to avail our- 
 selves of all the pressure produced. In other words, if we 
 give a certain amount of heat to a volume of steam, creating 
 thereby a high pressure and temperature, we shall be unable 
 to get all the heat energy back in the form of mechanical mo- 
 tion of the engine simply because of the necessary limitations 
 imposed by our conditions with respect to pressure and tem- 
 perature. Aside from these the energy lost is small in 
 amount. 
 
 It is found by experiment that if we were to have a unit 
 volume of some gas, like air, at the temperature of melting ice, 
 one degree Fahrenheit rise of temperature would increase its 
 pressure by i-493d part. Consequently, there might be some 
 temperature so low that if it were possible to reach it the gas 
 would have no pressure ; in other words, it would have lost 
 its elasticity and given up all its energy. This point is 461 
 degrees below zero Fahrenheit, as appears from the rate of 
 increase just given. This particular temperature is known 
 as the absolute zero. If we were able to start at this point, 
 and by applying heat to raise the temperature of any gas up 
 to our present working temperature, we should give to that
 
 CONCERNING PRIME MOVERS. 31 
 
 gas a certain amount of heat energy, all of which could be 
 recovered if we were able to let it expand and cool itself off 
 by expending its heat as mechanical work until all its press- 
 ure was exhausted, when its temperature would be back to 
 the starting point of minus 461 degrees. Unfortunately, we 
 are unable either to start with a gas or vapor at such a low 
 temperature, or to reduce its pressure after heating any- 
 where nearly to zero. 
 
 Obviously we cannot use all its pressure unless we get 
 down to the absolute zero, and in practice we are compelled 
 to cease utilizing its expansive power when the temperature 
 goes down to the every-day temperature at which we live ; so 
 only part of the whole amount of heat can be recovered as 
 mechanical work, the exact proportion depending on the ex- 
 tent to which we are able to cool by expansion the working 
 gas, allowing it to do work and lose pressure. If, for example, 
 we work with gas at 400 degrees Fahrenheit, and let it 
 expand until it cools down to zero Fahrenheit, we shall utilize 
 that portion of the pressure of the gas that corresponds to a 
 change in temperature of 400. All the pressure which cor- 
 responds to the temperature from zero down to the point 
 where pressure ceases is unavailable.* 
 
 The efficiency of such a transformation of energy is, then, 
 somewhat less than 50 per cent. To be exact, the effi- 
 
 'p 'p 
 
 ciency of the heat engine is equal to the fraction ' , 
 
 where T, is the highest temperature to which the gas or 
 vapor has been heated, and T Q the temperature at which we 
 are compelled to stop utilizing its pressure, T being always 
 reckoned from the absolute zero. In practical engineering, 
 T 2 is the temperature of exhaust or condensation. The 
 general law of the efficiency of the utilization of heat by en- 
 
 * In ordinary engines a nearly saturated vapor, such as steam, is used, and in produc- 
 ing it much heat is taken to drive the substance from the liquid to the gaseous state, so 
 that, although we begin to give heat to the gas at a comparatively high temperature, the 
 total heat required is much as if we had started at or near the absolute zero. As various 
 liquids require different amounts of heat to vaporize them, one mi^ht think that there 
 could be a great gain by properly choosing the liquid In fact, however, the liquids 
 that vaporize with little heat per unit weight give a considerably smaller total volume 
 of vapor, so that, so far as working in engines ib concerned, the particular liquid used 
 makes little or no difference.
 
 3 2 THE ELECTRIC RAILWAY. 
 
 giiies is that the heat transformed into mechanical energy 
 is to the whole heat received by the working fluid as the 
 range of temperature is to the absolute temperature reached by 
 the fluid. The perfect efficiency theoretically possible can 
 never be reached because of our inability to command suffi- 
 cient range of temperature, but the best modern engines 
 come quite near to the maximum efficiency possible between 
 the temperature limits attainable in practice. Taking nature as 
 we find it, and even condensing our working gas at ordinary 
 temperatures, we can really recover as mechanical work a 
 very respectable proportion of heat. Sufficient has been said, 
 however, to show that the higher the initial pressure given to 
 the steam and the lower the pressure at which it is exhausted 
 or condensed, the more efficient will be the engine in con- 
 verting the heat energy of coal into mechanical power. 
 
 Now in every-day work the pressure of steam is utilized in 
 only one way, that is, by employing it to drive a piston along 
 a steam cylinder, thereby working a driving wheel of some 
 kind by means of the piston rod. The rudimentary steam 
 engine, then, consists of a closed cylinder containing a mov- 
 able piston, with a piston rod reaching into the outside air. 
 The pressure of the steam, admitted to the cylinder by valves, 
 drives the piston to and fro. In some of the earliest engines 
 the steam was admitted to the cylinder through a valve oper- 
 ated by hand, and when the end of the stroke w T as reached 
 the steam, instead of being allowed to escape, was condensed 
 by a spray of water thrown inside the cylinder; afterward 
 the steam was allowed to escape at the end of the stroke 
 through an exhaust valve, or else to flow into a separate 
 chamber, where it was condensed. Finally, about one hun- 
 dred years ago, the engine was provided with valves to admit 
 automatically the steam to the cylinder, and to let it escape at 
 the end of the stroke. With various modifications, that is 
 the form of machine used to-day. 
 
 Fig. 14 shows a diagram of the working parts of a modern 
 
 engine very extensively used for electric light and power pur- 
 Its cylinder is provided with a steam port A, at each 
 
 tid, communicating with the steam chest S S. This latter is 
 placed immediately next the cylinder, and receives the steam
 
 CONCERNING PRIME MOVERS. 
 
 33 
 
 at full boiler pressure through the supply pipe. The entrance 
 of the steam into the cylinder is controlled by the valve V, 
 which is a hollow cylinder closed at the ends and contracted 
 somewhat in the middle. This piston valve has a slight to- 
 and-fro motion communicated to it by an eccentric on the en- 
 gine shaft, and slides backward and forward nra cylindrical 
 cavity in the steam chest ; at each end of this is an exhaust 
 pipe communicating with the outer air, or with the condenser. 
 Observe that in this arrangement the* valve is surrounded 
 on all sides by live steam, and consequently is not pressed 
 against any of the wearing surfaces: in other words, it is 
 .what is called a balanced valve. In the position shown the 
 valve is just admitting steam to the piston end of the cylin- 
 der; a very even flow is secured, not only around the shoul- 
 der of the valve as its enlarged portion passes beyond the 
 
 fin rm rrn 
 
 FIG. 14. WORKING PARTS OF SIMPLE ENGINE. 
 
 steam port, but also through the ports P P at the opposite end 
 of the valve and through the inner space. At the same time 
 it will be seen that at the other end of the cylinder the valve 
 has moved in its bearing completely past the steam port, 
 allowing the steam to flow out freely into the exhaust ; when 
 3
 
 24 THE ELECTRIC RAILWAY. 
 
 the piston has moved a little way on its return, the port A t 
 through which the steam has been flowing, will be closed as 
 the solid part of the valve slides over it, and open for the 
 exhaust at the other end of the stroke when the valve slides- 
 clear beyond. It will thus be seen that the same mechan- 
 ism admits the steam to the cylinder or allows it to flow into- 
 the exhaust, according to the relative positions of the valve. 
 This rudimentary description of the steam engine will be- 
 unnecessary to most* of our readers, but will prove useful in. 
 connection with points to be presently mentioned with refer- 
 ence to the economical use of steam. 
 
 Suppose, now, that we have communicated heat to a steam. 
 boiler to produce a quantity of steam at an ordinary working- 
 pressure such, for example, as 100 pounds to the square 
 inch in excess of the ordinary pressure of the atmosphere 
 a very practical question arises : how can that steam be so- 
 utilized in an engine as to give the maximum return in. 
 mechanical work? From what we have already seen, it will 
 be evident that the steam must not be allowed to cool off and 
 lose pressure, because all the heat so lost is simply so much 
 possible mechanical work thrown away. Consequently, ta 
 begin at the beginning, the pipe leading the steam from the 
 boiler to the engine ought to be thoroughly jacketed, so that 
 the steam will not get cooled on its way to the engine. Hav- 
 ing arrived at the steam chest, it is evidently best to employ, 
 if possible, the full pressure of the steam to push along the 
 piston; naturally, therefore, the steam ports and other aper- 
 tures through which the steam has to pass must be large 
 enough to permit its passage without serious check and 
 reduction of pressure, such as is known as wire-drawing- 
 among engineers. And in general, the efficiency of the 
 engine will be increased by receiving the steam at the high- 
 est temperature, that is, at the highest pressure possible, and 
 discharging it at the lowest temperature possible ; in other 
 words, after having obtained all the heat available from it in 
 the form of mechanical work. 
 
 The fundamental principle of the efficient working of 
 steam depends on following out this very obvious suggestion, 
 that has been admirably stated by Rankine as follows;
 
 CONCERNING PRIME MOVERS. 35 
 
 41 Between given limits of temperature the efficiency of a 
 thermo-dynamic engine is the greatest possible when the 
 whole reception of heat takes place at the higher, and the 
 whole rejection of heat at the lower limit." 
 
 It is desirable, therefore, in admitting steam to the cylin- 
 der to let it in, so far as it is admitted at all, at full boiler 
 pressure, and to exhaust it as completely as may be at the 
 lowest available pressure. A vast amount of ingenuity has 
 been spent in arranging the steam engine so as to produce 
 this result ; in other words, to move the valves in such a way 
 that the steam admitted may come in at a uniform high 
 pressure and be exhausted at a uniform low pressure. The 
 difference between a good and a bad engine lies largely in the 
 success with which this has been accomplished. In the early 
 days of the steam engine it was usual to leave the admission 
 valve open, allowing the steam to rush in direct from the 
 boiler, until the piston rod reached the end of its stroke. 
 Such a plan permitted a given engine to do a great deal of 
 work, because the piston was all the time subjected to the full 
 pressure of the steam, but at the same time it is obviously 
 very uneconomical, since the steam when we are through 
 using it will have nearly the same temperature and pressure 
 as at first, the deficiency, which has gone into work in push- 
 ing the piston along, being kept up by fresh supplies from 
 the boiler during the stroke. An engine of this character 
 requires, evidently, enormous amounts of fuel, so one of 
 the early improvements in engineering practice was to use 
 the steam expansively ; in other words, to admit to the cylin- 
 der a comparatively small amount of steam, and let it finish 
 up the stroke of the piston by its own expansion. 
 
 Under these circumstances, steam enters at full boiler press- 
 ure only, perhaps, for a quarter of a stroke; the supply 
 valve being then closed, the steam expands and uses up its 
 heat quite completely in doing work on the piston, so that 
 when the stroke is completed the steam exhausted will be 
 low in temperature and pressure or, in other words, work- 
 ing steam expansively enables us to increase the range of 
 temperature through which our heat engine works, and hence 
 increases its efficiency. In so using steam, however, the
 
 3 6 THE ELECTRIC RAILWAY. 
 
 pressure produced on the piston is not uniform, and should 
 we desire to compute the rate at which the engine is doing' 
 work, it is necessary to know the average pressure that the 
 expanding steam has exerted. An instrument for finding 
 this average will be described later. 
 
 If an engine is allowed to condense, that is, to exhaust its 
 steam into a vessel that is kept cool, it is possible still further 
 to increase the working range of pressure by very nearly 
 the pressure of the atmosphere, and hence the efficiency. 
 Wherever plenty of water for keeping the condenser cool is 
 available condensing engines are very frequently used, and 
 are especially useful when the engine is but lightly loaded. 
 By so getting rid of the steam, there is a very perceptible sav- 
 ing. From what has been said we are prepared to pass at 
 once to the practical discussion of every-day engines. 
 
 Following Rankine's general principle, we may begin at 
 once by saying that the ideal valve gear for an engine is that, 
 which can open instantaneously when steam is wanted, 
 admits it to the cylinder at full boiler pressure, and closes, 
 instantly, to avoid letting any steam in at a lower pressure 
 when it is desired to shut off the steam supply. When the 
 expansion of the steam is completed the exhaust valve should 
 open instantly and let the pressure down to that of the out- 
 side air, or the condenser, at once. From the diagram of the 
 engine just given you will see that these conditions are 
 but partially fulfilled ; for so long as the same ports are used 
 for admission and exhaust, and are controlled by a single 
 sliding valve, the engine is not always capable of proper 
 adjustment to the conditions under which it is working, 
 
 because the four functions performed by the single valve 
 
 the opening and closing of the supply ports and the opening 
 and closing of the exhaust ports are not independent of 
 each other, and, consequently, have to be so related as to pro- 
 duce the best average effect, not always the most desirable one 
 in a particular case. It was for the purpose of avoiding the 
 difficulties met in these single-valve engines that more com- 
 plicated valve gears have been introduced. The best type 
 of these is the Corliss, which has been used for over forty 
 years.
 
 CONCERNING PRIME MOVERS. 
 
 37 
 
 In this form of engine, shown in Figs. 15 and 16, four 
 separate valves are employed, two to admit the steam to the 
 
 FIG. 15. CORLISS VALVE MOTION. 
 
 cylinder and two others to let it out when it nas been used : 
 they operate by a slight turning motion about their axes, and 
 are usually in the form of slightly tapering cones, having the 
 steam ways cut as long longitudinal slots. It is apparent that 
 valves of this sort can be thrown wide open very quickly and 
 closed as speedily, and that they are capable of any sort of 
 independent adjustment ; the result is apparent in the supe- 
 rior economy that is obtained by such mechanisms. 
 
 ,'579414
 
 38 THE ELECTRIC RAILWAY. 
 
 The Corliss valve gear has various modifications, but all 
 retain the distinguishing characteristic of four independ- 
 ent valves separately adjustable. The admission valves are 
 opened by an eccentric, but are closed by gravity or, in later 
 designs, by vacuum pots. All four valves are driven (as 
 shown in Fig. 15) from a wrist plate pivoted on a pin project- 
 ing from the center of the cylinder this wrist plate is itself 
 actuated by the eccentric on the shaft, and has attached to it 
 four links connecting it with the four valves. The result of 
 this method of driving is that the valves are opened very 
 promptly, held open, and closed rapidly. The two upper 
 valves in the figure are the admission valves, and are closed 
 by the vertical rods seen attached to them, which pass into 
 the vacuum pots below the floor. The links leading to the 
 
 FIG. i6.-SECTioN OF CORLISS CYLINDER AND VALVES. 
 
 steam valves are provided with catches. When these are dis- 
 engaged the valves are free to close, and each catch takes 
 hold on its backward stroke ready to open its valve. The 
 rods A and B control the position of these catches, so that 
 the valve may be released and closed at any point in the 
 rtion of the piston up to nearly half the full stroke. These 
 rods are controlled by an ordinary centrifugal governor usu- 
 Uly belted to run much faster than the engine in order to 
 t sensitive. This Corliss valve gear is used only on 
 omparahvely low-speed engines, 140 revolutions per minute 
 less, because the vacuum pots or weights that- close the
 
 CONCERNING PRIME MOVERS. 39 
 
 valves do not work rapidly enough to operate at a much 
 higher speed. 
 
 For most engines running two or three hundred revolu- 
 tions per minute a simpler valve gear, much like that shown 
 in Fig. 14, is almost universally employed. This may 
 take many forms, most of them, however, retaining the 
 same general characteristic of a balanced valve simply and 
 strongly made, and controlling a pair of ports that serve both 
 for admission and exhaust. The particular diagram shown 
 refers to the Armington & Sims engine, which is used to a 
 very large extent in electric light and power work. But 
 something more than proper valve motion is necessary to 
 secure the best results from an engine. One of the most 
 serious losses met, particularly in engines driving electric 
 railway plants, is cylinder condensation, or loss of heat due 
 to the steam being cooled by the metallic portions of the cyl- 
 inder; the result is that a certain amount of heat which ought 
 to be utilized on the piston is wasted. 
 
 Considerable loss of this kind is almost certain to take 
 place where an engine is underloaded. The steam is admitted 
 at a high pressure and expanded altogether too much, during 
 which process a portion of it gets condensed. There is for 
 every steam pressure a particular amount of expansion that 
 gives the best results; if more than this be employed, the 
 engine generally suffers from condensation and excessive 
 cooling of the steam by radiation of heat through the sides 
 of the cylinder ; while if less expansion be used, the steam 
 is rejected while it still has a considerable amount of valu- 
 able energy which is thus lost. 
 
 When high steam pressures are to be used, as is most 
 desirable on the score of economy, it is best that the expan- 
 sion should not be carried out in a single cylinder, but con- 
 secutively in two or more. Engines so arranged are called 
 compound. The advantage of /this arrangement lies largely 
 in avoiding the losses incidental to changes of temperature 
 that take place during expansion ; the more nearly uniform the 
 temperature of the cylinder can be maintained, the less loss 
 from radiation and condensation. Where the temperature 
 range of steam in the cylinder is great, as in the case where
 
 4 THE ELECTRIC RAILWAY. 
 
 high initial pressures and high expansions are used, it is a diffi- 
 cult matter to keep the cylinder at a steady enough temperature 
 to allow economical expansion of the steam. If, however, 
 the expansion is distributed over two or more cylinders, only 
 one of them will communicate with the condenser, in itself a 
 ready source of cooling, and each cylinder can be kept at a 
 temperature much more favorable to escaping loss than if the 
 entire temperature range of expansion were in a single cylin- 
 der. Engines in which this useful plan is carried out are 
 coming into very extensive use. The compound engine, 
 either non-condensing or condensing, has found its way into 
 a large number of electric central stations, and the triple- 
 expansion engine, in which the work of the steam is still 
 further subdivided, is gradually displacing the compound 
 engine where large powers are required. The compound 
 construction is generally more advantageous in condensing 
 than in non-condensing engines, and under favorable circum- 
 stances will save something like 40 per cent, of the fuel that 
 would otherwise be required. 
 
 A very important part of every engine is the governor. In 
 its original form it consisted simply of a pair of balls sup- 
 ported by arms hinged to a vertical shaft put in rotation 
 'from the main shaft of the engine ; links from each ball led 
 to a collar capable of vertical play up and down the shaft. 
 This collar gave motion to a lever controlling the main steam 
 valve, and when, owing to the high speed of the engine, 
 the balls were thrown outward by centrifugal force, the col- 
 lar was raised and the steam partially cut off. The objec- 
 tions to this form of governor are numerous ; two of them 
 being lack of sensitiveness, owing to the considerable amount 
 of work that has to be done by the governor in operating a 
 steam valve, and the fact that engines equipped with this 
 throttling governor work at a fixed cut-off, that is, the steam 
 is always admitted to the cylinder for a definite part of the 
 stroke, but at a pressure depending on the amount the steam is 
 wire-drawn in passing the half -closed valve. By thus low- 
 ering the initial pressure the efficiency of the engine is decid- 
 edly diminished. One of the great improvements made in the 
 introduction of the Corliss valve gear was a successful way out
 
 CONCERNING PRIME MOVERS. 41 
 
 of both of these difficulties. In the first place, the work done 
 by the governor is very little, only sufficient to release the 
 catch on the link that works the admission valve and permit 
 the latter to close. This gives such sensitiveness that a 
 very slight change of speed "vill suffice to move the govern- 
 ing levers. The second advantage is that instead of admit- 
 ting steam to the cylinder for a fixed portion of the stroke, 
 but at varying pressure, the governing is accomplished by 
 changing the cut-off. In other words, when the load grows 
 heavy and the engine begins to slow down, the catch that 
 holds the admission valve is shifted so as to keep the valve 
 open during a greater portion of the stroke, thus furnishing 
 a higher average pressure during the stroke. The steam is 
 thus used at nearly the full boiler pressure and temperature, 
 but in amounts depending on the work to be done. 
 
 Although the Corliss form of governor is capable of very 
 excellent N work, and is a vast improvement over any of the 
 throttling governors, it still leaves, in its usual form, much to 
 be desired as regards keeping the engine absolutely constant 
 in speed. 
 
 We may as well state at once that so far as most elec- 
 tric railway service is considered an engine cannot be 
 kept at really constant speed by any device yet invented. 
 There are good and bad governors, but there is no perfect 
 governor. One of the reasons for this is the enormous rapid- 
 ity of the variations in load encountered in driving railway 
 generators for a small road. Quadrupling the load will slow 
 down any engine with any kind of governor. The decrease 
 in speed is, of course, only temporary, and in a few seconds 
 the engine recovers itself. But even with the most reliable 
 and sensitive forms of governor now constructed the moment- 
 ary slowing down is quite noticeable, although counted min- 
 ute by minute the speed may be almost uniform. In the 
 Corliss arrangement for governing the faults are two. First, 
 the speed of the engine in revolutions per minute is so slow 
 that opportunities to govern are too infrequent, for the drop 
 cut-off used in Corliss engines is seldom worked above 100 
 revolutions per minute, and if forced beyond this point it is 
 apt to fail in promptitude. Hence a very perceptible fraction
 
 4 2 . THE ELECTRIC RAILWAY. 
 
 of a second elapses between the action of the governor and the 
 next ensuing opportunity to shift the valve. Besides this, the 
 inertia of the governor itself causes some delay. The second 
 fault is one that is common to all fly-ball governors of the kind 
 described ; they act too uniformly. In other words, the force 
 tending to pull the balls downward is uniform, and the posi- 
 tion taken by the balls for any given speed of the engine is 
 always the same, so that there is continually a fixed relation 
 between the action of the governing balls and the position 
 of the valves. 
 
 Now, if instead of pulling the balls down by gravity a 
 force can be substituted for it that will vary with the change 
 in the position of the balls, so that the shifting of the valves 
 may go on rapidly until the engine comes to the required 
 speed, we shall have as the result a governor much more 
 sensitive than the ordinary form, especially as regards 
 changes in steam pressure or in load. On all modern high- 
 speed engines such a governor is in use. Most of the smaller 
 engines nowadays are high-speed machines, running from 
 two to four hundred revolutions per minute according to size. 
 The governor is, in such machines, usually placed where it 
 will act directly on the eccentric to change the throw of the 
 valves. Fig. 17 shows a governor of that class. Its opera- 
 tion will be seen at a glance. Two weighted arms pivoted 
 near the periphery of the fly-wheel are thrown outward by 
 centrifugal force, and their motion is resisted by a pair of 
 powerful spiral springs, acting in this case in tension ; the 
 motion of these arms is transmitted through a pair of levers, 
 clearly shown in the cut, to lugs on the eccentric, and tend 
 to shift the latter's position on the shaft, and thus vary the 
 valve motion. The engine can be conveniently adjusted for 
 various speeds by varying the tension of the springs and the 
 position of the weights on the arms, or both. These govern- 
 ors are much wider in range of action and far prompter in 
 motion than the fly-ball governors previously described. 
 
 vSome of them permit the admission of steam for over four- 
 fifths of the entire stroke, thus taking care of an overload 
 very effectively. We thus see that there are two quite dis- 
 tinct types of engine in use to-day for power purposes. In
 
 CONCERNING PRIME MOVERS. 43 
 
 the first place, engines with Corliss or equivalent valve gear, 
 having separate admission and exhaust valves and running 
 at a relatively low speed, seldom over 100 revolutions per 
 minute ; and, second, the great group of single-valve engines 
 in which the admission and exhaust are not independent of 
 each other, running at speeds usually over 200 revolutions 
 per minute, and furnished with fly-wheel governors of a 
 
 FIG. 17. GOVERNOR OF HIGH-SPEED ENGINE. 
 
 very effective character. The choice between these two kinds 
 of engine is somewhat difficult. There are a few excellent 
 transition forms between the types, but nearly every engine 
 with which the reader is likely to have to do belongs quite 
 distinctly either to one or the other. As will be shown later, 
 the slow-speed engines with the Corliss valve gear, or its 
 equivalent, utilize the steam rather more efficiently than any 
 single-valve engine has yet been able to do. On the other
 
 44 
 
 THE ELECTRIC RAILWAY. 
 
 hand, they suffer more from condensation than high-speed 
 engines of the same capacity for doing work, on account of 
 the greater size of cylinder necessary to compensate for the 
 decreased speed. They do not govern quite so well, and, as 
 a class, do not permit of admitting steam throughout a por- 
 tion of the stroke anywhere nearly as great as that which can 
 be reached with the other mechanisms. For electric-railway 
 service these are serious disadvantages, because the load is 
 apt in small roads to vary greatly and with great rapidity. 
 
 As the average load on an engine in an electric-railway 
 power plant is frequently only a quarter or a third of the 
 maximum load, the slow-speed engine is put somewhat at 
 a disadvantage, for if the engine is large enough for the 
 maximum power required to be developed within the avail- 
 able range of cut-off, the average power will be developed at 
 a cut-off much too short and an expansion of the steam con- 
 sequently far too great for the highest economy. 
 
 High-speed engines, with governors permitting the ad- 
 mission of the steam during a large part of the stroke, can 
 under great variations of load be worked on an average 
 much nearer the ratio of expansion that corresponds to the 
 greatest economy than their low-speed competitors. As a 
 class, however, they do not utilize the steam as efficiently. 
 If an engine has to be used in situations where the changes 
 of load are very violent, as in small electric railways, the 
 high-speed engine can usually be run more economically 
 than even the best Corliss engines. If the changes of load 
 are not very rapid and relatively rather small, the latter type 
 of engine will generally give the better result. Such condi- 
 tions are found now and then in large electric roads where 
 the number of cars is so great that the starts and stops of 
 individual cars are not important factors in change of load. 
 This subject will, however, be more thoroughly discussed in 
 the chapter on station design. 
 
 It is important that every man who is to use a steam engine 
 
 should understand thoroughly how the power it produces can 
 
 be measured. Ordinarily an engine is designated by its 
 
 available horse-power, but this practice really gives very lit- 
 
 e useful information about the machine, and at present
 
 CONCERNING PRIME MOVERS. 45 
 
 -makers either do not specify the horse-power, but give the 
 dimensions and speed, or in mentioning the horse-power 
 mention also the corresponding point of cut-off of the steam. 
 By one horse-power, of course, the reader will understand a 
 rate of doing work equivalent to 33,000 foot-pounds per min- 
 ute. This power is derived from the pressure on the piston 
 of the engine, and the following almost self-evident rule 
 gives the method of finding the horse-power of any given 
 engine. 
 
 Take the area of the piston in square inches, multiply it 
 by the number of feet that the piston travels each minute, 
 multiply this product by the mean pressure on the piston 
 during its stroke, divide the entire product by 33,000, and 
 the result will be the horse-power of the engine in question. 
 The area of the piston in square inches can be immediately 
 obtained from the bore of the cylinder as given by the maker, 
 and the number of feet of piston travel per minute from the 
 stroke and the number of revolutions. The only uncertain 
 and troublesome factor, then, is the mean pressure on the 
 piston. 
 
 If the steam entered the cylinder throughout the entire 
 stroke, the mean pressure against the piston would be very 
 nearly the pressure of the steam in the boiler, and the prob- 
 lem would reduce itself to a pusji equal to the number of 
 square inches in the piston multiplied by the pressure of the 
 steam per square inch, and working as many feet per minute 
 as the piston travels. But inasmuch as steam is always used 
 expansively, entering the cylinder at nearly boiler pressure 
 and leaving it at only a very slight pressure, the mean push 
 of steam against the piston is not so easy to find. It is evi- 
 dent, however, that if we could measure the pressure of the 
 steam at each point of the piston's travel, we could get the 
 mean pressure at once, for we should merely take the press- 
 ures at a sufficient number of points and average them. We 
 have fortunately a method of doing this and ascertaining at 
 the same time a vast variety of other valuable facts concern- 
 ing the performance of the steam engine. This is found in 
 the indicator, which is really nothing more nor less than an 
 instrument for registering the steam pressure against the
 
 4 6 THE ELECTRIC RAILWAY. 
 
 piston at each point of its path. So important an instrument 
 is the indicator that no engineer is fit to care for a modern 
 engine who does not understand its principle and use, and it 
 is worth while devoting some little space to the consideration 
 of the information it can give us. 
 
 But first let us follow roughly the actual cycle of operations 
 that goes on in the cylinder of a steam engine, beginning 
 with the moment when the valve that admits steam to the 
 cylinder opens. Let us suppose that we are possessed of an 
 eye quick enough to follow the fluctuations of a steam gauge 
 attached to one end of the cylinder. The first effect, of 
 course, is that steam at practically full boiler pressure rushes 
 in, and we should find the steam gauge darting up to that 
 pressure; then, as the piston started and the steam from the 
 boiler continued entering, the pressure would remain almost 
 exactly constant up to the point of the stroke where the 
 admission valve began to close. The pressure would then 
 fall, as the steam in expanding and pushing on the piston 
 loses its temperature and pressure together. We should 
 thus find the steam gauge sinking gradually down until very 
 near the end of the stroke. Then as the piston neared the 
 extreme limit of its motion the exhaust valve would open and 
 the pressure would fall immediately very nearly down to that 
 of the outside air. Meaiywhile the piston would be on its 
 backward stroke, driven by the steam at the other end of the 
 cylinder, and the pressure at the end we are investigating 
 would remain, so long as the exhaust port remained open, 
 very small in amount and almost uniform. Finally, as the 
 piston almost reached the point where it started the exhaust 
 port would close, and the small amount of steam left in the 
 cylinder would be slightly compressed, until at the end of the 
 stroke the admission valve would open again, and the press- 
 ure would rise as at first. 
 
 This is the very cycle of operations that goes on in either 
 end of the cylinder of a steam engine, and it is the duty 
 of the indicator to register it; the instrument, in fact, is 
 nothing more nor less than a registering steam gauge 
 arranged in such form as to be convenient for the purpose. 
 Fig. 1 8 shows an excellent modern indicator, and Fig. 19 a
 
 CONCERNING PRIME MOVERS. 
 
 47 
 
 sectional view showing its exact construction. It consists, 
 primarily, of a cylinder, seen at the left of Fig. 19, contain- 
 ing a snugly fitting piston, the upward motion of which is 
 resisted by a spiral spring; this forms a true steam gauge, as 
 the compression of the spring is exactly proportional to the 
 
 FIG. 18. THOMPSON INDICATOR. 
 
 FIG. 19. SKCTION OF THOMPSON INDICATOR. 
 
 pressure applied to it. If we screw this indicator cylinder 
 vertically into the end of a steam-engine cylinder, or to a 
 pipe communicating with it, and then allow the engine to 
 run, the position of the piston in the indicator cylinder will 
 depend on the pressure of the steam applied to the engine 
 piston, and at each moment of the stroke the compression of 
 the spring will be exactly proportional to the pressure of the 
 steam in the cylinder at that same instant. A little piston 
 rod passing through the top of this indicator cylinder moves 
 a lever carrying at its extremity a pencil point. The lever is 
 not a simple one, but a combination of levers so arranged 
 that the pencil shall travel vertically up and down instead of 
 in an arc. The point of this pencil rests, as seen from Fig. 18, 
 on the surface of another cylinder, over which, in practice, a 
 piece of paper is tightly stretched and held in place by the 
 pair of slender springs shown in the cut.
 
 48 THE ELECTRIC RAILWAY. 
 
 If, now, the steam be admitted to the engine, the pencil 
 point will fly up as far as the tension of the spring will allow 
 the indicator piston to push it, and as the steam expands 
 and the pressure falls the position of the pencil in its down- 
 ward path will be at every point exactly proportional to the 
 steam pressure in the cylinder. Now the cylinder is made 
 to revolve through nearly an entire revolution ; by the cord 
 shown it is attached to the piston rod of the engine, so that 
 as the piston moves the cylinder turns uniformly under the 
 point of the pencil, while a spring within keeps the cord taut 
 and takes up the slack on the back stroke, so as to bring the 
 cylinder exactly back to its starting-point. The effect of this 
 arrangement is as follows : The position of the pencil point 
 in a vertical line is always proportional to the pressure in the 
 steam cylinder ; its position with reference to the rotation of 
 the cylinder is always proportional to the distance through 
 which the piston of the engine has moved, so that the press- 
 ure in the steam cylinder at every point of the piston stroke 
 is registered upon the paper stretched on the revolving drum. 
 On taking off the paper and unrolling it, it is possible at once 
 to see exactly how the pressure has varied throughout the 
 stroke; and if the pressure required to produce a given com- 
 pression of the spring in the indicator be known, we can also 
 tell by the heights at various points of the indicator diagram 
 just the steam pressure in pounds per square inch that was 
 exerted on the piston. 
 
 With every indicator are furnished two or three springs 
 of strengths accurately known, and graduated so that each 
 inch of motion of the pencil point shall correspond to an 
 exact number of pounds pressure on the spring. A valve is 
 inserted between the indicator and the steam cylinder, so 
 that the piston of the instrument will not be in motion except 
 at the moment when it is desired to trace an indicator 
 diagram. As an added precaution against injury to the 
 indicator a little handle, shown at the top of the indicator 
 cylinder, enables the upper portion of it to be turned so as 
 to draw the pencil away from the paper except when it is 
 wanted. As the stroke of most engines is greater than the 
 periphery of the indicator drum, the latter is usually attached
 
 CONCERNING PRIME MOVERS. 49 
 
 to the piston rod, or its cross-head, through the medium of a 
 reducing motion of some sort, frequently a pendulum arrange- 
 ment, of which the lower extremity is operated by the cross- 
 head of the engine, while the cord to the indicator is attached 
 to an arc further up toward the center of motion. It should 
 always be remembered, however, in arranging anything of 
 this kind that it must be so put together that the motion im- 
 parted to the indicator drum shall be exactly proportional to 
 that of the cross-head. For details of the practical adjustment 
 of the indicator it is best to refer to one of the excellent trea- 
 tises upon it, or to the plain working directions that are fur- 
 nished by the makers of indicators, as an elaboration of this 
 sort of thing is hardly within the scope of the present chapter. 
 On admitting the steam to our indicator and pressing the 
 pencil down lightly against the drum during a single com- 
 plete stroke by means of the little handle before alluded 
 to, a tracing of the indicator diagram will be obtained, and 
 its general shape will be that shown in Fig. 20. In this 
 diagram the line F A represents the rise in pressure as the 
 admission valve is opened, and the line A B shows the 
 
 FIG. 20. SPECIMEN INDICATOR CARD. 
 
 period during which steam is freely admitted to the cylin- 
 der and the pressure kept constant; the smooth curve 
 B C shows the gradually falling pressure as the steam 
 expanded up to the point where, at C, the exhaust port be- 
 4
 
 5Q THE ELECTRIC RAILWAY. 
 
 gins to open; from C to D the pressure falls to a minimum, 
 and continues constant along the backward stroke up to E, 
 where the exhaust port closes, and the small amount of steam 
 that remains is slightly compressed until the opening of the 
 admission valve. The object of this compression is to cushion 
 the motion of the piston slightly, so as to prevent too violent 
 strains upon the engine. In taking an indicator diagram it 
 is customary to turn one of the valves of the instrument so as 
 to admit the outside air to the indicator cylinder before steam 
 is turned upon it; then, the drum being in motion, the pen- 
 cil point is swung against it, and a single line showing the 
 position of the spring for atmospheric pressure drawn near 
 the bottom of the paper. This furnishes a point of reference 
 both for showing the exact length of the indicator diagram 
 and the pressure of the steam above that of the atmosphere. 
 
 Now, having the indicator card taken, the first use to which 
 we can put it is the determination of the average pressure of 
 steam on the piston for the purpose of finding the horse- 
 power by the process indicated previously. 
 
 The simplest way of obtaining this mean pressure is to erect 
 at right angles to the atmospheric line ten pencil lines at 
 equal distances from each other, one being at each end of the 
 diagram, then measure the lengths of each of these, add 
 them together and divide by 10; then the quotient in 
 inches represents the mean pressure on the piston reckoned 
 on the scale of the spring that is used. If the indicator 
 spring gives a motion of one inch at the pencil point for fifty 
 pounds, and the mean of the vertical heights be one inch, 
 fifty pounds will be the mean pressure per square inch which 
 has to be used in computing the horse-power. A far more 
 accurate way of obtaining the true mean pressure is to 
 measure the area of the entire diagram by means of a plani- 
 meter, a little instrument made especially for the purpose; 
 then divide the area in inches by the length in inches, and 
 the mean pressure is at once found. The planimeter is not 
 always at hand, however, and the other method gives a toler- 
 able approximation. 
 
 In the next place, suppose the piston of the engine does 
 it well and the steam leaks past it: the result will be,
 
 CONCERNING PRIME MOVERS. 51 
 
 evidently, that the pressure behind the piston will lessen 
 very rapidly, and consequently on the indicator diagram the 
 line B C will fall toward the atmospheric line much more 
 rapidly than in the figure given. If, on the contrary, the 
 piston is tight but the steam admission valve leaks, the press- 
 ure will not fall by any means rapidly enough. If the steam 
 is only admitted for a very small portion of the stroke, and 
 there is considerable condensation in the cylinder, the press- 
 ure will again fall too rapidly. 
 
 If the valves do not open and close at the proper time the 
 result will be at once seen in the indicator card. For exam- 
 ple : Fig. 2 1 shows a card taken from a well-built modern 
 
 FIG. 21. INDICATOR CARD VALVES WRONGLY SET. 
 
 engine in which the valves were not set correctly ; the vertical 
 line at the left shows on the scale used for the diagram the 
 full boiler pressure. The admission valve, as will be seen 
 from the sluggish rise of pressure, only opens after the piston 
 lias started on its trip ; the steam pressure never catches up 
 effectively with the piston, and steam does not enter the cyl- 
 inder as freely as it should, although after the valve is wide 
 open the pressure has risen a little. The line showing the 
 expansion of the steam does not look promising, and the 
 pointed toe of the diagram shows that the exhaust port was 
 very late in opening, for the pressure did not fall anywhere 
 near the atmospheric pressure until the piston was well 
 started on its backward stroke. Besides opening too late, 
 the exhaust port closed too late, so that there was not the 
 slightest cushioning at the point where the admission valve
 
 52 THE ELECTRIC RAILWAY. 
 
 began its rather slow opening motion. Fig. 22 shows the 
 same diagram after the valves had been reset by reference 
 
 FIG. 22. INDICATOR CARD VALVES READJUSTED. 
 
 to the indicator card. You will see that in the first place the 
 boiler pressure is utilized much more fully, the expansion 
 gives a much smoother curve, and the steam is expanded 
 far more nearly down to the atmospheric line. 
 
 The exhaust is apparently not quite free enough, but the 
 exhaust valve opens and closes earlier, so that there is a 
 slight cushioning at the end of the stroke, and when the 
 
 FIG. 23-CARD FROM COMPOUND CONDENSING ENGINP, HIGH-PRESSURE CYLINDER. 
 
 admission valve opens it moves rapidly and opens fully, so 
 that the steam pressure rises sharply and is fairly well main- 
 tained. Figs. 23 and 24 show two very perfect indicator dia-
 
 CONCERNING PRIME MOVERS. 53 
 
 grams taken from a compound Corliss pumping engine. 
 Fig. 23 is the high-pressure diagram and Fig. 24 the low- 
 pressure. In the latter, of course, the atmospheric line cuts 
 through the diagram as the engine was condensing, and the 
 pressure consequently fell below the atmospheric pressure. 
 In the former of these diagrams the proper working of the 
 valves is very elegantly shown. The steam pressure rises 
 almost up to full boiler pressure instantly, and is maintained 
 almost absolutely constant until the admission valve closes. 
 The expansion line of the steam is almost perfect, and when, 
 just at the end of the stroke, the exhaust valve opens, it 
 opens fully and drops the pressure to a nearly constant 
 
 FIG. 24. CARD FROM COMPOUND CONDENSING ENGINE, LOW-PRESSURE CYLINDER. 
 
 amount, where it remains until the exhaust port closes, and 
 produces a slight cushioning at the end of the stroke. 
 
 The diagram from the low-pressure shows almost the same 
 beautiful characteristics. The admission line is vertical and 
 turns sharply at the top into the steam line. The steam 
 admitted, however, falls off somewhat in pressure, inasmuch 
 as it comes not from the boiler, but from the high-pressure 
 cylinder. The expansion line is as perfect as in the previous 
 diagram, and the exhaust works with the same beautiful 
 promptness. Altogether, these are model diagrams. The 
 engine from which they were taken has a high-pressure cyl- 
 inder 1 5 inches in diameter and a low-pressure 30 inches in 
 diameter, both of 3O-inch stroke. The high-pressure cylinder 
 takes steam at 125 pounds. The engine makes 46 revolu-
 
 THE ELECTRIC RAILWAY. 
 
 tions per minute, and the consumption of coal is but 1.8 
 pounds per horse-power per hour, a result very seldom 
 approached in stationary engines of this size. Pigs. 2 5 and 26 
 show a pair of excellent cards taken from an Armington & 
 Sims engine with a single piston valve, such as was shown 
 
 FIG. 25. CARD FROM SINGLE-VALVE ENGINE. HEAVY LOAD. , 
 
 in the early part of this chapter. The present diagrams 
 show perhaps as good results as can be obtained from this 
 class of valves in actual practice. In this case the admission 
 line of the steam is vertical, and runs almost up to full boiler 
 pressure; the expansion curve will be seen to be somewhat 
 waved, owing possibly to the slight irregularity of motion 
 in the pencil of the indicator, from the very high speed at 
 which the apparatus is worked. 
 
 The cylinder was 9 1-2 by 12 inches, the speed 200 revolu- 
 
 FIG. 26. CARD FROM SINGLE-VALVE ENGINE. LIGHT LOAD, 
 
 tions per minute, with steam at 85 pounds. The noticeable 
 feature of the cards from this and nearly all other high- 
 speed engines is the relatively large amount of compres- 
 sion, the exhaust port closing rather early in the stroke;
 
 CONCERNING PRIME MOVERS. 55 
 
 especially is this true when, as in Fig. 26, the load on the 
 engine is exceedingly light, hardly more than the friction of 
 the moving parts. In this case the compression begins 
 almost at half stroke, the cut-off of the steam admitted being 
 very early in the stroke. These cards are most excellent for 
 a high-speed engine, but do not show the economy of steam 
 of the pair of cards just given. The expansion of the steam 
 and the operation of the exhaust valve fail to give the good 
 result that is obtained by the use of separate admission and 
 exhaust valves. 
 
 Fig. 27 shows an indicator card of a somewhat less desir- 
 able kind taken from an engine actually engaged in driving 
 an electric-railway plant. In this machine the valves were 
 not adjusted anywhere nearly as well as in the example just 
 given. The compression begins very early in the stroke and 
 runs insensibly into the admission line ; the steam line falls 
 
 FIG. 27. CARD FROM SINGLE-VALVE ENGINE IN RAILWAY POWER STATION. 
 
 off in a curve, leaving the point where the admission valve 
 closed rather indefinite. The expansion curve is bad and 
 the opening of the exhaust valve rather gradual. All this 
 means lack of economy, for neither is steam admitted at as 
 high an average pressure as is desirable, nor is it expanded 
 properly before the exhaust begins to open ; besides this, the 
 compression is much larger than is ordinarily desirable. 
 
 High-speed engines seldom give indicator diagrams show- 
 ing anything like the efficiency found in using some other 
 forms of valve gear; nevertheless, the practice in this respect 
 is improving, and some of the modern engines give very 
 good results. At all events, the difference between the two 
 classes is not nearly so formidable as it once was. The weak 
 point of the low-speed engine has already been mentioned 
 the relatively large cylinder and consequent internal waste, 
 lack of range of cut-off, and sluggish governing. The weak
 
 5 6 THE ELECTRIC RAILWAY. 
 
 point of the high-speed engine with its single valve is that 
 the steam is not used as effectively because of the inter- 
 dependence of all the movements of the valve. If the cut-off 
 is changed the action of the admission and exhaust is changed 
 throughout, with the result that in most cases the steam is 
 not admitted freely up to full boiler pressure, the exhaust is 
 apt to take place at the wrong time, and there is often an 
 objectionable amount of compression. All these mean that 
 the steam is not used in the way to secure the maximum 
 efficiency, and the difference, rather evident from an inspec- 
 tion of the indicator cards, is often seen in the respective 
 coal bills met in using these two types of machine. 
 
 When economically run, the modern high-speed engine 
 will produce one horse-power per hour with a trifle over 30 
 pounds of steam, seldom or never below that figure. An 
 engine fitted with Corliss or equivalent valve gear, and of 
 similar size, will, under favorable conditions, do the same work 
 with the expenditure of 27 to 30 pounds of steam. Other 
 things being equal, large engines are rather more economical 
 than small ones. If compound engines are employed, used 
 non-condensing, the consumption of water per horse-power 
 hour is usually from 22 to 26 pounds; if used condens- 
 ing, from 1 6 to 20. Condensation in either simple or com- 
 pound engines is likely to save something like 1 5 per cent, of 
 the fuel. Very large triple-expansion condensing engines may 
 produce the horse-power hour on as low as 12 or 13 pounds 
 of steam, but such figures are rather unusual. These data 
 can, of course, be regarded only as approximations, as the 
 results obtained are influenced by the care exercised in set- 
 ting the valves of the engine to do the most economical work. 
 For example, in Figs. 21 and 22 the changes made in the 
 valve motion reduced the amount of fuel per day from 16 
 tons to 12 tons for the same work, which of itself is a suffi- 
 cient lesson in the importance of using the indicator and 
 working the engine in the very best manner of which it is 
 capable. 
 
 In choosing an engine for operating an electric-railway 
 power station very many things have to be taken into consid- 
 eration ; the general principle to follow, however, is to em-
 
 CONCERNING PRIME MOVERS. 57 
 
 ploy an engine of such size and character that it can be worked 
 with its average load at nearly its maximum point of economy. 
 With large roads, where the load does not vary very much, 
 these conditions can be fulfilled admirably by large Corliss or 
 similar engines, either simple or compound. If, however, the 
 load is likely to vary excessively, and the maximum load will 
 probably be three or four times the average load, such 
 engines if used will be so seriously underloaded as to be really 
 less economical than the high-speed engines, which, although 
 intrinsically less effective in their use of steam, more than 
 make up for this loss in their greater range of power and 
 consequent ability to handle heavy loads on occasion, while 
 on the average working near their point of maximum effi- 
 ciency. This principle will be elaborated further in the chap- 
 ter on station design. 
 
 A very large portion of this chapter has been devoted to 
 the steam engine because it is at present, and is likely to be 
 for some time to come, the principal prime mover for elec- 
 trical power stations, especially electric railways. Gas 
 engines, which have been before alluded to, and water- 
 wheels, of which we will now speak, are occasionally used for 
 such purposes. The difficulty of governing under a very 
 variable load is so great that the advantage of cheap power 
 rendered available by these means is materially lessened. In 
 speaking of water-wheels, text-books and popular treatises 
 generally begin by a formal mention of overshot and breast 
 wheels, both vertical wheels on which the water acts partly 
 by weight and partly by impulse, and finally give some brief 
 mention of turbine water-wheels. At the present time the 
 latter constitute the only class worth the slightest discussion 
 for our particular purposes, with the exception of a peculiar 
 form of impulse wheel frequently used for enormously high 
 heads of water. 
 
 The overshot, breast and undershot wheels are almost 
 as obsolete to-day as single-acting engines and steam 
 pressures of 1 5 pounds per square inch. The turbine is a 
 water-wheel generally, but not always, arranged on a ver- 
 tical axis, and receiving and discharging water in various 
 directions around its circumference. It differs from all other
 
 THE ELECTRIC RAILWAY. 
 
 water-wheels in that all the vanes or paddles provided are 
 acted on simultaneously instead of portions of the periphery 
 being used in succession. The wheel consists in general of 
 a drum or annular passage containing a set of curved vanes 
 shaped in such a manner that the water, after glancing from 
 them, is left behind with a minimum amount of energy. 
 They have the great advantage of being of very small bulk 
 for their power and being equally efficient for quite high and 
 low heads of water. On such wheels the water acts by vir- 
 tue of its weight and velocity, and the supply is received 
 either directly from a reservoir, in which case the wheel is 
 placed close to an opening in its bottom, or through a supply 
 pipe and wheel case. The former method, involving, as it 
 does, no loss from friction in pipes, is preferable for low falls ; 
 the latter for considerable heads. In most turbines the open- 
 
 FIG. 28. PARALLEL-FLOW TURBINE. 
 
 FIG. 29. ARRANGEMENT OF GUIDE VANES. 
 PARALLEL-FLOW TURBINE. 
 
 ing through which the water rushes upon the wheel is fur- 
 nished with guide blades to give the stream the most effective 
 direction. Turbine water-wheels are generally divided into 
 three classes, according to the manner in which the water 
 acts upon them. First, the parallel-flow turbines, in which the 
 water is supplied and discharged in a current parallel to the 
 axis of the wheel; second, outward-fiow turbines, in which 
 the water is supplied and discharged in currents radiating 
 from the axis; and, third, inward-flow turbines, in which the 
 water is supplied at the periphery and discharged through 
 the center. To these may be added the modern class of hori- 
 zontal turbines, in which the turbine principle is used in 
 Is upon a horizontal shaft, thus avoiding the necessity of 
 -earing. These may belong to any one of the three classes 
 above mentioned. Turbines frequently work "drowned,"
 
 CONCERNING PRIME MOVERS. 59 
 
 that is, entirely submerged; but inasmuch as some efficiency 
 is lost in this way, the practice is not to be generally rec- 
 ommended. 
 
 Fig. 28 gives a diagrammatic view of a parallel-flow tur- 
 bine, and the accompanying Fig. 29 shows the way in which 
 the guide blades and vanes are arranged to produce the proper 
 reaction. The water enters the guide blades around the 
 periphery of the casing, and from these passes to the wheel 
 itself, putting it in powerful rotation. The drum A is a sup- 
 ply chamber, containing the guide blades; B is the wheel 
 proper. Fig. 30 shows a horizontal portion of an outward- 
 flow turbine wheel ; A is a supply chamber in the interior, 
 being in the form of a vertical cylinder, with a ring of open- 
 ings around its lower end ; C shows the guide blades corres- 
 ponding to those in the figures previously shown, and D the 
 
 FIG. 30. DIAGRAM OF OUTWARD-FLOW FIG. 31. DIAGRAM OF INWARD-FLOW 
 
 TURBINE. TURBINE. 
 
 buckets of the wheel. Passing to the inward-flow turbine, 
 Fig- 3 i gives a similar diagram for it. The water is led, as 
 in the outward-flow turbine, by a vertical chamber; B is the 
 wheel, provided with vanes D D and free to discharge its 
 water through the centre ; C is one of the guide blades to 
 direct the water against the vanes. Fig. 32 gives a view 
 of a horizontal inward-flow turbine, showing how these gen- 
 eral principles are applied in a modern wheel often used for 
 electrical purposes. The efficiency of all these wheels is 
 about the same, and. equals say 80 per cent, of the theoretical 
 energy of the water as it strikes the wheel. No simple rules 
 can be given for computing the water-power of a given fall, 
 as the data regarding the velocity and amount of water can- 
 not be readily ascertained without a careful survey, which 
 should be made by a qualified engineer. The general prin-
 
 60 THE ELECTRIC RAILWAY. 
 
 ciple of supply, however, is to utilize as high a fall as prac- 
 ticable with as little intervening pipe as possible, 
 horizontal turbine is rather a favorite on account of the con- 
 venient position of the main shaft. Turbine water-wheels 
 not only are enormously powerful in proportion to their bulk, 
 but have a very high velocity of rotation ; so great, in fact, 
 that in some large electric stations the dynamos are belted 
 direct to the water-wheels, but furnished with larger pulleys, 
 the speed of the wheels being actually greater than that desired. 
 Where water-power can be cheaply had it is generally 
 
 FIG. 32. A TYPICAL HORIZONTAL TURBINE (LEFFEL FORM). 
 
 advantageous to use it, and in certain localities, where fuel 
 is very dear, the use of water is almost imperative; but 
 it must not be forgotten that there are circumstances under 
 which no advantage can be gained over steam. It is prob- 
 able that the water-wheel will be used more, rather than less, 
 for electrical power purposes, as there are many parts of the 
 country where fuel is particularly expensive, and by the 
 transmission of power for a few miles, water-falls can be 
 utilized with the greatest advantage. With the electro-motive 
 force at present used for railway work, the distance of prac- 
 tical transmission is somewhat limited ; but there are cases in 
 which it might even be worth while to supply power from a 
 distant dynamo at a very high potential for driving a railway
 
 CONCERNING PRIME MOVERS. 
 
 Cl 
 
 power station at the usual electro-motive force. The prin- 
 cipal disadvantage of the water-wheel is the difficulty of gov- 
 erning it. This difficulty is real and great, and arises from 
 the large mass of falling water that has to be controlled, and 
 the consequent mass and inertia of the gates. The result of 
 this is that a water-power governor lacks sensitiveness, and 
 it is only with great difficulty that the wheel can be held at 
 constant speed. This is of course a serious objection for 
 electric uses, and more especially so where the load, as in 
 electric railways, may vary with great rapidity. All the 
 disadvantages of bad governing in engines are exaggerated 
 when these are replaced by water- wheels, because the gov- 
 ernor does not act so quickly nor is it so effective a regulator 
 
 FIG. 33.-FOOTE SPEED REGULATOR. 
 
 as the isochronous governor of a steam engine. Neverthe- 
 less, on occasion, good work can be done with water-wheels. 
 
 There are ingenious electrical devices for controlling the 
 flow of water at a distant point by a governor at the water- 
 wheel ; and there is at least one speed regulator that obviates 
 the necessity for a sensitive governor at the wheel by regu- 
 lating the speed of the driven dynamo, irrespective, within 
 certain limits, of the velocity of rotation of the driving pulley. 
 
 The Foote speed regulator, shown in Fig. 33 is an apparatus 
 for accomplishing this end that has met with considerable 
 favor in electric light and power stations. The prime mover 
 is belted to a pulley, shown on the left-hand end of the
 
 6 2 THE ELECTRIC RAILWAY. 
 
 regulator, while the dynamo is belted to the larger pulley at 
 the other end. The condition for operation is that the driving- 
 shaft shall have a speed greater at all times than that of the 
 driven shaft. The large central drum has around its periph- 
 ery friction brakes adjusted by the centrifugal governor. 
 The brake shoes are of sheet steel, with a layer of paper pads 
 that fit closely on the surface of the friction pulley ; they are 
 pressed down upon it by means of the levers shown, and 
 actuated by the coiled spring on the shaft just at the right of 
 the drum. The tension can be changed by means of a screw 
 collar, and the controlling device is the centrifugal governor, 
 the pressure of which counteracts that of the spring when the 
 machine is in motion. When the driving pulley is running 
 faster than the required speed the arms of the automatic gov- 
 ernor are thrown out by the centrifugal governor, and par- 
 tially release the pressure of the brake shoes on the friction 
 wheel, so that enough slipping ensues to keep the driven 
 shaft at its proper speed. If the load on the dynamo varies 
 enough to cause the slightest decrease of speed the governing 
 weights fall toward the shaft and the brake shoes gripe the 
 friction pulley more closely. Of course such a regulator is 
 at its best when the variations of speed are not great, and 
 it is in successful use in a number of water-power elec- 
 tric light and power stations. For railway w r ork such an 
 apparatus may sometimes prove desirable, and would doubt- 
 less be effective in keeping the speed nearly constant unless 
 the changes of load were exceptionally violent. 
 
 Before closing this brief mention of water-wheels, one 
 should be mentioned that has made a remarkable name for 
 itself in service under certain peculiar conditions. This is 
 the Pelton water-wheel shown in Fig. 34. It is purely a jet 
 wheel, acting by the tremendous impulse of a water jet under 
 a powerful head. The buckets, as will be seen at a glance, 
 are numerous, small, and doubly concave, sweeping gradually 
 away from a ridge in the center; this is for the purpose of 
 allowing the water jet to divide and move smoothly off to the 
 sides instead of spattering, as would be the case if it impinged 
 on a flat or smoothly concave vane. These wheels are made 
 in a single massive casting, with buckets bolted firmly to the
 
 CONCERNING PRIME MOVERS. 63 
 
 periphery, and have proved very successful where an im- 
 mense head of water is to be utilized. They are not employed 
 for replacing turbines under ordinary conditions, but where a 
 small quantity of water under a very large head is available 
 the Pelton wheel is almost the only one that has in practice 
 given a thoroughly good account of itself. It is in use already 
 in a number of power-transmission plants in and about the 
 Rocky Mountains and on the Pacific coast, and replaces the 
 turbine with admirable effect when the head of water amounts 
 
 FIG. 34. PELTON WATER-WHEEL. 
 
 to a hundred feet or more. Its efficiency is quite equal to 
 that of the best overshot or turbine wheels. 
 
 Wherever good water-power is available it is certainly well 
 worth investigation as a source of power for electrical enter- 
 prises in the vicinity. To decide on its ability to compete 
 successfully with steam may not always be easy, but it is 
 worth while to make careful estimates whenever there is 
 a reasonable chance of good results, from the use of water- 
 power. Under favorable circumstances its advantage of 
 great cheapness is obvious ; its principal disadvantage is the 
 difficulty of proper governing, although this can be some- 
 what reduced by proper design of the station and by govern- 
 ing appliances such as those mentioned.
 
 CHAPTER III. 
 
 MOTORS AND CAR EQUIPMENT. 
 
 To drive a shaft when speed and load are desired to vary 
 both widely and independently is the work of the electric- 
 railway motor. 
 
 As to variation of speed, it has already been explained that 
 in the design of a dynamo electric machine, or in the com- 
 parison of one motor with another, variation in the number 
 of armature loops must be considered as affecting the speed, 
 other things being equal. But in studying the action of any 
 particular motor, when finally in operation, this cause of 
 speed variation may be neglected, since the armature, once 
 wound, carries in the magnetic field at all times the same 
 number of conductors.* 
 
 The other quantities, changes in which are connected with 
 changes in speed, are (i) strength of field, (2) the rate of 
 work, i.e., the quantity of work done in a given time. As 
 has been already shown, the rate of work is measured by the 
 product of (3) the current, and (4) the counter electromotive 
 force. The current, however, is readily expressed in terms 
 of the counter E. M. F., the resistance of the machine, and 
 the applied electromotive force, since it is always such a cur- 
 rent as will flow over the given resistance under a pressure 
 equal to the difference between the two opposing pressures, 
 - that applied or impressed by the dynamo through the line, 
 and that generated by the motor armature itself. (5) As 
 seen from the above, change in the applied E. M. F. is also 
 connected with change in the speed. The quantities (i), ^2), 
 (3), (4) and (5) are interdependent, but are separately men- 
 
 * It is of course possible to devise mechanism such that, in connection with the com- 
 mutator, it would afford a means of changing at will the number of active loops, but, 
 thus far. practice has not justified such a complication. We may, therefore, in examin- 
 ing the action of a railway motor, take the number of armature wires moving in the field 
 as constant. 
 
 64
 
 MOTORS AND CAR EQUIPMENT. 65 
 
 tioned, since convenience requires reference first to one, then 
 to another. 
 
 Of these five quantities, perhaps that which in practice is 
 most constant is strength of field. As has been shown 
 (Chapter I . ) , the efficiency of a motor depends upon the relation 
 of the counter E. M. F. to the applied E. M. F. High effi- 
 ciency, or relatively high counter E. M. F., is desirable at 
 any and all speeds. But high counter E. M. F. goes with a 
 large product of the three factors, (a) number of armature 
 loops, (b] speed of rotation, and (r) strength of field. As 
 noted, (a] the number of loops is fixed; (b} the speed is lim- 
 ited by practical requirements, hence (c) great strength of 
 field is constantly desirable. It is therefore good practice so 
 to wind a street-railway motor, by putting a relatively large 
 number of turns around its magnets, that a maximum 
 strength of field is attained, even when the current flowing 
 is small as compared with the maximum current. 
 
 This is equivalent to saying that, the magnetization given 
 by a relatively small current is yet sufficient to saturate or 
 nearly saturate the iron of the magnetic circuit. If, however, 
 the field be kept below saturation and be varied in strength, 
 this variation may be used to accomplish a certain degree of 
 speed regulation. Let us suppose a car moving on a level 
 (or on a uniform grade), and at such a speed as to produce a 
 counter E. M. F. of 400, the applied E. M. F. being 500. 
 For convenience assume the internal resistance of the arma- 
 ture circuit to be 10 ohms. Then the current flowing will be 
 
 lgg-400 = ^ZI^: = L = I0 amperes. The mechanical 
 10 R 10 
 
 work done would be 400 X 10 = ExC = 4,000 watts. 
 
 Now suppose it is desired to run more slowly. Increase 
 the field strength by 10 per cent. The counter E. M. F., at 
 
 5 oo 440 
 the same speed, would be 440 volts, the current = 
 
 6 amperes, and the work 440 X 6 = 2 ,640 watts. But since the 
 car requires 4,000 watts to maintain the previous speed, it is 
 evident that it must now decrease its speed until there shall 
 be an equality between the work required to maintain the 
 lower speed and the work done by the motor at the lower 
 5
 
 66 THE ELECTRIC RAILWAY. 
 
 speed due to the greater field strength. Let us learn what 
 this speed is. 
 
 It was seen above that work at the rate of 4,000 watts = 
 
 4' 000 =5.36 horse-power = 176,880 foot-pounds per minute, 
 746 
 
 must be performed in order to maintain the speed existing 
 before the change of field strength. Suppose the car to 
 weigh eight tons, and suppose that on the particular track in 
 question a horizontal effort of 25 pounds per ton is required 
 to overcome all resistances, including those of gears or other 
 mechanism between the armature and the axles; then, a total 
 horizontal effort of 200 pounds must have been exerted. This 
 quantity, multiplied by the number of feet traveled per 
 minute, must be equal to the number 176,880, representing 
 the total foot-pounds of energy utilized per minute. Hence, 
 
 176,880 
 the travel per minute = = 884.4 feet = 10.05 miles 
 
 per hour. At the new speed we must have a similar rela- 
 tion- 
 Work done in foot-pounds 
 
 ^^ = feet traveled per minute ; or r 
 
 since i watt-minute = 44.24 foot-pounds 
 
 Work in watts 
 
 7oo 44 24 = ^ eet trave l e d per minute ; whence, the 
 
 work in watts = 4.52 X feet traveled per minute. 
 
 The number of watts may always be expressed as the 
 product of the counter E. M. F. by the current, or thus, 
 watts = E' C'. 
 
 Let Z' = strength of field before change. Then, 
 i.i Z'=strength of field after change. 
 Let S'= speed of armature before change. 
 S"= speed of armature after change. 
 C' = current in armature before change. 
 C"= current in armature after change. 
 '= counter E. M. F. before change. 
 E"= counter E. M. F. after change. 
 
 X S'. (We may omit the number of armature 
 
 urns.) The speed of armature has a fixed ratio to speed of 
 
 ' m all practice of to-day. Hence, we may now assume
 
 MOTORS AND CAR EQUIPMENT. 67 
 
 some convenient ratio, such that; for example, the speed of 
 car above calculated, i.e., 884.4 feet per minute, shall cor- 
 respond to 1,768.8 revolutions per minute of the armature (i 
 revolution = 0.5 of a foot travel). Then we may write E' = 
 Z' X i , 768 . 8 . Also the work in watts = 4.52x0. 5 x revolutions 
 per minute = 2.26 S'. 
 
 ^=10= SOP Q, 768.8 Z'^ C , = 5QQ LI Z'S* 
 
 10 10 
 
 Z'X 1,768.8 = 400. .\Z' = 0.226. 
 
 This is a purely relative expression for the strength of field, 
 resulting from the assumed values and ratios as given above. 
 Then i.i Z' = Z" = 0.2486. 
 
 (i) "--=0.248 x S", 
 (2 \ r" = SOP (-248X8") 
 
 10 
 
 (3) E"xC"=2.26xS". 
 Placing in (3) the values taken from (i) and (2), we have 
 
 (4) (0. 24 8XS")X ( -^ =^^ ) = 2.26X8'; 
 
 500 (0.248XS") 
 
 = Q.I I. 
 10 
 
 .-. (5)408.9 = 0.248x8". .-. S"=i,649. 
 
 i ,640 
 
 Feet per minute = = 824.5 = 9-4 miles per hour ap- 
 proximately. The counter E. M. F. will then be found thus: 
 E"= 0.248 X 1,649 = 408.9. Also, C"= = 9.11, 
 
 and the work in watts = 408.9X9.11 = 3,725. 
 
 In like manner, if the field be weakened 10 per cent., we 
 
 (500 0.203 X S"') 
 may find: (0.203x8'") X - - = 2.26x8"; 500 
 
 .203x8"= in. .-.S'"= 389 -r- .203 = 1,916. Feet per 
 minute = 958 = 10.9 miles per hour. Counter E. M. F. = 3$9, 
 current = n.i amperes, and the work in watts = 389 X 1 1. 1 
 *= 4,3i7- 
 
 It should be noted that these results, increase of speed with 
 decrease of magnetic strength, and rice rcrsa, will be reversed 
 if the changes be made from a state in which the counter 
 E. M. F. is less than one-half the initial E. M. F. When the
 
 68 THE ELECTRIC RAILWAY. 
 
 counter E. M. F. of any motor is just one-half the initial 
 E. M. F., the rate of work, that is, the output per unit of 
 time, is greater than at any other efficiency. Since this output 
 is proportional to the product of counter E. M. F. by the cur- 
 rent, and the current is proportional to the difference between 
 the fixed E. M. F. of the line and the counter E. M. F., we 
 have the output varying as the general expression E'x 
 (E _ E') , the resistance being supposed to be constant. Now, 
 the product of a number as expressed by E', and of another 
 number which is equal to the difference between the first 
 number and a second fixed number, will always be a maxi- 
 mum when the first number is just half the second. Suppose 
 that E 500, for all values of E'. 
 
 Take E'= 249, then the product = 249X251 = 62,499; 
 
 Take E'= 250, then the product = 250X250 = 62,500; 
 
 Take E' = 251, then the product = 251 X249 = 62,499; 
 and any greater departure in either direction from 250, as 
 the value for E', will be followed by a further departure 
 from the maximum value for the product. While E' is 
 greater than the difference between E and E' (that is, when 
 it is greater than one-half E), any change of absolute value 
 made in E' is less, proportionally, than the corresponding 
 change in the smaller quantity, E E'. Thus when E', as 
 above, was taken at 400, and the magnetic strength was in- 
 creased, E 7 also increased to 409, a change of 2 per cent., 
 tending to increase the product E' C', measuring the output. 
 But this change of 9 volts was necessarily followed by an 
 equal change in the difference E E', reducing it from 100 to 
 91, a reduction of 9 per cent., and as the current flowing is 
 directly determined by the quantity E E', its value must also 
 change by 9 per cent. Since, therefore, we have increased 
 one factor, E', by 2 per cent., and decreased the other, C', 
 by 9 per cent., the product representing the new output must 
 be less than the former. But if the original value of E' were 
 200, instead of 400, then the difference E E' must be 300, 
 and any addition to E' resulting from an increase of field 
 strength would increase this factor, E', in greater propor- 
 tion than it could decrease the other factor, which depends 
 on the larger number, 300. The net result would therefore
 
 MOTORS AND CAR EQUIPMENT. 69 
 
 be an increase in the product E" C", and the car would con- 
 sequently run faster. 
 
 The fact that the rate of work of a motor is a maximum 
 when its efficiency is 50 per cent, is an important one. A 
 motor which has been properly rated as a 1 5 horse-power 
 machine should do that work at, say, 90 per cent, efficiency, 
 since that is an easily attainable figure. It then receives 
 about 1 7 horse-power of electric energy to perform 1 5 horse- 
 power of mechanical work. It is possible to obtain a rate of 
 20 horse-power from such a machine, but only by lowering 
 its efficiency, say to 60 per cent., in which case about 33 
 horse-power of electric energy must be supplied for the 
 20 horse-power of work. 
 
 Having discussed the relation between changes of field 
 strength and of speed, we now pass to changes of load and 
 field strength. It is evident that if in the case above assumed 
 the weight of the car or condition of track or steepness of 
 grade be changed, then the horizontal effort, 200 pounds, 
 required to maintain uniform motion, will also change. Let 
 us suppose it to be increased. Then if nothing be done to 
 change the magnetic strength of field, or the impressed 
 E. M. F., there must be, in a series-wound motor, a decrease 
 of speed, and that whether the original efficiency be above or 
 below 50 per cent. In the former case, increase of output 
 (due to the lowering of the speed, and hence of the counter 
 E. M. F.) will meet the larger demand, which is itself less 
 than if the speed were maintained at the previous rate, and 
 an equilibrium will again be established, when E" C" = Kx 
 revolutions per minute. 
 
 In this, K is constant, corresponding to 2.26 in the pre- 
 vious calculations, and depending on the particular value 
 of the horizontal effort required. If, before change of load, 
 the motor be working at less than 50 per cent, efficiency, the 
 decrease of speed must be greater, since the equality between 
 supply of and demand for power can be re-established only 
 because the one decreases less rapidly than the other. Thus, 
 the supply, varying with the product, E' C', will tend to 
 decrease with decrease of E', which, in a constant field, will 
 vary directly as the speed; but decrease of E' means increase
 
 jr THE ELECTRIC RAILWAY. 
 
 of C', though this increase will be less proportionally than 
 the decrease of E', as was explained above when considering 
 field strength. The product of these two factors will then 
 evidently diminish less rapidly than in the direct ratio to the 
 speed. The demand for power will, on the other hand, 
 decrease in exact ratio to the speed, since we have supposed 
 the horizontal effort, after the first change considered, to 
 remain constant. Another change would produce another 
 lowering of speed, another adjustment between supply and 
 demand ; and this process may continue until the horizontal 
 effort becomes greater than the rolling friction between wheel 
 and rail, in which case the wheels will "skid;" or greater 
 than the motor with maximum current in the armature and 
 fields is capable of overcoming, in which case no motion 
 whatever will be produced. 
 
 If we suppose these changes of load to take place when the 
 field is not saturated, we may, of course, use the changes of 
 field to counteract, if so desired, this speed reduction which 
 would otherwise follow. Diminution of field strength would 
 tend to increase or decrease the speed, as the efficiency is 
 greater or less than 50 per cent. 
 
 The second important method of regulating speed under 
 varying loads is found in the variation of the E. M. F. 
 applied to the armature. From what has been said concern- 
 ing the measure of work done, it will be readily seen that, 
 other things being equal, decrease of applied E. M. F. must 
 be followed by decrease of output, and increase of E. M. F. 
 by increase of output. Let us again consider the case of a 
 motor developing 400 volts counter E. M. F., the applied 
 E. M. F. being 500 volts, the current being 10 amperes, and 
 the resistance to motion being uniform. If now we reduce 
 the applied E. M. F. (E) to 450 volts, and if we conceive 
 that for a moment the counter E. M. F. (E') remains at 400, 
 
 the current would be reduced to 5 amperes thus : 45 ~ 4 = s 
 
 10 
 
 Such a change would reduce the output to one-half its former 
 
 value. The speed would then necessarily drop to one-half, 
 
 unless this decrease of output be checked in some way. This 
 
 k would come through the fact that a decrease in speed
 
 MOTORS AND CAR EQUIPMENT. 7 1 
 
 would be followed by a corresponding decrease of E', which 
 would at once increase C'. To learn what the change in 
 speed would actually be, let us proceed as before. 
 
 Using the new value for E, and assuming other condition* 
 as before, we may write 
 
 or E'x(45o E') = 22.6x8. 
 
 We suppose now that the field will remain constant. There 
 will then be a very simple relation between E' and S, that is, 
 they will vary directly each with the other. It was proved 
 above, under the conditions assumed, that 
 
 S = 4.42 E'. Substituting above, we have 
 
 '(450 E') = 22.6X4-42E', 
 or 450 '=22.6x4.42. .. '=350.11. 
 Calling this for convenience '=350, we have, 
 
 C' = '- - = 10 amperes, as before. 
 
 Output = 350X10=3,500 watts. Speed of car, in feet per 
 minute = = 774 = 8.8 miles per hour. 
 
 The original speed, it will be remembered, was 10.05 miles 
 per hour. Now, let us suppose a much greater change in 
 applied E. M. F., say to 250 volts. Then 
 
 250 E'= 22.6x4-42 = 99.9. .'. E'= 150. 
 
 250 _ 1 50 
 
 C'= - - = 10 amperes, as before. 
 10 
 
 Output = I50X 10 =1,500 watts. 
 
 Speed of car = = 332 feet per minute =3.8 miles 
 4.52 
 
 per hour. 
 
 Again, suppose E = 200 volts, then 
 
 200 E'=ioo (taken for 99.9). .-. E'= 100 volts. 
 
 200 100 
 C = = 10 amperes, as before. 
 
 Output = ioox 10 = i ,000 watts. 
 Speed = =22 1 feet per mint 
 Generally, we see that E' = E 100 =E C'R. R is a con- 
 
 Speed = = 2 2 1 feet per minute = 2.5 miles per hour.
 
 -, THE ELECTRIC RAILWAY. 
 
 stant by the construction of the machine (usually), and C', 
 under a constant load, also remains constant, while E varies, 
 as seen above. The output = E' C'= (E C' R) C' will then 
 vary in proportion to the difference (E C' R) : and this will 
 vary the more rapidly as E approaches the constant value C' 
 R. Thus, in dropping from 500 to 450, the change of 50 
 volts is applied to an original value of (E C' R)=45O 
 that is, a change of one-eighth. But in dropping from 250 
 to 200, the difference of 50 volts is applied to an original 
 value of (E C' R)= (250 ioo)= 150 that is, a change of 
 one-third. Looking again at the expression E C' R = E 
 100, we see that if E drops to 100 volts, the output will be 
 
 zero that is, the car will not move, unless the resistance to 
 
 motion be in some way diminished. 
 
 We may say then, generally, that in any motor, the 
 strength of field being fixed, a certain current, C, is required 
 to produce a certain torsional effect in the armature (or a 
 certain corresponding horizontal effort through the car axle). 
 To produce this current over the internal resistance of the 
 motor, R, a certain E. M. F. measured by C R is required. 
 If the impressed E. M. F. be just equal to C R, then the 
 rotating force in the armature and the resistance to rotation 
 will just counterbalance each other, and there will be no 
 motion. Excess of pressure over that measured by C R will 
 produce motion, and in the ratio of this excess. The in- 
 ternal resistance will cover that of the armature and field 
 windings, in case of a series motor, but only that of the 
 armature in the case of a shunt motor. During changes of 
 impressed E. M. F. the constancy of field strength supposed 
 above will follow in the case of a series motor, if the number 
 of turns and resistance of the field windings remain constant 
 (since we have seen that C will then be constant) . 
 
 Having seen the effects produced by changes in field 
 strength and in impressed E. M. F., we should now con- 
 sider the convenient and customary methods of producing 
 those changes. 
 
 Suppose three coils of wire to be wound around the mag- 
 net cores of a motor, as shown in Fig. 3 5 . Let the resistance of 
 each be one ohm. Connect D to B and E to C. Suppose a
 
 MOTORS AND CAR EQUIPMENT. 
 
 73 
 
 difference of potential of 100 volts be maintained between 
 the terminals A and F, then the total resistance of this field 
 circuit would be three ohms, the three coils being arranged 
 
 in series. The current flowing would be =33.3 amperes. 
 
 The number of turns would be two for each coil (one above, 
 one below) or six altogether. The magnetizing effect would 
 be measured by CxT =33.3x6 =199.8. 
 
 Now suppose the connections be changed by disconnecting 
 D and B, E and C, and connecting A, B, and C together, and 
 D, E, and F together. The resistance of the three coils, now 
 in multiple, will be only one-third of an ohm, or one-ninth 
 
 F c 
 
 FIG. 35. CONNECTIONS FOR COMMUTATING FIELD COILS. 
 
 the former value. The current will be nine times its value, 
 or 299.7. The number of turns which this whole current 
 makes around the magnet is only two, one above and one 
 below, or one-third the former number. The magnetizing 
 effect will be measured by 299.7x2 = 599.4, or three times 
 the former value. If the magneto-motive force, 199.8, were 
 itself sufficient to saturate the magnets, then the increase to 
 599.4 would be useless, so far as the strength of field is con- 
 cerned. But if a force of say 2,000 ampere turns be required 
 for saturation, then increase from 199.8 to 599.4 will t> e fol- 
 lowed by a proportional increase in the field strength. In 
 either case, the drop of potential required to force a given 
 current over the field windings will be much less, with the 
 ?econd arrangement of coils, in which the resistance is rela-
 
 74 THE ELECTRIC RAILWAY. 
 
 lively low, than with the first arrangement, in which it is 
 relatively high. If, therefore, these coils be placed in series 
 with the armature, we have a ready means of varying the E. 
 M. F. applied to the armature itself, and also of varying the 
 magneto-motive force applied to the field. The method of 
 regulation thus afforded is usually known as that by " com- 
 mutation of field circuits." 
 
 In its practical application, provision has been made for 
 other arrangements, intermediate in effect between the two 
 above described. Thus, (i) starting with the three coils in 
 series, (2) one may be cutout or short-circuited, the other 
 two remaining in series; (3) two may be placed in multiple 
 with respect to each other, the third being in series with this 
 multiple combination ; (4) the third may then be cut out or 
 short-circuited, leaving only the two in multiple; (5) the 
 arrangement of least resistance is of course that in which all 
 the coils are in multiple. If through any suitable switch the 
 various combinations be effected in the order above men- 
 tioned, the resistance of the field windings becomes less at 
 each successive step. Suppose a car moving at a constant 
 speed, the coils being in arrangement No. i, and the current 
 required being great enough to saturate the fields, even if 
 the number of turns were only one-third the number given 
 by this arrangement. If w r e now pass through No. 2, No. 3, 
 etc., successively, and there be no change of load, then the 
 E. M. F. applied to the armature w r ill increase, the speed will 
 correspondingly increase (as explained above) ; the current 
 will remain constant, for, by supposition, the field is contin- 
 uously saturated. The changes, then, do not affect it, and 
 affect the armature simply by varying the applied E. M. F., 
 as would be the case if the field circuits were independent, as 
 in a shunt motor, and some resistance external to the motor 
 were used to vary the drop of potential on the lines connect- 
 ing the dynamo and motor. The increase of output without 
 increase of current is due to a progression in motor efficiency. 
 In the first arrangement the resistance of the field circuit 
 (hence the waste of energy therein) is greater than would 
 be permissible in good design, were it not plainly set forth 
 as a means of regulation. Only in the last arrangement does
 
 MOTORS AND CAR EQUIPMENT. 75 
 
 the resistance become as low as good design would give were 
 efficiency alone in question. 
 
 If the current flowing before the changes be less than that 
 required for saturation with any number of 'turns less than 
 that given by the first arrangement, then the increase of 
 speed in motors in practice will be more rapid than in the 
 case just considered, and the current will not remain con- 
 stant, but will increase. For there are now two causes tend- 
 ing to increase the speed: first, the increase of applied 
 E. M. F., as before; second, there is decrease of the field 
 strength, due to the decrease of turns, which tends to in- 
 crease the speed (provided the efficiency of operation be 
 greater than 50 per cent.). 
 
 As explained, the increase of speed due to this cause 
 requires an increase of current. These changes tend, indeed, 
 to check each other. Thus, increase of current tends to 
 diminish the E. M. F. applied to the armature, just increased 
 by lowering resistance, and to strengthen the field, just 
 weakened by decrease of turns. To determine the constant 
 speed and current which would be established with arrange- 
 ment No. 2 or No. 3, etc., it would be necessary to know the 
 relation between the number of turns and the resistance of 
 the three coils. Assuming them to be equal in all respects, 
 the changes of turns and the resistances may be seen from 
 the following table : 
 
 ARRANGEMENT. TURNS. RESISTANCE. 
 
 1 6 3 
 
 2 4 2 
 
 3 4 i-5 
 
 4 2 0.5 
 
 5 2 0.3 
 
 In moving from step to step, the rate of change is evidently 
 not uniform, but is continuous in the direction of higher 
 speed. By making the coils unequal, it is possible more 
 equally to divide the total change effected from first to fifth, 
 but in the very limited space usually available for windings, 
 such inequality may require dangerously small cross-section 
 for the maximum currents required. 
 
 In the early motors of the Sprague Electric Railway and 
 Motor Company, rated at 7.5 horse-power, at ordinary street-
 
 THE ELECTRIC RAILWAY. 
 
 car speeds, say ten miles per hour, and for soo-volt circuits, 
 the turns and resistances were as follows : 
 
 Positions 2 and 3, and 5 and 6, are effectively 
 
 Switch No... i ; 2 and 3 ^4_^and_^;_7_ ! the same. 
 
 ^T".! I !. ; , i 3 .Proportional numbers, not absolute values. 
 
 Armature R = 1.4, constant. 
 
 R es ! 20 o ' 12 i 9.5 \ 5-0 i 3.2 (Actual R in ohms, including armature. 
 
 In a later machine, rated at 15 horse-power, under the 
 same conditions the field resistance of each motor was 8.42 
 ohms, total, made up of three coils of 2.16, 2.65, and 3.61 
 ohms, respectively, the number of turns being as 11.5, n, 
 and 15, respectively. 
 
 In using this method, the principal difficulty has been met 
 in disposing of the excessive heat necessarily generated in 
 the compact mass of field windings. Many ordinary forms 
 of insulation have been found to deteriorate under the influ- 
 ence of the high temperatures maintained. When, however, 
 the incidental difficulties have been met by use of proper 
 material, the method is a convenient one for securing a con- 
 siderable range of regulation. The mechanism required, 
 external to the motor, consists only of such a switch as will 
 in a simple manner produce the changes of connection, such 
 as described above. In the practice of the Edison Electric 
 Company (successors in this business to the Sprague Electric 
 Railway and Motor Company, who most widely introduced 
 the method considered), these changes of connection have 
 been produced by taking all the coil terminals to a series of 
 spring contacts, pressing against metal surfaces arranged on 
 a cylinder. These surfaces are so shaped that rotation of the 
 cylinder produces the desired combinations. It is of course 
 possible to use a smaller or larger number of independent 
 coils than three. In the former case, the number of com- 
 binations giving different effects is not sufficient for a smooth 
 control. In the latter, the complication of switches becomes 
 excessive. 
 
 The control by external resistance, thus lowering or raising 
 the E. M. F. applied to the motor, has been widely used and 
 scarcely needs extended description. The practical problem 
 has been to secure a convenient rheostat, made of such
 
 MOTORS AND CAR EQUIPMENT. 
 
 material as will give the high electric resistance required, 
 and at the same time withstand the heat produced. Open 
 coils of iron or German-silver wire seem to offer the most 
 ready means of attaining the end in view, but the space 
 demanded is found excessive as compared with the meagre 
 dimensions required by the condition that the whole mech- 
 anism shall go under a car floor. Plates of thin iron, bent 
 into a " U" shape, insulated from each other by mica, and 
 arranged in a semi-circular frame, have composed the rheo- 
 stats used by the Thomson-Houston Electric Company, which 
 has widely used this method of control. An arm moving 
 around the center of the semi-circular frame carries contact 
 pieces over the convex surface of the resistance plates, leaving 
 all, a part, or none of them in circuit, according as a low, or 
 higher, or maximum. E. M. F. is desired to be applied to the 
 motor. When this method is used, the field winding should 
 be permanently of low resistance, but sufficient to nearly satu- 
 rate the cores at all loads. 
 
 One of the two methods of control thus described, or a 
 combination of them, has been used in almost every effort 
 at electric-railway work. The chief exception is found in 
 storage-battery cars, in which the batteries themselves may 
 be thrown into various combinations, thus changing the 
 E. M. F. applied to the motors. There are few examples, if 
 any, of the exclusive use of either method for motors now 
 being manufactured. Thus, the Edison Company uses an 
 external resistance to prevent too great a rush of current into 
 the motors when starting ; their proper resistance, even with 
 all the coils in series, being so low when the two motors are 
 connected in multiple on a 5oo-volt circuit as to cause a cur- 
 rent of about 100 amperes to flow. This is usually much in 
 excess of that actually required for a comfortable start of the 
 car, and is of course highly objectionable in respect both to 
 its effect on the motor and its demand upon the station. 
 This "starting coil," as it has been called, is cut out of cir- 
 cuit as soon as the start has been made. The switch then pro- 
 duces the first combination of field coils, as above described, 
 and as it rotates further, produces the others successively 
 tnis commutation of the windings being the chief means of
 
 7 g THE ELECTRIC RAILWAY. 
 
 control. On the other hand, the Thomson-Houston Electric 
 Company, which has used the external rheostat as the 
 chief means of control, in the last position of the switch, 
 when the full line potential has been applied to the motors, 
 short-circuits or cuts out one of the two parts of the field 
 winding. Up to this last point these two have been con- 
 nected in series, acting, indeed, as one coil. Now, in order to 
 increase speed, a large part of the turns are made ineffective, 
 hence the field is weakened and the increase of current and 
 increase of speed take place, as already explained. The total 
 resistance of the rheostat is fixed at such a value as will allow 
 to flow the maximum current required for starting from rest, 
 the line potential being taken, as in practice, at 500 volts. 
 This total resistance may then be so subdivided as to produce 
 any desired rate of change. 
 
 In case two or more motors be used on one car or train, 
 there remains a third method of producing a considerable 
 range of control; namely, by changing the machines from 
 the multiple to the series arrangement, with respect to each 
 other, and vice -ccrsa. The applied E. M. F. is just one-half 
 in the second case what it is in the first, if there be two 
 motors, one-third if there be three, and so on. A variation 
 of this method might be had by changing the field windings, 
 leaving the armatures permanently in series or permanently 
 in multiple. The series arrangement of the two machines is 
 of special value in starting a car and in maintaining low 
 speeds. In both cases it may become the equivalent in effect, 
 and the superior, in economy, of either of the above-described 
 methods of control. Through this method it is indeed pos- 
 sible to make the same motors fulfill widely different service 
 conditions. Thus, suppose cars are to begin their trips in 
 the most populous portions of a city, where many stops and 
 relatively slow running are unavoidable, while, on reaching 
 the suburbs, long runs at high speeds are desired. We may 
 then design motors which, if placed in multiple and per- 
 mitted to run at say 2 5 miles per hour, will develop each 2 5 
 horse-power. If placed in series, they would each develop 
 12.5 horse-power, at 12.5 miles per hour, and at both speeds 
 their efficiency may be high. To meet speeds lower than 1 2 .5
 
 MOTORS AND CAR EQUIPMENT. 79 
 
 miles per hour, the load being such as would permit only 
 that speed if the impressed E. M. F. be 250 volts (half of 
 500 for each), we may resort to either of the two methods of 
 general control, the motors as a whole being kept in series. 
 So, for speeds between 12.5 and 25 miles per hour, the tor- 
 sional effect required being still the same, we may in like 
 manner control the two motors placed in multiple. In case 
 of any heavy and long grade, this latter arrangement would 
 be used. 
 
 Early in 1888 multiple-series switches were placed by the 
 Sprague Electric Railway and Motor Company on many of 
 the cars installed by it. They were subsequently abandoned, 
 not because of failure to produce the anticipated results, but 
 because at that date, when many other things were giving 
 much trouble, it seemed wise to sacrifice variety of effect to 
 simplicity of mechanism. 
 
 Others, when studying the problem, have desired to make 
 the change from series to multiple connection a step in the 
 progression constantly taking place after and during every 
 stop of the car, and hence, to be produced when the car is in 
 motion and current flowing. This may involve a quick jump 
 of the car, in one case, or a corresponding drag in the other, 
 both occurrences being objectionable. The greatest value 
 of the method seems to be as pointed out above, in which 
 cases the breaking of the circuit and operation of a switch, 
 separate from that used for general control, would be per- 
 missible though shown by recent experience to be needless. 
 
 Now that many previously troublesome features of railway 
 practice have been made easy, while at the same time there 
 is constant increase in the variety of service called for, at- 
 tention is actively devoted to this method. It appears in 
 several important installations recently made, and its use 
 will doubtless become more general. Its advantages are 
 evident. See. Appendix F. 
 
 More than nine-tenths of all the installations thus far made 
 have shown two motors on a car, and these two have almost 
 invariably been placed in multiple. Each is therefore wound 
 so that it may produce the maximum speed required, having 
 between its terminals the full line potential. If they are
 
 8o THE ELECTRIC RAILWAY. 
 
 placed in series, there is, of course, no harm done, the effect 
 being only to produce a lower speed in this way rather than 
 by high-resistance field coils, or an external rheostat. When 
 the motors are placed in multiple, each is nearly independent 
 of the other. One may have its circuit broken while the 
 other is propelling the car.* 
 
 If, on the other hand, the motors are so wound as to give 
 maximum required speed when placed in series across the 
 line, and this be considered the normal relation, it then be- 
 comes practically impossible to place them in multiple. If 
 one machine must, by reason of injury, be cut out, then the 
 other must be protected by some external resistance, or left, 
 at best, to run with a high-resistance field. The advantage 
 to be gained lies in this, that for a given quality of work- 
 manship there will be fewer failures of insulation, since the 
 pressure on each motor is only half that due to the line poten- 
 tial. Further, this prevents effectively, save in a special 
 case, that widely unequal division of labor which has often 
 been observed between two motors on the same car. This 
 is generally due to a difference of resistance in the two mag- 
 netic circuits. 
 
 Its effect can be best understood by considering the fact 
 that the current flowing over an armature is due to the differ- 
 ence between impressed E. M. F. and counter or self-induced 
 E. M. F. The speed of the two armatures must, in car ser- 
 vice, be practically equal. The number of turns, save by 
 gross carelessness in manufacture, will also be equal. Now 
 suppose that by reason of either or both of the causes men- 
 tioned, the two armatures are moving in fields of force dif- 
 fering in strength by 5 per cent, of the stronger, the number 
 of amperes being the same. Consider armature (A) as gen- 
 erating a counter E. M. F. 490. Then armature (B) would 
 generate 490 (490 X 0.05)= 465.5. The effective E. M. F. 
 would then be 10 volts for (A) and 34.5 for (B) the line E. 
 M. F. being as usual taken at 500 volts. The current would 
 
 * A trouble that has sometimes occurred with alarming frequency, however, consists in 
 this: that the commutator of one motor begins to "throw solder" i.e., melts the solder 
 at the connections with armature wires. In such case, look for a loose connection in the 
 other motor. Such a loose connection, being of high resistance, may have thrown an un- 
 due load on the motor which has shown the trouble at commutator, this load causing an 
 excess of current resulting in the melting of the solder.
 
 MOTORS AND CAR EQUIPMENT. 
 
 81 
 
 be -~- for A and -^- for B, or, since the resistances of the 
 
 armatures are practically equal, we have for a 5 per cent, 
 difference in the field strength a difference in the current of 
 345 per cent. This supposes the currents in the fields of the 
 two machines to be identical, and, indeed, such has generally 
 been the case, the connections of the two machines, however 
 controlled, having been usually as shown by Fig. 36, the effect 
 mentioned not having been fully appreciated until lately. 
 
 The field windings being simply so many turns of wire 
 having equal resistance in the two motors, the division of 
 
 FIG. 36. FIG. 37. 
 
 METHODS OF CONNECTING MOTORS IN PARALLEL. 
 
 current from A to B will be equal, while from B to C the 
 division will be determined by strength of field, as just 
 explained. 
 
 If the connections are as shown in Fig. 37 the possible dif- 
 ferences are very largely reduced. 
 
 In this case, each field winding with its armature consti- 
 tutes a separate circuit, and the excessive current which 
 would tend to flow in B, under the condition of unequal 
 reluctance of field, will be checked by its own effect in 
 increasing the ampere turns in circuit B over that in circuit A, 
 thus increasing the counter E. M. F. The current in B will,
 
 3 2 THE ELECTRIC RAILWAY. 
 
 unless saturation for both machines has been reached, be 
 greater than that in A, since by supposition the field strength 
 is less for a given current in B than in A, and the counter 
 E. M. F. must then be less. But the difference will be only 
 proportional to the difference of permeability below saturation. 
 The connections as shown in Fig. 36, in which the machines 
 are cross-connected, have the advantage of offering greater 
 facility for reversing the current. Thus, one switch, revers- 
 ing the current over both fields or both armatures, through 
 the terminals A and B. or B and C, controls both motors. 
 If the machines be not cross-connected, then a separate 
 switch must be used for each machine. This, however, is a 
 matter of small consideration compared to the importance of 
 the result secured by it. The connection of the two motors 
 in series of course renders unnecessary the double reversing" 
 switch, and makes the current in the two motors identical. 
 Another method of meeting this particular difficulty was 
 used, indeed, is still used, on some of the cars equipped by 
 the Edison Company. 
 
 In this method, the two motors remain in multiple: a coil 
 is placed around the field of motor (A') and in series with the 
 armature of motor (B) : likewise on B, a field coil is in series 
 with armature (A). The machines remained cross-connected 
 as shown above. 
 
 These additional coils were known as "balancing coils." 
 If, due to its relatively weak field, armature (B) tends 
 to take more current than its companion, this current tra- 
 versing the coils around the magnets of A, and in a direction 
 opposite to that of the principal windings of A, diminishes 
 the field strength of that machine. Likewise, the current 
 m armature A circulates around B and diminishes its field 
 strength, but by a less quantity than that subtracted from A. 
 The tendency, then, is plainly toward equality. The method 
 has been applied with considerable but varying success. 
 Thus, suppose one machine, A, to be over-saturated; the 
 other, B, to be below saturation. Its armature current, 
 though greater than that of A, may have a less effect 
 in demagnetizing A than will the current of A have in de- 
 magnetizing B, and the inequality would thus be increased.
 
 MOTORS AND CAR EQUIPMENT. 83 
 
 This may be taken to be an exaggerated case, but yet it is 
 possible ; and since it shows an actual reversal of the intended 
 effect, it is plain that in cases of smaller and more probable 
 differences, the device may at least vary widely in its effect 
 without actual reversal. 
 
 If, further, each pair of machines before being mounted 
 were carefully studied, the adjustment might certainly be 
 made very accurate. The practical necessities of manufacture 
 and installation prevent this. The chief objection to this 
 method may be stated in this: that while it introduces a 
 complication of wiring equal to or greater than that demanded 
 by the double reversing switch, it does not cure the evil so 
 efficaciously or so simply as by independence of circuit and 
 double reversing switch, or by placing the motors in series. 
 
 This unequal division of current becomes an evil only 
 when occurring under heavy loads, such that one machine 
 may be required to do much more than its normal work, in 
 which case the efficiency of its performance may be consider- 
 ably lowered ; or the over-load may go so far as to injure the 
 insulation by excessive heat. 
 
 Before leaving the subject of speed control, it should be 
 stated that numerous designs have been made (and a few 
 executed) looking to continuous and approximately constant 
 speed of armature, mechanical devices being used to effect a 
 varying speed reduction between the armature and the axle. 
 In some cases, usually by the use of friction plates or 
 hydraulic gearing, the change of ratio between the armature 
 and the axle speed is effected by some continuous movement, 
 carrying the ratio by changes of insensible value from unity 
 to infinity or the reverse. 
 
 In other cases, usually by the use of some form of the so- 
 called sun-and-planet or external and internal gearing, the 
 changes are abrupt and limited in number. This variation 
 of leverage, however obtained, is certainly in itself an incon- 
 testable advantage. By means of it an armature of minimum 
 weight and working at a maximum efficiency may be made 
 to serve for a given range of work. 
 
 If used in connection with a shunt field, the armature 
 speed might be as constant as in the case of stationary motors
 
 g 4 THE ELECTRIC RAILWAY. 
 
 doing variable work. Change of work requirement would be 
 followed by change of current consumption, which would 
 correspond to slight changes of field strength, the motor 
 efficiency and the speed of armature remaining nearly con- 
 stant. The proper determination of relations in such case 
 would be as follows: First consider the maximum allowable 
 speed of armature, as determined by electrical and mechanical 
 effects in the armature, and its directly attached devices. 
 Calculate the maximum output that might be required for 
 such a length of time as would cause the efficiency of the motor 
 during that time to be of importance. Assume that this 
 output should be at an efficiency of, say, 90 per cent. 
 The current required then becomes known, likewise the 
 resistance of the armature conductors. Their cross-section re- 
 sults from making proper allowance for the heating effect of 
 such a current over such a resistance. Their length is deter- 
 mined by the relation between the cross-section and the 
 resistance. The size of the loop, or of the body of the arma- 
 ture, is calculated from the conditions that the given length 
 of conductor rotating in a field of assumed strength, say 
 one-half that of saturated iron, shall generate a counter 
 E. M. F. of the required magnitude. The other dimensions 
 of the motor follow from ordinary calculations. 
 
 The average output, and those below the average, are all 
 obtained at high efficiency, the field strength being increased, 
 the counter E. M. F. being slightly increased, and the cur- 
 rent largely decreased thereby. 
 
 In case series motors are used with this method, it would 
 seem best so to wind the field magnets as to produce satura- 
 tion with a relatively small service current. The armature 
 would then for the most part rotate in a constant field ; slight 
 changes of the speed of the armature would be followed by 
 considerable changes of current, corresponding to changes 
 in the required output, practically the whole speed regulation 
 being effected through the particular mechanical device used. 
 Should the field strength become constant only at relatively 
 large currents, there would be a considerable variation of the 
 armature speed, which might or might not be in the same 
 direction as that desired to be effected.
 
 MOTORS AND CAR EQUIPMENT. 85 
 
 Summarizing concerning this general method, it may be 
 said that if variable gearing could be had, in which the com- 
 plications would be insignificant, while its reliability and 
 efficiency of transmission would be great, then its use would 
 become general. Active minds are at work on the problem, 
 and it may soon be solved. Meanwhile, the controlling de- 
 mand for simplicity excludes the devices thus far presented. 
 
 The action of shunt motors has just been compared to that 
 of series motors in the particular case of mechanical regula- 
 tion of speed. It remains to inquire why they have not been 
 used in the already extensive application of electric motors 
 to traction work. At the outset it may be stated that the 
 particular quality named by Silvanus P. Thompson as advan- 
 tageous to the series motor, in comparison with the shunt 
 machine, is in fact of no practical value. On p. 539 of his 
 great work, "Dynamo Electric Machinery," Thompson says: 
 
 " The fact that the torque of a series motor depends only 
 on the current, is of advantage in the application of motors 
 to propulsion of vehicles (such as tram-cars) which at start- 
 ing require for a few seconds a power greatly in excess of 
 that needed when running. To start, a large current must 
 be turned on. One convenient way of arranging this is to 
 use two motors, coupled habitually in series. When start- 
 ing, they are, by moving a commutator, coupled in parallel. 
 This doubles the electromotive force for each, and at the 
 same time halves the resistance. For a few seconds a very 
 strong current flows much stronger than that which the 
 motors would stand for any prolonged work and so pro- 
 vides the needful additional torque." 
 
 As has been pointed out above, the practical difficulty 
 met with in electric-railway design is to get sufficiently high, 
 not sufficiently low, initial resistance. The normal resist- 
 ance, even of two series motors, placed in series with each 
 other, is not high enough to prevent needlessly large starting 
 currents from flowing. The absence of opportunity for any 
 considerable experience in this direction offers ready ex- 
 planation for so slight an error in what is, perhaps, the most 
 satisfactory all-round practical electrical treatise in the Eng- 
 lish language.
 
 8 6 THE ELECTRIC RAILWAY. 
 
 A disadvantage which was early discovered in the use of 
 shunt motors is stated thus by Kapp (" Electric Transmission 
 of Energy," pp. 316, 317), speaking of the little Blackpool 
 electric tramway : " It has been explained that the speed 
 of a shunt motor, whei running light, can never exceed a 
 certain limit; whereas a series motor may, under the, same 
 conditions, assume a dangerously high speed. On purely 
 theoretical grounds shunt motors are, therefore, more suitable 
 for tramway work. But a serious practical difficulty was 
 soon encountered. It arose from the uncertainty of elec- 
 trical contact between the wheels and the rails. When a 
 current of electricity has to pass through two pieces of metal 
 in contact, the first condition is that the surfaces should be 
 clean, and that is precisely the condition which cannot 
 always be fulfilled in a tramway exposed to the weather and 
 overrun by other traffic. It would thus occasionally happen 
 that the current was interrupted for a very short time, per- 
 haps only a fraction of a second, but the interval was suffi- 
 cient to cause the field of the motor to lose its magnetism. 
 The consequence of this was, that when contact was restored 
 and the current began again to flow, the armature was not 
 able to offer any counter electromotive force, and an abnor- 
 mal rush of current took place before the field magnets had 
 had time to again become excited. It will be noticed that 
 the injurious effect here described will be the greater, the 
 bwer the resistance of the armature; that is to say, the more 
 efficient the motor, the more will it suffer from an occasional 
 interruption of current. Since it was impossible to abso- 
 lutely avoid these interruptions, the use of shunt motors was 
 discontinued, and series motors were substituted. In a series 
 motor the intensity of the field and, therefore, the counter 
 electromotive force of the armature are at once restored 
 when the current begins to flow, and no abnormal rush of 
 current can take place. To prevent racing when lightly 
 loaded, variable resistances, placed below the platform at 
 either end of the car, have to be used. These resistances 
 are also employed for regulating the speed when the motor 
 is doing a fair amount of work." 
 
 It seems that this difficulty could readily be overcome by
 
 MOTORS AND CAR EQUIPMENT. 8/ 
 
 a simple automatic device, which would open the armature 
 circuit unless the field circuit is closed. Such a device would 
 doubtless have been used, were there any marked advantages 
 in other respects of the shunt over the series winding. The 
 reasons for the use of the latter are : i . The early prejudice 
 founded upon the facts related by Kapp. 2. That in a series 
 motor the regulation of field strength and applied E. M. F. 
 may be effected through one mechanism, the current in the 
 armature and the field coils being the same. 3. The facility 
 of using the field windings as a rheostat, as explained above. 
 
 4. The fact that the insulation of the field coils is somewhat 
 easier in the series machine, the difference of potential 
 between the terminals being much less than in the shunt 
 motor. 5. The fact that up to saturation there is an auto- 
 matic field regulation. 
 
 These reasons are good and sufficient to determine present 
 practice, though it is certainly true that good service would 
 "be had from a shunt-wound motor, speed control being 
 effected mechanically or through rheostats placed in the 
 armature and the field circuits, to which may or may not be 
 added commutation of field windings. 
 
 In considering the motor, in place, in relation to the other 
 necessary parts of the car, we are led to discuss : 
 
 (A) The manner of connecting the armature shaft with 
 the axle. 
 
 (B) Convenience of position with respect to inspection, 
 removal and repairs. 
 
 The means used for transmitting power from the armature 
 shaft to the axle have been : 
 
 i. Spur gears. 2. Sprocket chains. 3. Beveled gears. 4. 
 Connecting rods, as from one locomotive driver to the other. 
 
 5. Worm gears. 6. Ordinary belting. 7. Ropes and pul- 
 leys. 8. Friction plates, etc., as explained, for variable speed 
 reduction. 9. Centering of armature directly on the axle. 
 
 i. Spur Gears. Perhaps 99 per cent, of all the rail- 
 way motors now running are connected by spur gearing. 
 From this general use, it follows that the burden of proof 
 lies upon any other method before it may be considered as 
 being practically in the field. The efficiency of this method
 
 38 THE ELECTRIC RAILWAY. 
 
 may be roughly stated as 90 per cent, for each couple of 
 intermeshing gears, if exposed to grit, and provided the load 
 be reasonably large. When the load is very small, the ^xed 
 loss in friction causes the efficiency to be low. 
 
 If the gears be inclosed in oil-tight cases, as is now largely 
 practiced, the couple efficiency is from 90 to 95 per cent, at 
 medium or heavy loads. In mechanical simplicity and relia- 
 bility, under reasonably good design, there is little to be 
 desired. The serious problems that have been met are: 
 
 (a) Choice of material, regard being had to strength, free- 
 dom from noise, and cost. Cast iron for the heaviest gear 
 (that on the axle), with machine-cut teeth, has been almost 
 exclusively used. For the smaller gears, running at higher 
 speeds, cast iron, steel, brass, bronze, wood (for the teeth), 
 rawhide all have been tried, with a gradual settling down 
 to steel or gun-metal. The evolution from the train of 
 four gears, with intermediate shaft, to the train of two gears. 
 (armature pinion and axle gear) has removed almost en- 
 tirely the principal cause for endeavoring to depart from 
 the metals for gear material. That cause has been, still is, 
 on earlier apparatus, the excessive noise of the high-speed,, 
 much-worn teeth. It seems probable that on the single- 
 reduction gear motors, now so widely introduced, the best 
 practice will use steel wheels with machine-cut teeth both, 
 on axle and armature shaft. 
 
 (/;) Diminution of noise. This is connected with the mat- 
 ter of (i) speed, (2) material, (3) relative exposure, (4) sus- 
 pension method for the whole machine, (5) tooth-design, (6) 
 alignment and (7) load transmitted. 
 
 (1) In designing for a speed of about 500 armature revolu- 
 tions per minute, geared to produce a car velocity of about 
 ten miles per hour, it is probable that the manufacturers have 
 gone as far as is wise, so long as gears are used at all. The 
 noise at this speed may be inconsiderable. The next step 
 after passing the single-reduction motor, geared as just sug- 
 gested, seems to be the gearless or axle motor. Of its de- 
 sirability, notice will be taken later. 
 
 (2) As to material, we have only to add to what has been 
 said above, that the quietest substitute for metal thus far
 
 MOTORS AND CAR EQUIPMENT. 89 
 
 found is wood. This material, however, is not strong enough 
 to be used for the armature pinion, the diameter of which 
 scarcely exceeds five inches. Rawhide has been the most 
 successful rival of metal for this hard service. In this, as in 
 other substitutes, it has been difficult to obtain uniform qual- 
 ity in the finished product. Consequently, widely different 
 opinions have been held in regard to its value. The contest, 
 we think, will cease with the passing away of the older 
 high-speed machines, leaving metal undisturbed in the 
 field. 
 
 (3.) It is easily practicable tightly to cover the gears of a 
 single-reduction motor. Nor would the problem have been 
 difficult for the four-gear machines, had it been held in view 
 during the original design. That, however, was not the case 
 in respect to the motors most largely used up to date, those of 
 the Thomson-Houston and of the Sprague or Edison com- 
 panies. Efforts have been made to add suitable covers, but 
 every mechanic knows the difficulty of a "patch job." 
 
 Unless the cover be very firmly attached, and of material 
 not particularly resonant, its use may be of no benefit in the 
 matter of diminishing noise, but its great usefulness in ex- 
 cluding dust, pebbles, etc., will of course remain. 
 
 (4) That form of motor suspension which places the whole 
 weight of the machine on springs will readily be understood 
 to be that which best conduces to soft-running gear wheels. 
 Until very recently, it has been an almost invariable rule to 
 place one-half of the motor weight dead on the axle, as ap- 
 pears in several of the figures illustrating familiar railway 
 motors. The other half rests on springs which permit a 
 rotation of the motor around the axle. This method has 
 doubtless been of more value in saving the teeth from break- 
 age by sudden strains, than in diminishing noise. 
 
 (5) As to tooth-design, there has been little variety, an 
 approved epicycloidal tooth having been most largely used. 
 By using two sets of such teeth on the same gear, " stag- 
 gered " with respect to each other, the Edison Company has 
 been able largely to reduce the rattle often noticed in well- 
 worn or badly-adjusted gears of the double-reduction type. 
 The design is shown in Fig. 38.
 
 9Q THE ELECTRIC RAILWAY. 
 
 (6) Good alignment is of value in this as in all similar 
 
 cases. 
 
 (7) In the matter of load transmitted, it is observed that 
 a car having been put in rapid motion by the work of the 
 motor, the gears, grinding noisily while the work is being 
 
 ;UE-EDISON MOTOR. 
 
 done, become almost noiseless if the current is cut off, though 
 the car may continue its rate of motion unchanged, as on 
 a down grade. 
 
 Lubrication of Gear Journals. For convenience, we may 
 refer here to armature shaft journals also, and state that 
 the best results have been had from the use of grease rather 
 than fluid oil. The self-oiling bearings, such as are so widely 
 used on dynamos, are not so well adapted to service on a 
 much-jolted car motor. Oil is lost rapidly from the basin, 
 and it has seemed, further, that the feeding rings or chains, 
 supposed to revolve eccentrically around the shaft while dip- 
 ping in the oil, are prevented from traveling by the constant 
 shocks, which tend to keep them playing vertically rather 
 than to rotate them. The grease, placed in a very simple 
 box or cup, should be followed by spring pressure applied 
 to the cover or to some disk inside the box. A pin resting 
 on the shaft (but pressed lightly, so that it will not cut) and 
 passing up through the body of the grease, causes the melted 
 particles to follow easily down its sides. 
 
 For the journals of the armature shaft, globe or conical 
 bearings, as shown in Fig. 39, have been much used, while
 
 MOTORS AND CAR EQUIPMENT. 91 
 
 for intermediate and axle shafts straight bearings have been 
 preferred. 
 
 There are, however, good authorities in favor of straight, 
 babbitted bearings throughout. If babbitt or other soft 
 metal be used, the greatest care must be exercised as to its 
 quality. 
 
 2. Sprocket Chains. Four or five years ago, and in the 
 practice of Mr. Charles J. Van Depoele, one of the earliest 
 workers in. the field, sprocket chains were used on perhaps 
 seventy-five cars. Examples of quite tolerable service from 
 these early machines may still be found. The sprocket chain 
 was perhaps the best medium of transmission, as long as the 
 motor remained above the car floor, on the platform or inside 
 the car. Nothing better was available for covering the con- 
 siderable distance between armature shaft and axle. But 
 the speed was high, and the art of making sprocket chains 
 for such trying service was not far advanced. Hence this 
 device disappeared when the motors came to be placed under 
 the car floor (as was done by the Sprague Electric Railway 
 and Motor Company in its earliest work), and the spur gear 
 came to the front. There seems good reason to suppose that 
 the chain may still have its sphere of usefulness, as in trans- 
 mitting power from the directly-driven axle to a second 
 one, the speed between axles being less than elsewhere in 
 the train. Useful and apparently successful applications of 
 
 ___" 
 
 FIG. 39. FORMS OF BEARINGS. 
 
 sprocket-chains have been found in the construction of elec- 
 tric mining machinery, improvements having been made in 
 the manufacture of the links of such chains. 
 
 3. Bevel Gears. In using these, the armature is placed 
 parallel to the axis of the car. The only reason for endeav- 
 oring to do this is that one motor may thus be connected to 
 two axles of a truck, thus utilizing the whole weight on each
 
 9 2 THE ELECTRIC RAILWAY. 
 
 truck for adhesion, or for "traction," as it is frequently but 
 improperly expressed. On grades lower than 3 per cent, 
 this double connection is of relatively small importance, since 
 on such grades one-half the weight of the car, if resting on 
 the driven wheels, is found sufficient to prevent skidding, on 
 average tracks, even when wet with snow a reasonable 
 supply of sand being available. For an ordinary 1 6-foot car, 
 operating on such grades, a single motor of such capacity 
 as are those generally rated at 15 horse-power (as by the 
 Thomson-Houston, Edison, Short, and Westinghouse com- 
 panies in the United States), and geared to one axle of an 
 ordinary four-wheel truck, is sufficient, both as to adhesion 
 and traction. Under favorable track conditions, and unless 
 unusual speeds be required, the same motor will handle, with- 
 out slipping and without overheating, one trail car of about 
 the size of the motor car. Where greater effort is required, 
 or under exceptionally bad track conditions, it is best to 
 apply the driving power in some way to both axles of a four- 
 wheel truck ; if a long car be in question, requiring a six- 
 wheel truck, or two four-wheel trucks, then the power should 
 be applied to those axles, which carry from 70 to 90 per 
 cent, of the whole weight. As is familiarly known, this re- 
 quirement has been met in the designs most widely used by 
 using a separate motor for each driven axle. 
 
 There was, in the early days, four or five years ago, some 
 prejudice in favor of two motors per car for other reasons 
 than that of securing the needed adhesion. Thus it was 
 argued that if one machine should be injured, the other 
 would be available to carry the car to the hospital. This rea- 
 soning supposed a degree of unreliability in electric service 
 which would, if permanently, inherently true, rule it out of 
 good practice. The proper reliance in case of accident is 
 upon the next car following, and accidents should be, indeed 
 are, rare enough to make this occasional double duty unob- 
 jectionable. 
 
 In face of other manifest advantages of a single motor, 
 this prejudice would soon have disappeared, and with it the 
 double motor equipment for ordinary cars, except for this 
 difficult problem of sufficient adhesion on heavy grades.
 
 MOTORS AND CAR EQUIPMENT. 
 
 93 
 
 The device of beveled gears was early considered and tried, 
 especially by Bentley & Knight, who thus equipped a car in 
 1886. But the mechanical difficulties are great. The truck 
 framing must be rigid to secure alignment of gears at the 
 opposite ends of the armature shaft ; the weight of the motor 
 must be practically uncushioned, the efficiency of beveled 
 gears is low ; it is difficult to get sufficiently strong teeth in a 
 beveled pinion, of such diameter and pitch as required for 
 use in connection with a 3O-inch car wheel, which limits the 
 
 FIG. 40. RAE MOTOR TRUCK. 
 
 diameter of the axle gear to about 24 inches, thus leaving 
 very little clearance. 
 
 Many students of the problem thought these difficulties 
 such as to warrant, without trial of the beveled gears, the 
 adoption of other methods. Recently, however, the method 
 lias been given some prominence by the efforts of Mr. Rae, 
 electrician of the Detroit Electrical Works, Detroit, Michigan. 
 There are now running perhaps two hundred cars equipped 
 by that company. Curiously enough, they have been most 
 largely used on comparatively flat roads, where this peculiar 
 merit is of least value, since on such roads a single motor could 
 be geared in the ordinary way. The weakness of gear teeth 
 has given some trouble in cases of the heavier work under-
 
 94 THE ELECTRIC RAILWAY. 
 
 taken. It remains to be demonstrated whether satisfactory 
 work will be had from similar installations on very heavy 
 grades. The progress of these efforts will certainly be a 
 matter of great interest. (See Appendix F.) 
 
 4. The use of connecting rods instead of gears was tried 
 some years ago by Bentley & Knight, later by Mr. Leo Daft. 
 But it has remained for Mr. Rudolph Eickemeyer again to 
 call attention to this method, by bringing out at the same 
 time a motor of such slow speed that there is no reduction 
 between armature and axle. The connecting rod transmits 
 the power to the two axles of a truck from a single armature, 
 driving forward and backward, as in the case of three driv- 
 
 FlG. 4 i.-ElCKEMEYER GEARLESS MOTOR TRUCK. 
 
 ing axles of a locomotive coupled together. (See Fig. 41 for 
 further information.) 
 
 [t will at once be inquired why, having reduced the arma- 
 ture speed to that of the axle, Mr. Eickemeyer does not then 
 center the armature on the axle and avoid any intermediate 
 mechanism. His reasoning is thus: That it is generally 
 
 ' to use one motor rather than two; that two motors 
 
 I be needed for grade work if the armature were con- 
 centric with the axle; that when the motor is placed midway 
 
 xm the two axles its whole weight may be very conven- 
 itly placed on springs, a matter of grave importance, by 
 
 n of the cost of the track repair due to the pounding of
 
 MOTORS AND CAR EQUIPMENT. 95 
 
 so much dead-weight ; and that such position for the motor 
 offers greater facility in the matter of removals than if on 
 the axle. 
 
 The few equipments thus far made by Mr. Eickemeyer have 
 been in satisfactory operation for some months. It seems 
 quite clear that the weight of these machines is small, capac- 
 ity and speed being duly considered. But in comparing this 
 or any other motor of only axle speed with motors requiring 
 reduction gearing, it must be remembered that the same 
 excellence of design that may be applied to the production 
 of a machine of say 20 horse-power at 100 revolutions, 
 and weighing 5,000 pounds, will produce a machine of 20 
 horse-power at 500 revolutions, weighing about 2,000 pounds, 
 or two motors of 10 horse-power, weighing about 1,200 
 pounds each. Now, suppose these two motors to be entirely 
 spring-supported, and geared, one to each axle, through spur 
 gearing, in dust-proof, oil-tight cases. We then have, in the 
 one case, one large motor (more difficult to handle), 5,000 
 pounds weight, spring-supported ; four connecting rods, both 
 axles driven. In the other, two small motors, 2,500 pounds 
 weight, spring-supported ; four spur gears ; both axles 
 driven. The future can best determine which of these de- 
 signs will be generally preferred. 
 
 In comparing the Eickemeyer equipment with other motors 
 of axle speed, we are brought to 9 the direct centering of 
 the armature on the axle. The most notable example of 
 this method now in operation is the City and South London 
 Electric Railway. While, indeed, the particular feature of 
 centering the armature on the axle is here illustrated, yet in 
 many ways it is not a case properly comparable with the 
 efforts no\v being made in the United States. He who has 
 felt the larger freedom which comes in designing a separate 
 locomotive, instead of a. machine placed out of sight under a 
 car floor, will appreciate the fact that gearless motors for 
 ordinary street-railway service cannot be copied from the 
 English example. We shall, then, for the present, return 
 to those designs which have just been built in this country. 
 
 Two companies, the Short Electric Company, of Cleveland, 
 Ohio, and the Westinghouse Electric and Manufacturing Com-
 
 9 g THE ELECTRIC RAILWAY. 
 
 pany, of Pittsburg, Pa. , have already manufactured such gear- 
 less motors. 
 
 The Short motor has a " disk armature " with pole pieces 
 on the sides, familiar in Brush arc-light dynamos. The arma- 
 ture is attached to a hollow sleeve, concentric with the axle, 
 but supported, through a framework, on springs, and of large 
 enough inside diameter to permit vertical play of about an 
 inch. The field magnets and pole pieces are supported in 
 like manner and move with the armature. The sleeve trans- 
 mits its power to the axle through six cushions or springs 
 connecting two pairs of disks, one keyed fast to the axle, 
 the other fast to the sleeve (Fig. 42). 
 
 The design of the Westinghouse Company shows a Sie- 
 mens or drum armature keyed directly to the axle. The 
 field is made by four pole pieces connected through an 
 external cylinder. The whole weight of the motor is directly 
 upon the axle (Fig. 43). 
 
 It will be noted that the Short design covers the important 
 point of spring suspension, which will be better appreciated 
 as time relentlessly forces attention of electric-railway owners 
 to track repair. It is intended to use one motor for light 
 grades, and two for heavy ones, as above explained. The 
 relative merit of this and the Eickemeyer design will then be 
 determined, in time, by the peculiarities in service of con- 
 necting rods on the one hand and spring connection between 
 pairs of disks on the other. 
 
 Ideally simple, the Westinghouse design will be called 
 upon to demonstrate whether or not some departure from 
 such simplicity is wai ranted by consideration for the track. 
 
 We do not here enter into discussion of the relative merits 
 of the three motors viewed simply as electric machines of 
 greater or less efficiency and weight per unit of output. 
 Little is known concerning these points. That it is difficult 
 to obtain high efficiency at such low speeds and in such lim- 
 ited space goes without saying. The necessary limitation 
 of weight also raises a serious question as to whether gear- 
 less motors will not eventually be restricted to comparatively 
 light service, unless the speed be considerably higher than 
 in ordinary surface tramway service. If the connecting
 
 MOTORS AND CAR EQUIPMENT. 
 
 97 
 
 FIGS. 42-44. TYPICAL MODERN MOTORS. 
 FIG. 42. SHORT GEARLESS MOTOR. FIG. 4 > WESTINGHOUSE GEARLESS MOTOR. 
 
 FIG. 44. EDISON SINGLE-REDUCTION GEAR MOTOR. 
 7
 
 gg THE ELECTRIC RAILWAY. 
 
 rods of the Eickemeyer design, or the sleeve, springs, and 
 disks of the Short design, are indeed necessary to save the 
 track, we have at once a complication fairly comparable with 
 the simple spur gears now used, and a total weight, for a 
 given output, considerably greater. 
 
 In regard to (5) worm gears, (6) ordinary belting", (7) ropes 
 and -pulleys, (8) friction plates, etc., little need here be 
 said. They have been used sporadically and with no marked 
 success. The worm gear is apparently the best of these four 
 methods. Its low efficiency has prevented many from seri- 
 ously studying it. Mr. A. Reckenzaun has constructed a 
 motor operating a worm gear, and has run the car by storage 
 batteries. Details of his results have been published, claim- 
 ing remarkably high efficiency as compared with that usually 
 given by mechanical authorities. It is perhaps unfortunate 
 that an untried gearing should have been experimentally 
 associated with storage batteries. It is hardly fair toward 
 the gearing. 
 
 (B) Convenience of position with respect to inspection, 
 repair and removal. In the earlier electric-railway practice 
 numerous examples are found in which the consideration 
 of accessibility seems to have been treated as important. 
 Motors were placed on the platform of the car, in the body 
 of the car, or in a separate cab. The desire to use spur 
 gears, and to leave untouched th'e space available for pas- 
 sengers, caused the Sprague Company to place motors under 
 the car floor; and there they remain in the general practice 
 of to-day. The position is one of the greatest exposure to 
 dirt and to foreign objects that may cause immediate de- 
 struction ; it also makes repairing a difficult matter. Never- 
 theless, it is the right position, pointed out alike by mechan- 
 ical and commercial reasons. Recent improvements in the 
 reliability of motor windings, in the mechanical construction 
 of motor frames, in the simplicity and protection given the 
 gearing all combine to diminish the troubles resulting- 
 from this difficult position'. If separate locomotives are used, 
 the armature must still be concentric with the axle, or 
 removed from it only by one train of gears. The general 
 body of the motor may, however, be extended vertically,
 
 MOTORS AND CAR EQUIPMENT. 
 
 99 
 
 FIGS. 45-47.-TYPICAL MODERN MOTORS. 
 
 FIG. 45. THOMSON-HOUSTON 'WATER-PROOF" MOTOR. FIG. 46. SHORT SINGLE-RE- 
 DUCTION GEAR MOTOR. FIG. 47. THOMSON-HOUSTON SINGLE-REDUCTION GEAR MOTOR.
 
 100 THE ELECTRIC RAILWAY. 
 
 instead of horizontally, as now; or it may lie between these 
 two positions, as in the City and South London Railway loco- 
 motives. This arrangement, of course, makes the machines 
 much more accessible. 
 
 It has been thought by some that separate locomotives 
 would eventually be introduced into the street service of 
 great cities. This does not seem probable. To bring it 
 about means to introduce greater dead-weight per passenger 
 carried, greater expenditure of power, greater difficulty as 
 to adhesion, and greater street obstruction. For suburban 
 service, or for elevated or underground railways, in which 
 several cars are to be run in a train over easy grades, the 
 problem is widely different, and seems to call for the separate 
 construction. 
 
 It may also appear in a sort of compromise case as follows: 
 a suburban line feeds a main cable line ; cars are brought to 
 and from the latter by an electric motor car having perhaps 
 a few seats for smokers. This method has already been 
 proposed for one of our large cities. 
 
 CAR TRUCKS. 
 
 Some of the earlier installations show the motor hung on 
 the axle on one side, and suspended by springs directly from 
 the car body on the other. This method was applied to cars 
 that had been built for propulsion by horses, and required no 
 change save an enlarging of the axle and a strengthening of 
 the floor beams. Subsequent practice has, however, devel- 
 oped a different method, now almost universally used. The 
 wheels and axles are mounted in a frame which carries a 
 cross-bar for the support of the motor, and springs which 
 receive directly the weight of the car body and cushion its- 
 oscillations. This separate truck offers facilities for mounting 
 and dismounting motors, for exchange of car bodies (as 
 between open and closed), and in a convenient manner sup- 
 plies the strength of frame needed for the support of the 
 motor. A great variety of designs may now be seen, differ- 
 ing rather in detail than in any important principle. It is 
 gratifying to note in the later designs a simplicity of con- 
 struction which was not observable in the earlier forms.
 
 MOTORS AND CAR EQUIPMENT. 
 
 IOI 
 
 FIG. 48. 
 
 FIG. 49. 
 
 FIG. 50. 
 
 FIG 
 
 TYPICAL ELECTRIC STREET-CAR TRUCKS. 
 
 PIG. 48.-BRILL TRUCK FIG 49 .-MASiF. R TRUCK FIG. 5 o.-BRiLL MAXIMUM TRACTION 
 BOGIE TRUCK. FIG S I.-STEFHF.NSON BOGIE TRUCK.
 
 I02 THE ELECTRIC RAILWAY. 
 
 Nuts, bolts, screws, even rivets, should all be considered as 
 .evils sometimes necessary, but to be kept down to a mini- 
 mum in numbers. There is room for endless discussion 
 concerning the details of truck design, but the scope of this 
 book permits that only some of the more familiar types shall 
 be shown, as in Figs. 48 to 56. 
 
 During the last year, there has been a very rapid increase 
 in the use of larger cars than the standard horse-car, which 
 may be said to be that having a 1 6-foot body. These larger 
 cars have brought with them the double truck and its varia- 
 tions. Since the maximum allowable wheel base may be 
 said to be 7 feet, it is evident that in going much beyond a 
 1 6-foot body, it becomes necessary to use more than two 
 axles, as the overhang, and consequent "teetering," would 
 otherwise be excessive. Steam-railway practice affords 
 abundant precedent in the matter of double-truck cars, but 
 some of the conditions for electric service are new. These 
 are mainly the provision of suitable space and framing 
 members for motor support, provision for sharper curvature 
 than is ever met with in steam practice, and especially 
 important is the problem of throwing a large part of the 
 total weight of the car on two axles, instead of equally 
 dividing it over the four axles of two trucks, as in general 
 steam-railway practice. It is not desirable to multiply the 
 number of motors on any one car ; it is desirable necessary, 
 indeed that the driven axles should bear such weight as will 
 prevent skidding of the wheels. On wet, especially snowy, 
 rails the frictional resistance between wheel and rail is found 
 at times as low as one -tenth of the total weight on the 
 wheels. That is to say, if an effort of 200 pounds per ton 
 were required to start a car resting its whole weight on the 
 driven wheels, these wheels may slip if the rails be wet. 
 If the rails be dry, the wheels may not slip until a horizontal 
 effort equivalent to 0.35 of their load has been applied; or, 
 expressed in pounds per ton, they may slip when the hori- 
 zontal effort is 700 pounds per ton. 
 
 ^Experience shows that on a level track, with ordinary run- 
 ning gear, an effort of about 70 pounds per ton is required to 
 start a car this overcoming the inertia and friction of the
 
 MOTORS AND CAR EQUIPMENT. 
 
 103 
 
 FIG. 52. 
 
 FIG. 53. 
 
 FIG. 54. 
 
 FIG. 55 . 
 
 FIG 5 6. 
 TYPICAL ELECTRIC STREET-CAR TRUCKS. 
 
 FIG. 52. STEPHENSON TRUCK. FIG. 53. ROBINSON RADIAL TRUCK. FIG. 54. TAYLOR 
 TRUCK. FIG. 55. PECKHAM CANTILEVER TRUCK. FIG. 56. McGuiRE TRUCK.
 
 I04 THE ELECTRIC RAILWAY. 
 
 parts. On grades, add 20 pounds per ton for each i per cent, 
 of grade to give the starting effort. Thus, on a 3 per cent, 
 grade the starting effort = 70 + (20 X 3) = i 3O pounds. Since 
 200 pounds may be the lower limit of adhesion, we may cal- 
 culate that on a grade of - ^ Q 7 '- 6.5 per cent, the wheels 
 
 may slip, though the whole weight of the cars be on the 
 driven axles, as in the ordinary 1 6-foot single-truck car 
 carrying one motor on each axle! A similar calculation 
 shows that if only half of the total weight be on the driven 
 axle, or axles, there may be slipping at the start, on a 1.5 
 per cent, grade. The adhesion in this case for the driven 
 axles is only 100 pounds per ton of total weight; of this take 
 70 pounds for starting on a level, leaving 30 pounds per ton 
 for gravity resistance, which would be met on a grade of 1.5 
 per cent. When once under way, the horizontal effort per 
 ton, exclusive of gravity effect, may be taken at 30 pounds 
 as a high figure for ordinary conditions. We have, then, 
 when the car is in motion, as possible slipping grades, for the 
 
 whole weight on "the driven axles, -- = 8.5 per cent., 
 
 for half the weight on the driven axles, =3.5 per 
 
 cent. 
 
 Fortunately, the conditions here supposed are rare, but the 
 sand box should be filled in order to meet them. If either 
 the start or the run must be made on a curve or on a very 
 rough track (or on both), the figures, 70 pounds and 30 
 pounds, for starting and maintaining speed, respectively, 
 may be exceeded, and the grades of possible slipping cor- 
 respondingly diminished. 
 
 Examples may be given of cars in daily operation on 
 grades as high as 13 per cent. Sand is frequently used. 
 The cars have ordinary four-wheel trucks, a motor on each 
 axle. There is rarely any commercial demand for operation 
 over any heavier grades. 
 
 For single-truck cars we may, then, summarize the case as 
 follows: 
 
 For summer service (i.e., when rails are generally in good
 
 MOTORS AND CAR EQUIPMENT. 10$ 
 
 condition) on grades up to 5.0 per cent., and for winter ser- 
 vice on grades up to 3 per cent., the required adhesion may 
 be had from one driven axle. On grades above 5 per cent, 
 for summer and 3 per cent, for winter both axles must be 
 driven; this may be done by using two motors, as is now 
 ordinarily the case, or by some connection of a single motor 
 to both axles, should any method of accomplishing this prove 
 successful. 
 
 The case of the long cars may be thus stated. For the 
 lower grades, as above mentioned, for summer and winter 
 respectively (requiring only half of the total weight to be on 
 the driven axles), adhesion may be had (a) by gearing a 
 motor to each of the two axles of one of the two ordinary 
 trucks; (b) or by gearing one motor to one of the axles of 
 each of the two ordinary trucks ; (c) or by gearing one motor 
 to both axles of one of the two trucks used, if such gearing 
 can be perfected. 
 
 For the heavier grades (requiring all or nearly all the 
 weight to be on driven axles) the requisite adhesion may be 
 had (a) by gearing one motor to each of the axles of the 
 two ordinary trucks, making four motors in all; (b) by 
 gearing one motor to both axles of the two trucks, if such 
 gearing is successful ; (c) by using two trucks, modified in such 
 a way that one axle of each shall carry nearly all the weight 
 of one-half of the car, this axle having a motor geared to it ; 
 the other axle carrying only enough weight to keep it on the 
 track, and having no motor geared to it; (d) by using only 
 three axles altogether, those under the ends of the car to 
 carry nearly all the weight, and having a motor geared to 
 each the middle axle serving only as a guide, and carrying 
 no motor. The first method (a) is objectionable by reason 
 of the great number of parts to be maintained and the rela- 
 tively greater losses in transmission of power to the axles. 
 The second (b) has not been generally demonstrated. The 
 third (c) may utilize 90 per cent, of the total weight, and has 
 been successfully applied on considerable grades, through 
 what is called the "maximum traction truck." 
 
 The fourth (d) may utilize about 90 per cent, of the whole 
 weight, and has been successfully applied on considerable
 
 IO 6 THE ELECTRIC RAILWAY. 
 
 grades through what is called the "radial truck." To 
 further explain these last two methods, Figs. 50 and 53 are 
 given, showing the trucks mentioned. Both these designs 
 are of great importance. As compared with ordinary truck 
 construction, they represent a departure which seems essen- 
 tial to the application by present methods of spur-geared 
 motors to long-car service (20 to 35 foot bodies) on grades 
 above 5 per cent. The use of a sprocket chain, connecting 
 the motor axle with an empty axle (as recently designed for 
 some special cases) , may indeed serve the same purpose ; its 
 success has yet to be demonstrated. The efforts about to be 
 made should be watched with great interest. Further, it is 
 of course possible to use four motors on trucks of ordinary 
 construction. This method has, indeed, been proposed by a 
 large railway company, for covering its winter service on 
 slippery tracks ; in summer, one motor truck is to be replaced 
 by an empty truck taken from a summer or open car; the 
 motor truck thus borrowed is to be put under the open car, 
 which would thus be ready for service. The closed car, when 
 needed during the summer, could also be run by its remain- 
 ing motor truck. In other words, both classes of cars, open 
 and closed, would at any time during the summer be ready 
 for immediate service the total equipment, counting both 
 classes, being only two motors per car. Further, it is argued 
 that the tractive power and the adhesion against slipping 
 are thus both kept in a reasonable proportion to the demand, 
 which, for both, is greater in winter than in summer. 
 
 It was also proposed that the individual weight and capacity 
 of the motors might be diminished as compared with present 
 standards. If each motor were of 10 horse-power capacity, 
 then in summer each car would have a total of 20 horse- 
 power, and in winter a total of 40 horse-power the supply 
 in both cases being considered sufficient for grades not ex- 
 ceeding 5 per cent., and for a schedule speed of about 8 
 miles per hour, including stops, the cars in question being 
 about 28 feet long in the body. While this method has 
 some advantages, yet on account of the objection to multipli- 
 cation of motors already pointed out, it does not seem as 
 desirable as that based on the use of trucks such as the max-
 
 MOTORS AND CAR EQUIPMENT. 1 07 
 
 imum traction type, limiting to two the number of motors 
 per car, winter and summer. 
 
 If it be still required that open and closed cars shall be in 
 constant readiness during the summer, two methods are 
 open. If the grades are light, one motor may at the open- 
 ing of warm weather be transferred to the open cars, each 
 car thus having for summer work one motor of, say, 20 horse- 
 power capacity. If the grades be heavy, requiring even in 
 summer much more than half the weight to be on the driven 
 axles, then a double equipment of two motors per car may 
 be had, each of, say, 15 horse-power capacity. If one motor 
 can be successfully geared to the axles of a truck as by 
 sprocket chains, by gearing or connecting rod, the shifting 
 of one " live " truck from the closed to the open car, as above 
 described, would, for the lower grades, produce the desired 
 result of having both classes of cars in readiness during the 
 summer months. If this condition.be required on roads of 
 heavier grade, a complete independent equipment must be 
 provided for open and closed cars, no matter how the motors 
 may be connected to the axles. It is not probable that such 
 an expense would be incurred by many companies; they 
 would in general prefer to run only the closed cars, or to run 
 the open cars with curtains down during summer rains ; or, 
 again, to run the open cars only as trailers. This latter 
 method has the disadvantage of attracting the live load away 
 from the motors, thus making more difficult the problem of 
 adhesion. 
 
 TROLLEYS. 
 
 To obtain a satisfactory " running contact " was four years 
 ago a serious question. To-day, the trolley apparatus as a 
 whole is in a satisfactory condition. A brass grooved wheel 
 about five inches in diameter is centered on rawhide or 
 graphite journals, and mounted at the end of a pole about 12 
 feet long ; this pole trails back from the middle of the car 
 roof at an angle of about 30 degrees with the vertical, when 
 the trolley wire is 18 feet above the ground; it is pivoted, a 
 few inches from the lower end, in a frame attached to the 
 car roof, and springs acting at the lower end press the wheel
 
 io8 
 
 THE ELECTRIC RAILWAY. 
 
 against the trolley wire ; by expansion or contraction of these 
 springs, the pressure is continued as the height of the wire 
 varies ; by a cam arrangement, the lever arm through which 
 the springs act against the pole may be changed in such a 
 way as to keep the pressure against the wire nearly constant, 
 whatever the degree of compression or expansion of the 
 operating springs. This general description covers a num- 
 ber of satisfactory forms varying in detail. The pole may 
 be of wood or steel ; the steel may be straight, tubular, in 
 sections of different diameter, or drawn tapering in one 
 piece ; the current may pass through the journal of the wheel 
 or be taken through brushes ; thence it may go through the 
 
 I J 
 
 FIG. 57. UNDERRUNNING ROD TROLLEYS. 
 
 pole, insulated at its base to the wire leading to the motor; 
 or it may pass from the wheel to an insulated wire passing 
 through the pole. 
 
 Two variations from this general form are important. In 
 one, the wheel just mentioned is replaced by a shoe which 
 slides against the wire. Usually a lining of soft metal is 
 placed in the shoe, being replaced every day or two at an 
 expense of about one cent. This has been used by the Short 
 Electric Company and seems to have given satisfaction. In 
 the other the wheel is replaced by a straight bar of consider- 
 J length, placed at right angles to the pole and to the
 
 MOTORS AND CAR EQUIPMENT. 109 
 
 longer axis of the car. To prevent the possibility of trouble 
 at crossings and switches, while doing away with overhead 
 frogs, this bar or cylindrical rod should be cuived down at 
 its ends, or the " T " form should be replaced by an inverted 
 "U." (See Fig. 57.) 
 
 This " T " form has lately been brought forward by Sie- 
 mens & Halske. It was used in 1887 by the Sprague Electric 
 Railway and Motor Company, and sacrificed, perhaps need- 
 lessly, to the desire to please in the matter of looks. While 
 the action of the trolley wheel is on the whole satisfactory, it 
 would certainly be a great gain in line construction to drop 
 the use of frogs, cross-overs, etc. 
 
 A few typical forms are shown in Figs. 58 to 63. 
 
 CAR WIRING. 
 
 As to the car wiring, it may be thus briefly described. 
 There are independent circuits between the trolley and the 
 rails, one for lamps, one for motors, one for the lightning ar- 
 rester, and there may be one for heaters. In the lamp circuit 
 there may be placed in series 16 candle-power lamps, each at 
 normal brilliancy when it has about 100 volts between its 
 terminals. Since the line potential is approximately 500 
 volts, if more than five lamps be required, they must be of 
 lower voltage than 100, or another circuit must be made. 
 
 The motor circuit is itself in duplicate when the motors 
 are in multiple, and may indeed be said to be constituted of 
 two circuits. One must of course vary its wiring according to 
 the system of regulation employed. Fig. 64 shows the stand- 
 ard wiring of the Thomson- Houston Company. Fig. 65 shows 
 that of the Edison Company. In the motor circuit are placed 
 the safety fuses and controlling switches. As to the use of 
 these, consult the rules of practice given elsewhere. Elec- 
 tric heaters have not come into very general use, but may 
 grow in favor. Thus far the practice has been to place two 
 or more heaters in multiple, in the first few hours of service, 
 causing a current of about 6 amperes to flow. When the 
 car has been well warmed, the heaters are thrown into the 
 series arrangement, reducing the current to about 3 amperes. 
 It is said that this quantity of current will keep a 1 6-foot
 
 no 
 
 THE ELECTRIC RAILWAY. 
 
 FIG. 62. 
 TROLLEYS AND TROLLEY BASES. 
 
 FIG. 6 3 . 
 
 FIG. 5 8. SHORT SLIDING TROLLEY. FIG. SQ.-BAKER TROLLEY. FIG. 60. SHORT TROL- 
 LEY. FIG. 61. "BOSTON" TROLLEY. FIG. 62. LIEB TROLLEY BASE. FIG. 63. 
 " COMMON SENSE " TROLLEY BASE.
 
 MOTORS AND CAR EQUIPMENT. 
 
 1 I I 
 
 car at a comfortable temperature in very severe weather. 
 Its cost may be roughly estimated on this basis. One ampere- 
 hour will cost from 0.8 cent to 1.5 cents. For convenience, 
 say i.o cent. If during an eighteen-hour run the current 
 averages 4 amperes, then the cost per day for the supply 
 of fuel would be 72 cents. The expenditure for fuel in heating 
 by stoves may be taken at about 10 cents. 
 
 Cleanliness and convenience are of course great considera- 
 
 ARTICLE 
 
 
 
 ARTICLE 
 
 
 
 ARTICLE 
 
 BOX (4 AMP.FUSE WIRE ) 
 
 ? 
 
 H 
 
 >' 
 
 2 
 
 N 
 
 
 FIG. 6 4 . THOMSON-HOUSTON CAR WIRING. 
 
 tions in favor of the electric heaters. These have been made 
 by imbedding wire or other metallic conductors in clay or a 
 similar substance, the whole being then inclosed in an iron 
 case which may be placed under the seats. The clay, by 
 preventing oxidation, permits a relatively larger current to 
 flow through a conductor of given cross-section, without 
 destroying it.
 
 112 
 
 THE ELECTRIC RAILWAY 
 
 FIG. 6 5 .-EDisoN CAR WIRING.
 
 MOTORS AND CAR EQUIPMENT. 
 
 LIGHTNING ARRESTERS. 
 
 The device used to protect the machinery from lightning 
 constitutes normally what is often called an open circuit. In 
 its simplest form, the' lightning-arrester circuit would show 
 simply a break of say one-eighth of an inch between two 
 points. In Fig. 66 this simple circuit is shown, and in its 
 relation to the other circuits of the car. 
 
 It has been demonstrated by experience, that if a high 
 potential be suddenly established at a point, A, Fig. 66, it can 
 more readily force a current across a break, B (of suitable 
 dimensions), than through the complicated windings of wire 
 around metal masses,, which together constitute the motors. 
 
 FIG. 66. LIGHTNING-ARRESTER CIRCUIT. 
 
 A steady and moderate potential, such as that supplied from 
 the dynamos, is on the other hand unable to force a current 
 across the air space, B, but sets up a proper current through 
 the motors. 
 
 The motor current thus produces in the two cases effective 
 resistances, varying widely as compared with that of the air 
 gap. This is traced to the self-induction of a path containing 
 many convolutions. 
 
 Seeking some analogy with more familiar mechanical phe- 
 nomena, we may liken this to the resistance known as inertia, 
 or the resistance to change of state. Its measure is the 
 measure of the work required to change the velocity of a 
 given mass by a given number of units of velocity in a given 
 time ; it is the time element which operates to our advantage 
 in the lightning discharge. The high potential at A can
 
 II4 THE ELECTRIC RAILWAY. 
 
 establish an arc at B in shorter time than it can establish a 
 current through the motor windings. The arc having been 
 established, the high potential is at once lowered by the flow 
 to. earth, there being nothing now to maintain a higher 
 potential than 500 volts. If between the trolley wire and 
 the ground no other circuit be prepared than that through 
 the motors, then the discharge would force itself through 
 some point in the windings where the insulation resistance 
 between the wire and the body of the motor or between the 
 wire and the general metal of the car (equivalent to earth) is 
 low. 
 
 Whether there is a lightning arrester or not, it is best to 
 connect the field windings next to the trolley wire, since the 
 discharge generally, though not always, enters from over- 
 head injury to the fields being less serious than injury 
 
 to the armature. 
 
 The arc at B having been established, and the motor thus 
 protected from Lne destructive effect of the discharge, a new 
 trouble arises, necessitating some means of quickly destroy- 
 ing the arc itself. This arc, or body of heated gas between 
 the points at B, is of very low resistance as compared with 
 the cold air previously separating them. The ordinary press- 
 ure on the line (500 volts) is therefore able to maintain this 
 arc, and will cause a very great current to flow over this path, 
 practically short-circuiting the motors. 
 
 The current, even if not great enough to melt the wires 
 leading to and from B, becomes great enough to materially 
 lower the potential at A, thus checking, perhaps stopping, 
 the motors, and causing a waste of energy. The effort to 
 prevent this secondary action has produced various lightning 
 arresters now on the market. Underlying nearly all these 
 forms may be found one of these two principles. First : to 
 destroy the arc by magnetic action repelling or attracting, 
 and thus weakening it. Second: to lengthen the air gap 
 so much that the arc cannot be maintained across it. 
 
 If the latter method be employed, means must be devised 
 for resetting the points with the proper distance between 
 them, or setting new points in position for the passage ot 
 another discharge. In the first method, the pole of a magnet
 
 MOTORS AND CAR EQUIPMENT. 
 
 Is placed near the air gap, and when the arc is formed the 
 reaction between the lines of force surrounding the magnet 
 and those surrounding the arc strains the latter to the point 
 of disruption. Unless the heat should cause the metal points 
 to " bead," and possibly to bridge over the gap, this device is, 
 of course, constantly ready for action. If the strength of the 
 magnet be properly proportioned, this beading can scarcely 
 occur. As a further precaution, the points may be of carbon, 
 although their disintegration by the heat may increase the 
 gap until the device becomes useless. One of the most suc- 
 cessful of the devices in which new points are put in position 
 after every discharge is theWasson arrester. The arc forms 
 between carbon buttons and the passage of the current 
 
 FIG. 67. WESTINGHOUSE LIGHTNING 
 ARRESTER. 
 
 FIG. 68. WIRT LIGHTNING 
 ARRESTER. 
 
 melts successively the fuses, causing a lever to drop into the 
 series of positions which bring the successive couples of 
 carbon buttons very near together. In resetting the lever to 
 its original position, the buttons may be rotated to place fresh 
 portions of their circumferences in the position of nearest 
 approach. This device seems to take care of as many dis- 
 charges as there are carbon couples which may be conven- 
 iently placed in the box, say four or five. In another device 
 the sudden expansion of a body of air, due to the heat of the 
 arc, caused an arm to rotate, as from A to B, Fig. 67, bring- 
 ing C and D successively into position in the compartments
 
 ! jg THE ELECTRIC RAILWAY. 
 
 E and F, respectively. This ingenious apparatus has not 
 been so widely tested as to demonstrate its relative or its 
 absolute merit. 
 
 In the Wirt arrester, Fig. 68, there is a departure from the 
 methods just described, in this : that the arc is supposed not 
 to be formed at all. It consists of a number of thin metal 
 plates laid alternately with thin sheets of insulation, the 
 whole arranged in compact cylindrical shape. One end plate 
 is connected to line, the other to ground. In order to com- 
 plete a circuit through this system of plates, the current must 
 jump from the periphery of the first plate to that of the sec- 
 ond, then on to the third, etc. After proper experiment, a 
 certain number and thickness of plates and insulating sheets is 
 found, such that the dynamo current cannot, while the ordinary 
 lightning discharge can, force its way from edge to edge of 
 the plates. That ordinarily no arc is formed, is due to the 
 fact that each of the many air gaps is very short, while its 
 breadth (equal to the circumference of the disk) is consider- 
 able. Occasionally there has been some " beading " across 
 from plate to plate, producing finally a short circuit with 
 respect to the motors. Generally speaking, however, this 
 device is reliable, and has the advantage of compactness and 
 economy. 
 
 The principle of "dividing the arc," exemplified in this 
 arrester, is important, and has other useful applications. 
 
 GENERAL INSTRUCTIONS FOR THE CARE OF MOTORS. 
 
 It should be well understood that it is as necessary properly 
 to care for electric motors as that a locomotive should be 
 kept clean and in good working order. Careful attention 
 given to all the parts of any piece of machinery insures to it 
 longer life and from it better service. Electric - railway 
 motors have a very hard duty to perform. Rough and dirty 
 tracks, severe strains, heavy loads, etc., all tend to wrench 
 and shake the motors to pieces, and it is therefore very nec- 
 essary that every individual part of the apparatus should 
 receive the most careful attention. In order that this matter 
 maybe well understood and emphasized, we have arranged a 
 number of rules, which, if followed in their spirit, we believe
 
 MOTORS AND CAR EQUIPMENT. I I/ 
 
 will greatly aid those to whom the care of motors is given in 
 keeping the apparatus in the best possible condition. 
 
 There are so many manufacturers making electrical rail- 
 way apparatus, and these change so frequently, in greater 
 or less degree, the details of their apparatus, that it is impos- 
 sible to give to-day any set of rules which may not in consid- 
 erable part be inapplicable within the next few months. It 
 must always remain necessary that the more minute instruc- 
 tions required for the operation of apparatus shall be obtained 
 directly from the manufacturer of that apparatus. 
 
 The rules given we therefore call " General Instructions 
 for the Care of Motors," and they are as follows: 
 
 A. Inspection of cars and their preparation for service. 
 
 The motors should be thoroughly cleaned and all oil, 
 grease, dust, etc., wiped from them; and all oil wells and 
 grease cups should be filled. The armatures, commutators, 
 and brush holders should receive especial care. An accu- 
 mulation of dust and oil is a good inducement for short cir- 
 cuits. All parts should be kept as dry and clean as possible ; 
 a very little vaseline or paraffme on the commutator, how- 
 ever, if carbon brushes are used, lengthens the life of the 
 brushes and seems to diminish the noise. 
 
 All nuts and bolts should be carefully inspected and seen 
 to be tight and in their proper places. Keep the gearing as 
 free from dirt as possible. Many motors now have their 
 gears inclosed in an oil bath, which besides diminishing the 
 noise keeps the gearing in very good condition. 
 
 Examine the wiring of the car to see that the connections 
 are correctly and securely made and the insulation of the 
 wires intact. See that the controlling apparatus is in good 
 condition ; this is very important, and too much care cannot 
 be taken to see that all controlling mechanisms, switch boxes, 
 rheostats, reversing switches, etc., are in the best possible 
 working order. At least once a week examine and carefully 
 clean the lightning arrester, remove all oil and grease from 
 the boxes, and carefully clean the bearings. Every fortnight 
 examine the armatures, insulation, and fields; repaint and 
 reshellac them if necessary. 
 
 The trolleys should also be inspected, and all bearings
 
 TI g THE ELECTRIC RAILWAY. 
 
 cleaned and properly lubricated especial attention being 
 given to the trolley wheel. There should be good contact 
 between the wheel and "leading-down" wires, otherwise 
 there will be injurious sparking at the wheel. The trolley 
 springs should be adjusted so that the trolley wheel shall be 
 pressed against the wire firmly enough to insure a good 
 working contact when running at full speed. 
 
 The brake mechanism in all its parts should be thoroughly 
 examined, for it is exceedingly important that this part of 
 the apparatus be in good condition. Cars should be fur- 
 nished with good sand if operated on hilly roads. 
 
 See that the car is supplied with a screwdriver, monkey 
 wrench, etc., an extra lamp or two, and an oil lamp for use 
 if the power should be cut off. 
 
 B. Operation of the cars. 
 
 ' i. When cars are left standing in the car house or on a 
 side track, see that the safety or cut-out switches are placed 
 so that the circuit is open. Generally there is some mark on 
 the switch to show the proper position of the switch lever. 
 The trolley wheel should be removed from the wire and left 
 in such a position as to relieve the trolley springs of their 
 tension. 
 
 2. Before placing the trolley wheel on the wire, be sure the 
 circuit is open at the switches, and before closing the switches 
 be .sure that the controller handle is at the "off " stop. 
 
 Before placing the trolley wheel firmly on the wire, let 
 the side of the wheel just touch the wire ; if any flash occurs, 
 except the spark which may be seen when the lamps are on, 
 then something is wrong, and the wheel should be kept off 
 the wire. Probably a switch which should have been open 
 will be found closed. 
 
 3- If there is a reversing switch on the car, see that it is 
 set properly before applying the power. 
 
 4- Before starting the car, raise the traps and see if the 
 commutators, armatures, gears, etc., are in condition for 
 work. See that the brushes and holders are in good order 
 and properly placed. 
 
 5- When ready to start, move the controller handle gradual- 
 ly but firmly. If the car will not move, throw the controller
 
 MOTORS AND CAR EQUIPMENT. 119 
 
 handle off and look at the several switches between the trolley 
 wire and the motors. Probably some of them will be found 
 open. If the car still refuses to move, throw on the lamp 
 circuit. It may be there is trouble at the power station, 
 and the lamps will tell the story, unless it happens that the 
 dirt on the track prevents a contact between the wheels and 
 the rails. If this is the case, press the switch stick between 
 the rear of one of the wheels and the rail, thus securing a 
 ground. On many roads, an insulated wire is furnished each 
 car for use in just such cases as this. 
 
 6. Do not attempt to run the car backward unless the trol- 
 ley is closely watched by some one who holds the cord in 
 his hand. 
 
 7. In throwing power on, move the controller handle step 
 by step, allowing the car to gain headway under one, before 
 advancing to the next step. Too sudden starting strains the 
 machinery and wrenches the gears, etc. 
 
 8. In throwing the power off, move the controller handle 
 gradually until nearly at the " off " stop, when it should be 
 turned the rest of the way with a snap; and care should 
 be taken that the power is off before setting the brakes. 
 
 9. The brakes should be set gradually, so as not to bring 
 an undue strain upon the gearing. 
 
 10. Never run down grade faster than the maximum speed 
 allowed on the level, and always keep perfect control of the 
 car. 
 
 1 1 . If the trolley jumps off the wire, a slight slowing of the 
 car may be felt, and at night the lights will go out ; stop 
 the car immediately, after throwing the controller handle to 
 the "off" stop. 
 
 1 2 . When by accident or otherwise the current from the 
 power station is cut off, throw the controller handle to 
 the " off "' stop, throw on the light switch, and watch for the 
 power. When the power is on, it will be well for a portion 
 of the cars only, say those having even numbers, to start 
 immediately, the others waiting a minute or two. If all 
 start at once, a severe strain may be placed upon the dynamos 
 and engines. 
 
 13. Never stop a car on a curve, except in case of accident.
 
 I2 o THE ELECTRIC RAILWAY. 
 
 This will save the gearing from unnecessary wear and tear, 
 and at the same time relieve the generators and motors from 
 excessive strain and resulting loss of power. Many break- 
 downs and troubles have resulted from unnecessarily stopping 
 on curves. The extraordinary amount of current required 
 to start a loaded car on a curve may endanger the insulation 
 of the motors. 
 
 14. Never reverse a car while it is in motion, except to 
 avoid serious accident, and then it must be done very care- 
 fully. It is very easy to overdo the matter, blow the fuses, 
 and thus perhaps become helpless to avert some second ac- 
 cident. 
 
 15. If there is a reversing switch on the car, be sure that 
 the power is cut off before throwing the switch. 
 
 1 6. In any case of reversal to avoid accident, turn the 
 reversed power on very gradually, as a little will be found 
 sufficient to stop the car quickly, while a sudden application 
 of the power might strip the gears. Break the current as 
 soon as the car stops. 
 
 17. Do not reverse a car with the brakes set, for the fuse 
 may be needlessly blown. 
 
 1 8. Run slowly over railroad crossings, curves, switches, 
 rough track, etc. Remember that the cars are heavy and 
 that shaking up the motors should be avoided as much as 
 possible. 
 
 19. Run through water very slowly and carefully, and in 
 examining the motors, be sure that no water drips from the 
 clothing or elsewhere upon them. Water on the field mag- 
 nets will soon cause them to burn out. 
 
 20. Be careful that nothing falls from the pockets on the 
 motors, and do not let any metal for example, an oil can- 
 touch the brass screws on the connection boards, or in any 
 way cross-connect parts of the motor circuit unless the trolley 
 is off. 
 
 2 1 . Do not run over sticks, stones, or wires ; they may be 
 caught by or knocked against the motors. Remember that 
 the motors are hung very low and are apt to strike such 
 obstructions. 
 
 22. If a motor is flashing badly at the commutator, or gives
 
 MOTORS AND CAR EQUIPMENT. 121 
 
 out a burning odor, or shows weakness in any way, it is be&t 
 to cut it out. Some companies provide switches to do this 
 easily. If the car has a cut-out switch, insert the key which 
 should be with it in the socket, with pointer up, and turn 
 the pointer around one-quarter of a circle toward the good 
 motor. Never attempt to move tliis sivitc/i unless the main motor 
 sivitc/i is off. 
 
 23. It may be that a fuse is blown, in which case try 
 another, and then two in multiple, which will generally be 
 sufficient for the necessary current required to start the car. 
 
 If the double one blows, there is probably a serious ground 
 or short-circuit, which needs careful attention. 
 
 24. Unless very familiar with their current-carrying capac- 
 ity, never put copper or iron wires in place of the lead fuse, 
 for it is the function of this fuse to " blow " whenever the 
 motors are endangered, and if a wire were used, perhaps the 
 current which it would allow to pass would damage the mo- 
 tors seriously. 
 
 25. A sure way to stop any electrical trouble in the car is 
 to remove the trolley from the wire. 
 
 26. In case of a lightning storm, keep cool, for there is 
 absolutely no danger ; a slight noise may be heard, however. 
 If the motors are damaged by lightning, the car will run 
 unsteadily or stop altogether. If lightning damages one 
 motor, cut it out. If both are damaged, pull the trolley 
 down and wait to be pushed back to the car house. 
 
 27. Never run down grade with the current on; but the 
 trolley should always be in contact with the wire on such 
 occasions, as it may be necessary to reverse the car. 
 
 28. Be careful in the use of sand on the track, as too much 
 will prevent good ground connections. A very small amount 
 will serve to keep the wheels from slipping. 
 
 29. The power should be shut off when passing a "trolley 
 break" or "section insulator." 
 
 30. In ordinary stopping of a car, always release the brake, 
 but do not let it fly just before coming to a dead stop. The 
 armature will then be able to gather up the lost motion in 
 the gears and shafting, and will be ready for a smooth start. 
 
 31. Do not attempt to make up time on grades or rough
 
 I22 THE ELECTRIC RAILWAY. 
 
 tracks; in fact, never, at the expense of the machinery, try to 
 make up lost time. 
 
 32. Supposing the car to be equipped with Sprague-Edison 
 apparatus (commutated fields), the following rules regarding 
 the use of the switch box will be found helpful : Never dally 
 in the movement of the handle. Move on slowly from step 
 to step, allowing the car to respond to each step and gather 
 headway before proceeding to the next step. In case the 
 switch turns hard, move the pointer to the notch by a suc- 
 cession of slight taps; this will prevent running over the 
 notch. 
 
 In case you do run over the notch one-third, when throw- 
 ing on power, move on to the next notch. In throwing off 
 power, pass the fourth notch quickly. In case (in throwing 
 off) you run over one-third of a space, go back to the last 
 one passed. 
 
 The best notches in throwing off are from five to three 
 and from three "off." Throw the pointer from the first 
 notch to " off " with a snap. When leaving the switch, even 
 for a moment, take the handle off and leave it in the car. 
 Always allow the car to gain speed at one notch before giv- 
 ing it another. Never throw the switch on the fifth notch 
 if the car does not start on the preceding ones. Never ap- 
 ply the brakes when the switch is turned on. 
 
 When necessary to go beyond the fourth notch, pass slowly 
 to the seventh, never stopping on the fifth or sixth: and 
 always use the seventh notch on grades. Of the lower 
 notches, the third should be used in preference to the first, 
 second, or fourth. Prepare for heavy grades by throwing 
 the switch on to the fifth and the last notch, when the car 
 is going its fastest at the foot of the grade. During wet 
 weather, however, if the car wheels slip while on the grade, 
 work the switch back to the fourth point, or even work it 
 back to the first point and "off " point, until the wheels get a 
 gripe, then work up gradually to the seventh notch. 
 
 33. Familiarize yourself with the peculiar noises made by 
 the apparatus, in order that you may detect in this way 
 whether the motors are acting properly.
 
 CHAPTER IV. 
 
 THE LINE. 
 
 OF an electric railway, the conducting system, as a whole, 
 has come to be known simply as "the line." 
 
 It may be considered : (A) Merely as a conductor. Tak- 
 ing this view of it, the proper material, its resistance, cross- 
 section, and weight are the points to be determined. 
 Copper is now, and will continue to be, the cheapest con- 
 ducting metal, so long as there shall be maintained, even 
 approximately, the present relations between the specific 
 resistances and prices of the various metals. (B) As a con- 
 ductor in place, insulated more or less highly, and supplying 
 current in a suitable manner to a number of cars. 
 
 Before calculations as to the line resistance can be made, 
 it is evident that the following conditions must be known : 
 (i) The initial pressure, i.e., the voltage at the station ; (2) 
 the average or the maximum demand for current; (3) the 
 allowable " drop " on the line at the average or maximum 
 load; (4) the average distance over which the current at 
 the average or the maximum load is to be transmitted. 
 
 I. Railway dynamos are now quite generally run at a 
 difference of potential, between the brushes, of 500 volts. 
 This value may fairly be considered as a compromise between 
 considerations of economy in copper on the one hand, and 
 safety to life and facility of insulation on the other. The 
 question of danger to life has been thoroughly and frequently 
 discussed in public hearings before municipal authorities. It 
 .seems that nothing need here be said further than this, that 
 as yet no human being has either been killed or seriously 
 injured by the shock due to a 500- volt continuous current; 
 though, of course, many persons while handling electric 
 apparatus have received such shocks. Several horses have 
 been killed by railway wires, while in other cases special 
 efforts to kill a horse in this way have failed. 
 
 123
 
 I24 THE ELECTRIC RAILWAY. 
 
 In the direction of insulation against loss of current, prac- 
 tically nothing remains to be accomplished. A higher press- 
 ure than 500 volts could be satisfactorily insulated, so far as 
 loss of current is concerned; but the fear of danger to life, 
 in all systems requiring in any way the use of bare wires in 
 exposed situations, will doubtless keep the service pressure 
 for railways on the city streets where it now is, from 500 to 
 600 volts. 
 
 2. The predetermination of the average or maximum load 
 cannot be made with exactness. Given a certain schedule to 
 be observed by a given number of cars, the power required 
 to perform that service varies with the condition of the track 
 and the running gear, the efficiency of the motors used, the 
 skill of the motorman, the weight of the cars and passengers, 
 the magnitude of the grades and curves, the number of stops, 
 and the relation between the number of cars simultaneously 
 ascending and descending grades. And again, even if all 
 these variables may be given particular values for a given 
 schedule, we must recognize the fact that quite frequently 
 the schedule itself is not maintained. In spite, however, of 
 these uncertainties, it has been possible to adopt rules which 
 are fairly satisfactory guides in the calculation of the copper 
 required. When the service to be performed over given lines 
 is known in a general way, we should then seek to define 
 (a) the maximum number of cars that may be simultaneously 
 on the line ; (b) the number that may be ascending known 
 grades and the speeds required ; (c) the number descending 
 grades on which gravity is sufficient to propel them ; (d) the 
 number on levels and the speed required; (e) the weights 
 (live or dead) to be carried; (f) the horizontal effort in 
 pounds required to propel a given weight, say one ton, over 
 tracks such as may be in question. This last is known as 
 the traction co-efficient. Given the above values, we may 
 easily calculate the whole work to be done, and may express 
 it in terms of horse-power, or finally, in terms of current 
 and E. M. F. 
 
 Thus, let us suppose twenty-five cars on the line, of which 
 ten are ascending grades averaging five per cent., and at the 
 rate of five miles per hour; ten are moving down grades,
 
 THE LINE. 125 
 
 requiring no current; five are on the levels, making ten 
 miles per hour. We will suppose each car and its motors to 
 weigh 10,000 pounds, and the passengers of each car to weigh 
 also 10,000 pounds. The work to be done is of two kinds: 
 first, that against gravity; second, that against the friction 
 of the running gear. Atmospheric resistance at street-car 
 speeds is small, and may be neglected, or considered as 
 combined with that of friction. The work done against 
 gravity is thus calculated: Weight lifted = 10 (cars) X 
 20,000 pounds = 200,000 pounds. 
 
 Height of lift per minute equals the rise per foot on the 
 grade(o.05 feet) multiplied by the number of feet traveled 
 per minute. Now, five miles per hour is equivalent to 440 
 feet per minute ; hence the total rise per minute will be 
 440 X 0.05 = 22.00 feet. 
 
 The work done against gravity per minute will be, then, 
 200,000 X 22 = 4,400,000 foot-pounds, a rate of 133.3 horse- 
 power. 
 
 For the track in question, let us assume that a horizontal 
 effort of 30 pounds per ton will overcome friction. (This is 
 a safe outside figure for all ordinary tram-rail track.) Then, 
 for the 10 cars, we have 100 (tons) X 30 ( pounds) X 440 feet 
 per min. = 1,320,000 foot-pounds per min., a rate of 40 horse- 
 power; also for the five cars on the levels, we have 50 X 30 X 
 880 feet per min. = 1,320,000 foot-pounds per min., a rate of 
 40 horse-power. Thus, for keeping all the cars in motion, 
 we need, 133 + 40 +4 = 21 3 horse-power. 
 
 This power, however, is that actually needed at the axles, 
 while that delivered to the motors must include all losses 
 in the motors and gearing. Suppose this to be 25 per 
 cent, of the power delivered to the cars. Therefore, we 
 must supply 213 ^0.75 =2 84 horse-power. Since we premise 
 that the pressure under which the current flows is about 500 
 volts, and since 746 watts = i horse-power, we may, with 
 sufficient accuracy, say that 1.5 amperes on the line will be 
 equivalent to i.o horse-power. Then, for the total current 
 we shall have 284 X i 5 = 426 amperes. In this detailed way 
 the power required may be calculated. But it is usually 
 sufficient to allow without such detail 10 to 15 horse-power
 
 1 2 6 THE ELECTRIC RAILWAY. 
 
 as the maximum per car for service over considerable grades 
 and in case the total number of cars is in the neighborhood 
 of 20. This is quite sufficient for 1 6-foot cars. For roads 
 having practically no grades, and for a large number of cars, 
 the allowance may be as low as 7 or 8 horse-power per car. 
 
 3. As to the allowable " drop " of potential, it is determined 
 principally by the requirement of a close uniformity of con- 
 dition, in order that the same motors, moving all along the 
 line, shall be able to give practically uniform results at all 
 points. It has been found that a variation of 10 to 15 per 
 cent, from the station to the point of lowest potential will 
 not injuriously affect the speed demanded of the motors. 
 
 In making the following calculations for the copper to be 
 used, we must have clearly in mind just what conditions we 
 wish to produce. There has been a great deal of loose 
 thought and loose calculation in the matter. Thus, have we 
 in view simply a fixed maximum "drop" of say 100 volts? 
 This may seem a definite condition, fixing the copper accu- 
 rately, yet it is not enough, save in very simple cases, to 
 insure uniformity in the determinations by different engi- 
 neers, even though they make the same assumptions concern- 
 ing the value of the rail return. By one of them the line 
 may be wired in such fashion that the maximum drop should 
 occur at many places under many conditions, while by an- 
 other it may be so wired that the maximum drop could 
 occur only in a few places and under few conditions. The 
 weight of copper in the first case would be considerably less 
 than in the second. 
 
 Again, have we in view a certain average drop, say fifty 
 volts? This in turn, standing alone, is not definite enough; 
 for we must ask whether we shall consider the average drop 
 to be determined under a fixed maximum load (expressed in 
 amperes or horse-power), or when a fixed number of cars is 
 in ordinary service; and further, shall this average be deter- 
 mined from readings taken on one car while running over 
 all the line or lines, or shall it be determined by readings 
 taken on all the cars at the same instant, or shall the average 
 readings of a round trip on each of many lines be required 
 to fall within the fixed limit? Further, if we aim at an
 
 THE LINE. 127 
 
 average loss as the determining condition, we must still have 
 in view some limiting maximum drop, otherwise the condi- 
 tions concerning the average drop might be obtained, while 
 the resulting service might be poor indeed. 
 
 If the copper placed on the line has a uniform cross- 
 section along the whole length, then the observance of any 
 given condition as to average drop will carry with it a fixed 
 maximum drop, and vice versa. This is because, with a 
 imiform cross-section and, as is usually the case, a tolerably 
 uniform distribution of load along the line, there must neces- 
 sarily be an approximately regular drop of potential from 
 the station to the farthest point at which service is performed. 
 If, however, the line be broken into many sections, each 
 supplied with current separately, this necessary relation be- 
 tween the maximum and average drop ceases to exist. Thus, 
 let us suppose the governing condition to be that the average 
 pressure on the line when cars are in regular service shall be 
 400 volts. Let us further suppose the line to be divided 
 into four separately fed sections. On one of these, the aver- 
 age pressure might be 300 volts, while on the other three it 
 might be 433 volts, the average for the whole line still 
 remaining at 400 volts , yet this would certainly not be the 
 condition aimed at by good engineering, since it is not desir- 
 able that the pressure should on any section be so low as 
 300 volts. Generally speaking, indeed, this may be said: 
 that the more complex the wiring, the more necessary is it 
 accurately to define the conditions desired, in order that like 
 results shall be reached by different calculators. This whole 
 matter is of constantly increasing importance, since there is, 
 especially on the larger systems of railways, an increasing 
 tendency toward the subdivision of the lines into separately 
 fed sections. 
 
 A very reasonable specification, and one of wide applica- 
 tion, may be thus expressed namely: on any section the 
 average pressure shall not be less than 450 volts, and the 
 minimum pressure shall not be less than 400 volts. This 
 rule will apply as well to the case in which the whole line is 
 electrically continuous as to the case of subdivision, since 
 the whole line in the case supposed would constitute, in the
 
 I2 8 THE ELECTRIC RAILWAY. 
 
 sense in which the word is here used, one section. It should 
 further be specified, that the average and minimum pressure 
 above mentioned should be found under conditions of maxi- 
 mum load for regular traffic. Reasonable discretion must be 
 used in determining for any such case what shall be called 
 regular traffic. 
 
 It would perhaps be increasing unnecessarily the amount 
 of copper required, to insist that the given pressure should 
 be found in case of maximum extraordinary load; as, for 
 instance, when snow covers the tracks, increasing the power 
 required for the propulsion of the cars in service, and usually 
 also causing an extraordinary demand for current to operate 
 snow plows and snow sweepers. 
 
 4. Having in view these specifications, we may next con- 
 sider the distribution of the copper which is to be put up in 
 the line. The following cases arise in practice: 
 
 CASE I. (Fig. 69). A single bare conductor, the trolley 
 wire, usually of uniform diameter throughout its length, is 
 used to convey all the current required for the car service 
 over a given line (or section of line). Only in case of a com- 
 paratively small number of cars and a comparatively short 
 line can this method be economically used. The line pressure 
 in such case must always be found continuously lower from 
 
 FIG. 69. 
 
 the station to the farthest point of service. No two points on 
 such a line could have equal pressure. 
 
 This first condition, geometrically illustrated, may be rep- 
 resented by a straight line of uniform thickness, as A B (Fig. 
 70) , in which, let A represent the position of the station. Over 
 any section measured from A toward B must pass, not only 
 the current required by the cars on that section, but also the 
 current for all the sections beyond. If, therefore, the resist- 
 ance in the section A D be just equal to that in the section 
 F B, the drop in voltage along the first section would be 
 greater than along the second, and still greater than along 
 the third and fourth. Thus, if the tangent of the angle
 
 THE LINE. 
 
 129 
 
 A E C represents the whole current supplied to the cars, and 
 if we now take A B as representing the total and A E the 
 average resistance over which the total current must flow, 
 then the vertical line A C may be taken as representing the 
 50 volts total "drop," bearing in mind the geometrical rela- 
 tion that the tangent of the angle, A E C, varies with the 
 
 AC E 
 
 ratio -T- analogous to the expression of Ohm's law C =^> . 
 A. iii -K. 
 
 Following this geometrical method, draw through m 1 a 
 line to A C, making the tangent of the angle at m 1 equal to 
 
 one-fourth that at A EC. This tangent will correspond to 
 the current required for the first section flowing over the 
 resistance A D -^ 2, and the vertical line A H will represent 
 the " drop " over the first section for its otvn service, and will 
 be one-sixteenth of the total "drop, "or, in specific value, 
 3.125 volts. (The fact that the load of each section is pre- 
 sumably distributed along its whole length, instead of being 
 concentrated at one end, renders this illustration inexact, 
 but none the less valuable as setting forth the principle.) 
 
 Through m" draw the line to J m\ making an angle with 
 A B oqual to H m 1 A. Then A J will represent the number 
 of volts required to force the current for the second section 
 over the resistance A E. Draw other oblique lines at m 3 
 and at m 4 , parallel to the lines through m 1 and m", and like- 
 wise intersecting A C. Then draw vertical lines through 
 D, E, and F, intersecting the oblique lines. The total fall 
 of potential over the first section is now seen to be A H for 
 its own load, H J for the load of the second section, J K for 
 9
 
 j, THE ELECTRIC RAILWAY. 
 
 that of the third section, and K L for that of the fourth sec- 
 tioneach of these latter quantities being two-sixteenths of 
 the total, or twice as great as the quantity required for the 
 first section itself. This is because the current of the second, 
 third, and fourth sections must be carried over the whole 
 length of A D, while its own current is carried over an aver- 
 age distance of A m 1 , only one-half of A D. Of the total 
 "drop " it appears, then, that T V + (A X 3) = T'G takes P lace 
 on the first quarter of the total distance. On the second sec- 
 tion, the "drop" in like manner is found to be f 5 g ; on the 
 third, fV; on the fourth, -fa. The actual pressures in specific 
 values would be at A, 500 volts; at D, 500 ( T 7 g X 5) = 
 478.2: at E, 478. 2 ( T 'V X 50) =462. 6; atF, 462.6 ( T 3 g-X 
 50) = 453- 2 ; at B, 450.0. 
 
 While the " drop " when referred to equal sections of the 
 whole line is thus seen to be irregular, this condition is not 
 generally to be considered objectionable. 
 
 CASE II. A continuous trolley wire, uniform in diameter, 
 is connectc 1 t: intervals of, say, from 500 to 1,000 feet to 
 a continuous feeder wire, also of uniform diameter. If, in 
 any case, more than one feeder should be connected at inter- 
 vals to the same section of trolley wire, the case would still 
 be analogous to the simpler one above, since in both there 
 is, over the distance in question, a uniform total cross-section 
 of copper, over which the current passes for the car service. 
 This method will be understood by reference to the diagram, 
 
 P.S, 
 
 i i i i i i i: 
 
 FIG. 71. 
 
 Fig. 71. S T represents the trolley wire ; F F ' the feeder (or 
 feeders) connected to the trolley wire at relatively short inter- 
 vals, perhaps every 500 feet. In this case, the trolley wire 
 may be of small diameter, since the maximum current over 
 any of its parts is only that required for the service done in 
 about 2 50 feet that is, half the interval between sub-feeders, 
 A, B, and K. A car placed at D, midway between A E and 
 I. C, would receive a part of its current from A E (the sub-
 
 THE LINE. 131 
 
 feeder nearer the station) and part along the feeder F F ', 
 through B C and the trolley wire back to D, the proportions 
 being in inverse ratio to the resistances of the two paths 
 divergent at A. Another portion would flow directly from 
 the station along the trolley wire. 
 
 In practice, the trolley wire used in this system of com- 
 paratively numerous sub-feeders has varied in size from No. 
 6 B. & S. to No o. In the system first described, the trol- 
 ley wire, having to carry the whole current a greater dis- 
 tance, has rarely been less than No. o, and has been as large 
 .as No. oo B. & S. 
 
 CASE III. A trolley wire, of uniform cross-section, is 
 broken into short lengths insulated from each other, each 
 length being connected at one point to the feeder, one feeder 
 thus supplying the current for a considerable number of 
 trolley-wire sections. This is analogous to the second case 
 just considered, except that the conductivity of the trolley 
 wire itself is of service only over the distance from each sub- 
 feeder to the end of that particular trolley- wire section. 
 
 For a car. at B, Fig. 72, the whole current required must 
 
 E B C B' H 
 
 FIG. 72. 
 
 pass over the feeder A, thence along a sub-feeder to C, 
 thence to B or B '. Now, although the current cannot come 
 over the trolley- wire sections between the station and B, yet, 
 as in the two preceding cases, the " drop " of line pressure is 
 continuous from section to section of the trolley wire, since 
 the pressure on this wire must be determined by that of the 
 feeders at the points of junction with the trolley- wire sec- 
 tions. The car at B, however, must be working under a 
 lower pressure than the car at B', if the load from C to B is 
 greater than from C to B'; but a car at E must always be 
 working under a higher pressure than a car at C, and so on, 
 from section to section. 
 
 This subdivision into short and independent trolley-wire 
 sections has been required in certain cases, as a safeguard in
 
 !-, THE ELECTRIC RAILWAY.' 
 
 case of fires along the line. Each of these independent sec- 
 tions may be supplied with a switch, conveniently placed on 
 one of the supporting poles ; and the operation of this switch, 
 placed at K in the sub-feeder connecting the trolley wire 
 with the feeder wire proper, enables the trolley wire to be 
 thrown in or out of circuit at will. A fusible plug may 
 readily be placed in the same box containing the switch, and 
 any short circuit of this particular line section may thus be 
 made automatically to cut the section out of service. 
 
 The same remarks concerning the feasibility of cutting out 
 these independent trolley sections (either at will, by the 
 switch, or automatically by the fusible plug) that apply in this 
 case (III.) apply in case V., as will be seen when the latter is 
 discussed. 
 
 CASE IV. A continuous trolley wire of uniform diameter 
 
 is fed .at intervals by independent feeders, each running 
 direct from the station. It is plain that each of these 
 feeders may be so calculated as to give, under assumed con- 
 ditions of load, equal line pressures at the points of junction 
 
 (A, B,.C, D, and H) with the trolley wire, Fig. 73. Then a 
 car at E would receive its supply of current partly from A 
 and partly from the point B, and these parts of the current 
 supply would be dependent upon the distances of the car 
 from the points A and B, respectively, and in inverse ratio to 
 these distances. It is, of course, not convenient to run a 
 very large number of separate feeder wires. The trolley 
 wire, therefore, is usually fed at a comparatively small num- 
 ber of points, and consequently the distance over which the 
 current is conveyed on the trolley wire alone is much greater 
 than in cases II. and III. The size of the latter has, there- 
 fore, usually been greater, the most general practice having 
 been to use a No. o copper wire, with a distance between
 
 THE LINE. 
 
 133 
 
 feeding points varying from 1,000 to 10,000 feet. Care 
 should be taken, in using this method, so to limit the length 
 of the intervals that a trolley wire of convenient size shall 
 not be required to carry a current which, over the fixed 
 resistance, will produce an excessive "drop." 
 
 A variation of this case consists in connecting two or more 
 of the feeders together at the point of junction of any one of 
 them with the trolley wire. Thus, at F the four feeders 
 may be made practically into one conductor, the four being 
 themselves in multiple with the trolley wire, which is sup- 
 posed to have at a point very near the station a short connec- 
 tion to the dynamos. The effect of such a connection as that 
 indicated at F will, of course, be to lower the pressure pro- 
 duced at B, C, and E, with the three longer feeders entirely 
 independent each of the other; but such a cross-connection 
 maybe advisable if, after beginning operation, the load at any 
 point, as at A, is found to be much heavier than was orig- 
 inally assumed. 
 
 CASE V. The trolley wire is.broken into sections insulated 
 from each other (Fig. 74), each section being supplied by one 
 
 or more feeders not connected with the other sections. As in 
 case IV., the potentials at A, B, C, D, and H may be made 
 equal; and, as in case III., the conductivity of the trolley wire 
 for service is lost, except as it serves to convey current from 
 the points of junction with the feeders to the car in any partic- 
 ular part of a particular section. The current for all the cars 
 that may be found in any one section must be carried wholly 
 over the feeder or feeders of that section, the feeders and 
 trolley wire of other sections being electrically independent 
 each of the other. It has been usual to make the trolley 
 wire for this case of about the same size as that for case IV. , 
 namely: No. o B. & S
 
 J34 THE ELECTRIC RAILWAY. 
 
 The great advantage obtained by this system lies in this, 
 that the effects of several classes of accidents on the line may 
 thus be localized, and each feeder, being supplied with a 
 break switch in the station, may thus be thrown in or out of 
 circuit at will, the other trolley sections not being in any 
 way affected thereby. If, then, the trolley wire be broken at 
 any point, or if there be a short circuit to ground, the move- 
 ment of the cars over the other sections need not be 
 interrupted. 
 
 In any of the cases discussed, it is frequently possible to 
 carry the cars across the break by the headway gained before 
 reaching it. If, however, there be but one feeder connection 
 to any trolley- wire section (not itself connected to the station) , 
 and if the break should occur between this feeder connection 
 and any cars in question, as when the break is at L and the 
 car at N, Figs. 72 and 74, the car would be deprived of 
 power to obtain the necessary headway, unless the break 
 should be so near the adjoining trolley section that the car 
 could run by its headway from that adjoining section across 
 the insulating joint between the trolley wires, and also across 
 the accidental break. 
 
 With respect to this matter of the ability to run across a 
 break in the trolley wire, cases II. and IV., Figs. 71 and 73, 
 show an advantage. In either of these cases, a car is able to 
 receive current either from its rear or from its front, accord- 
 ing as the break is in front of or behind it. 
 
 CASE VI. This consists of a combination of either IV. or 
 V. with III. It is of great value on lines having very heavy 
 traffic. The trolley wire alone, when of convenient size to 
 handle, is (for heavy traffic) not able to supply, without too 
 great loss, the current required by sections (between feeding- 
 in points) of such length as would be needed in order to 
 keep the number of independent feeders within reasonable 
 limits. The remedy is plainly to place between the feeders, 
 at their point of junction, copper in excess of that found in 
 the trolley wire. Such a connecting conductor is called a 
 mam. It is connected at two or more points with the trolley 
 wire. This combination must recommend itself in all large 
 systems of electric street railways.
 
 THE LINE. 135 
 
 Let us next determine for the cases above described the 
 area and, consequently, the resistance of the conductors that 
 will be necessary to carry the required current to the cars 
 under the conditions given. 
 
 We will assume, for simplicity, that we have 10 cars uni- 
 formly distributed over a distance of 2 miles ; that each one 
 of these cars requires a current of 1 5 amperes, which, at 450 
 volts, is nearly equivalent to 10 horse-power. 
 
 E E Volts lost 
 From Ohm slaw, C = -^- (i), or-~- = ~ = R (2) 
 
 The resistance of any wire is 
 
 length in feet X 10.8 
 area in circular mils ' 
 
 10.8 ohms being the resistance of a wire of length one 
 foot and diameter .001 of an inch. 
 Equating (2) and (3), we have 
 
 Volts lost length in feet X 10.8 
 
 C area in c. m. 
 
 and by transposing we get 
 
 length in feet X 10.8 X C 
 c. m. = ;^ T = (O 
 
 \ olts lost 
 
 From this formula, which holds for any metallic circuit, 
 we may readily find the area necessary for any required cur- 
 rent, with any desired per cent, of " drop," when the distance 
 of transmission is known. 
 
 Let us now, for simplicity, further assume in these calcu- 
 lations that the resistance of the return circuit is nil. This 
 puts all the resistance into the overhead line or lines, or, in 
 other words, reduces the area of the wires to one-half of what 
 it would be if the circuit were a double metallic circuit, each 
 side having the same resistance ; or, briefly, we may consider 
 only the distance out one way, not the return. 
 
 The irregularity of the "drop," when referred to sections 
 of equal length and resistance, has been shown in the geo- 
 metrical treatment of case I. We can, however, obtain a 
 general expression determining the distribution of " drop" 
 as between any assumed divisions of a continuous wire of 
 uniform size. We may represent the ten cars as regularly 
 placed along the line in Fig. 71.
 
 136 THE ELECTRIC RAILWAY. 
 
 The distance from the station to No. i is 1,056 feet, and 
 that is the distance between successive cars. 
 
 Let C be the average current required by each car ; for the 
 general case illustrated above, the number of cars by N, the 
 interval by d, the resistance of d by R. The total current 
 from the station will be N C, from the station to the first car 
 the " drop" will be N C R, from the first car to the second 
 (N-i) C R, from second to third car (N-2) C R, etc., the 
 "drop" over the last section being C R. The total "drop," 
 which we may represent by E, is made up of the sum of 
 these quantities or may be expressed : 
 
 E = CR(N+(N-i)+ (N-2)+ + ) (6). 
 
 The sum of the series in parentheses we know from alge- 
 braic demonstrations to be represented by the expression, 
 
 We then have E = C R - (7) ; for the resistance 
 
 2 E 
 of any section, R = 
 
 Now let us consider any section whatever, as that which 
 is in sections distant from the station. The current flowing 
 over this section will be represented by C (N m). The 
 "volts lost" will be C (N m) R. In equation (8) we have 
 the value of R, true for all sections. Substitute this value 
 in the expression for " volts lost " and we have 
 
 Volts lost = (N f= ^)xCX2E _ (N-m)X2E 
 
 (N' -f N) x C N 2 + N 
 
 Now substitute this value for " volts lost " in equation (5), and 
 for "length in feet" substitute "d," and for current (N m) 
 x C ; we then have 
 
 c m = d X (N m) X C X 10.8 x (N* N) 
 
 (N m) x 2 E 
 _ d x 10.8 x C (N* -f N) 
 
 ~^E~ (9) 
 
 The supposition of uniform size of conductor" has been in- 
 troduced above, hence this equation, true for any section, 
 gives in fact the cross-section necessary throughout the line.
 
 THE LINE. 137 
 
 Now suppose E = 100, N = 10, d = 1,056, we then have 
 
 c. m. = ii X 264 X 3 X 10.8 = 94,087.6 
 It is plain that formula (9) may readily be converted into 
 one having more direct reference to the whole length of line. 
 
 D 
 Let D represent that length. Then d = ^; substituting 
 
 this value for d in (9) we have 
 
 D X (N + i) C 
 c. m. = 5.4- -^^H- (10) 
 
 The quantity (N -f- i ) C is simply the total current required 
 for one more car than the number originally in view. E, it 
 is to be remembered, represents the allowable " drop" at the 
 end of the line. In this shape the formula is of very con- 
 venient application. 
 
 The determination of the area of the conductor made above, 
 as applicable to case I., is evidently applicable also to cases 
 II. and III. The differences in actual construction would 
 be as follows: In case I., the whole of the area, 94~,oS7.6 c. 
 m., would be found in a single bare trolley wire. In case 
 II. this area would be found partly in the bare trolley wire 
 and partly in a parallel feeder. Thus suppose a No. 6 trolley 
 wire to be used, of about 40,000 c. m., then the feeder would 
 have to be about 55,000 c. m. in area. To insulate a pound 
 of copper with the material used ordinarily in weather-proof 
 wires costs about 7 cents. The total additional cost of the 
 feeder, over the bare wire, is found by taking this cost per 
 pound and adding it to the cost of placing the feeder in 
 position and connecting it to the trolley wire. These two 
 items may be roughly taken at $100 per mile. 
 
 In case III. we must have a feeder of about 94,000 c. m., 
 since the continuity of the trolley wire is broken. The in- 
 creased cost over either of the other two methods is evident. 
 
 It has been assumed above, for convenience, that the re- 
 sistance of the rail and earth circuit is zero. 
 
 Because the resistance, under favorable conditions, is ex- 
 tremely low it has been the habit of some engineers thus 
 to rate it as zero, placing all the allowable resistance in the 
 overhead wires and reducing them to one-half the cross- 
 section that would be required, if the outgoing and the return
 
 j^g THE ELECTRIC RAILWAY. 
 
 branches of the circuit were supposed to be of equal resist- 
 ance, as in the case of the double trolley system. The total 
 weight of copper overhead would thus be made only one- 
 fourth of that needed for the double trolley system, where 
 the length and cross-section are both doubled, as compared 
 with the single trolley or ground-return system. The facts 
 do not seem to justify entire neglect of the resistance of the 
 ground return. Especially does this seem true when we 
 consider the changes of moisture in the earth, the rusting of 
 the track, the gathering of dust and dirt upon it (causing an 
 appreciable resistance between the wheels and the rails) and 
 the resistance of joints from rail to rail. 
 
 The specific resistance of iron is well known, being about 
 six times as great as that of copper. The cross-section and 
 the length of the rails may also be readily determined, hence 
 the actual resistance of the rail return (neglecting joints) 
 may be accurately calculated. It may be taken fully into 
 account^upposing only that the joints be so bonded together 
 that the rails may be considered as practically continuous. 
 
 It is in assigning a value to the earth as a conductor that 
 the greatest uncertainty exists. Since the earth becomes a 
 conductor through contact with the rail, it is evident that the 
 manner of laying the track and the nature of the paving laid 
 along the track must materially affect the quantity of current 
 which actually leaves the rails. Further, the nature of the 
 soil itself, both in the surface stratum and in the lower strata ; 
 the number of iron pipes (as for gas, etc.) placed in the 
 ground; the proximity of large bodies of water all these 
 variables will affect this quantity. Realizing that the soil is 
 usually to be found permanently moist at a distance of sev- 
 eral feet below the surface, electrical engineers very early 
 adopted the practice of connecting the rails with plates or 
 rods, which were sunk to a convenient depth. There was 
 some gain in conductivity, but later it was thought best by 
 some, on account of telephone disturbances, to aim rather at 
 the restriction of the current to the rail than at its dispersion 
 through the earth. To utilize perfectly all the metal in the 
 rails, specifications were accordingly drawn up looking to 
 their thorough bonding both longitudinally and laterally.
 
 THE LINE. 139 
 
 To this bonding it has by some been thought wise to add a 
 continuous copper conductor, supplementary to the rails and 
 the earth. The effect of this, however, is generally small, 
 except in so far as it performs the function of a bond-wire, 
 rendering more certain the otherwise difficult maintenance 
 of a good circuit around the joints ; and further, in that it 
 will cause the potential of the rails and the earth to be so 
 nearly identical as to prevent the occurrence of the slight 
 shocks sometimes given to horses while resting some of their 
 iron-shod feet on the ground, others being on the rail. 
 
 In view of all the uncertainty as to the quantities involved 
 and all the variety of practice in treating them, it need 
 scarcely be added that there has been considerable variety 
 in the results. Fortunately, any error of under-calculation 
 is readily corrected. The first service of the cars over the 
 line may be purposely brought to the conditions of maximum 
 load. The speed then, or, more accurately, the readings of 
 a voltmeter on a car, will furnish immediate data as to the 
 sufficiency of the copper, or will detect local trouble in the 
 rail connections. The erection of an additional feeder, or 
 some intelligent work on the bonds, will soon set matters 
 right. If, on the other hand, the pressure on the line shows 
 somewhat higher than was calculated, no harm is done and 
 the coal consumption has been permanently reduced. 
 
 While the assumption of any definite value for the rail and 
 earth resistance (called for convenience the return circuit), 
 or of any fixed jatio between this value and that of the 
 resistance overhead, must generally be in error, it will not 
 be unwise to present a rule which has given fairly uniform 
 results. 
 
 Subject to modification, whenever it is possible by actual 
 measurement to determine the rail-earth resistance, we may 
 use without large error the following modification of formula 
 (10): 
 
 , D X (N + i) C 
 c. m. =6.5- g- (n) 
 
 This calls for about 20 per cent, more copper than is required 
 by the assumption of zero resistance in the return circuit. 
 Should there be considerable differences in the quantities of
 
 I40 THE ELECTRIC RAILWAY. 
 
 current required by uniformly distributed cars, sufficiently 
 accurate results will yet be obtained by the same formula 
 (n), if the average current C be properly determined. 
 Should there be considerable "bunching" of cars, the re- 
 quired copper may be calculated separately by the general 
 formula (5) (increasing it by 20 per cent., as above in (i i) ), 
 the "length in feet" being taken from the station to the 
 middle point of the section (supposed of moderate length) on 
 which "bunching" is found; and the required areas may 
 then be summed. If it be desired to change the size of the 
 feeder, after placing any length S A from the station to A, 
 the area up to A may be determined as above, a fixed " drop " 
 being then allowed (less than the total allowable) and the 
 point A may then be taken as a new datum point of initial 
 pressure. By diminishing the size of the feeder over the 
 part nearer the station, we may maintain the total "drop" at 
 100 volts, while increasing the average "drop" beyond the 
 value it would have if a wire of uniform diameter were placed 
 along the whole line or section. 
 
 Failure to understand this relation has not infrequently 
 caused copper to be placed where it could not do the greatest 
 good. Thus, suppose a uniform wire gives a line pressure of 
 450 volts at A and 400 volts at B. It is found that, there 
 being heavy grades at B, the motors become unduly heated. 
 It is desired to increase the pressure, thus decreasing the 
 current required for a given amount of work. A length of 
 wire, S A = A B, is bought and (i) is placed between the 
 station and A, and connected at intervals to the trolley wire; 
 or (2) it is connected only at S and at A ; or (3) it is con- 
 nected at A, thence at intervals along A B to B; or (4) it is 
 connected simply at A and B. For raising the pressure at B, 
 the fourth is the most efficacious method ; and if increase of 
 pressure is needed at B, and is not important elsewhere, the 
 fourth method should be followed. The formulae already 
 given will show the distribution of pressures. 
 
 In treating cases IV., V., and VI., there are required two 
 determinations of different characters. First, to determine 
 the diameter of the feeders, the function of which is to carry 
 a certain current over a certain distance at a certain percent-
 
 THE LINE. 141 
 
 age of loss, the whole current to pass out of the feeder at 
 one point, the end. Such a calculation, made by formula 
 (5), is very simple. It should be borne in mind, however, 
 as explained above, that some assumed value for the earth 
 circuit must be used. The general formula (5) is 
 
 length in feet X 10.8 X C 
 c. m. 
 
 volts lost 
 
 If we were to use a double metallic circuit, this distance in 
 feet would be the distance out plus the distance in (return) . 
 For the single trolley system, however, we have in mind the 
 distance measured only in one direction. We increase the 
 value of the constant 10.8 by 20 per cent., as above, and 
 then have 
 
 length in feet X 1 3 X C 
 
 c. m. = (12) 
 
 volts lost 
 
 The value of C in this formula depends upon the number 
 and service of the cars which it is proposed to supply by the 
 feeder in question. The volts lost must, of course, be less 
 than the total "drop," since a part of this total must take 
 place on the trolley wire, or on trolley wire and main, as in 
 case VI. The terminals (feeding-in points) of the feeders 
 become, with respect to the trolley wire, points of initial 
 pressure ; and the second calculation to be made consists in 
 the determination of the length of the trolley section between 
 feeding points, if the trolley wire (of fixed diameter) be used 
 alone/ or the determination of the diameter of the mains for 
 a fixed distance between feeding points. This latter deter- 
 mination would be entirely analogous to that made above for 
 cases I., II., and III. The formula to be used is that for 
 earth-return (u). The same formula, slightly transposing 
 its terms, becomes 
 
 volts lost (= E) 
 D = C - m -6. 5 X(N+,)C 
 in which C is the current per car, or per short section into 
 which the line may be supposed to be divided, the volts lost 
 being the difference between the maximum allowable and 
 the " drop " on the feeder alone, from station to feeding 
 points. If the pressures at these points be maintained prac- 
 tically equal, and if the work between them be distributed
 
 I4 2 THE ELECTRIC RAILWAY. 
 
 uniformly, the point of lowest pressure (maximum " drop ") 
 will be found on the trolley line midway between feeder ends. 
 
 By the system of independent feeders the pressure may 
 readily be made higher at some point distant from, the station 
 than at some point near it. And if the work at any distant 
 point be very trying, the establishment of such differences 
 may be wise. 
 
 Since No. o has been largely used for the trolley wire, it 
 may be well to deduce from formula (13) the distance over 
 which it alone may supply current to a given number of cars. 
 Let us assume, as heretofore, that 400 volts is the minimum 
 allowable line pressure; assume, further, that we maintain 
 450 volts at the feeder ends. Then the total " drop " over 
 the trolley must not exceed 50 volts. A mean between the 
 Birmingham and Brown & Sharpe gauges gives 110,000 
 for No. o. Suppose, further, that we have about 10 cars to 
 the mile, or 150 amperes total current. Then 
 
 D = 1 10,000 g-^^ - 5,641 feet. 
 
 For any other car service the distance is readily found. 
 
 In large systems, having lines crossing and recrossing each 
 other, there will arise some rather perplexing questions as to 
 potential distribution. The simple formulas (11), (12), and 
 (13) will serve all practical purposes; since, after all, the 
 whole matter consists in a sometimes complicated application 
 of Ohm's law. 
 
 Except in the matter of the rail-and-earth circuit, all that 
 has thus far been said applies equally well to the single and 
 double trolley systems. The latter has now been almost 
 wholly superseded by the former. There are three reasons 
 for the survival of the single trolley method as the fittest : 
 first, greater simplicity of overhead turn-outs and frogs, in 
 so far as the mechanical operation of the trolley is concerned, 
 and this is the controlling reason; second, greater facility in 
 insulating the out-going from the in-going side of the circuit, 
 for when the rail return is used these sides are about 1 8 feet 
 apart, while with the double trolley wire they are from 8 to 
 18 inches apart; third, greater economy of copper in large 
 systems. This advantage is not as great as has appeared
 
 THE LINE. 143 
 
 from the comparisons already given in discussing the copper 
 calculations. An offset must be made by considering the 
 cost of bonding the rails thoroughly, as compared with the 
 cost of supplying and erecting the copper return, which would 
 serve instead of the rails and earth. In case of very light 
 service, the first cost may be less for the double than for the 
 single trolley system. A fourth advantage is usually claimed, 
 in that fewer wires are actually required to be erected, thus 
 diminishing the objections made to the whole trolley system 
 merely on the score of "looks." To this, the advocates of 
 the double trolley (for there are a few) answer that in either 
 system all feeder wires may be buried ; that the comparison 
 would then rest as between bare wires alone ; that the single 
 trolley wire requires one or two guard wires, stretched par- 
 allel to each "live " wire, and these guard wires require span- 
 wires for their support this being done to prevent foreign 
 wires from falling across the " live " wire and to the ground, 
 where they rriay or may not make such contact as will cause 
 them to be entirely burned out ; that the double trolley sys- 
 tem does not require guard wires, since in case of any foreign 
 wire crossing the two live wires it would at once be destroyed 
 and the trouble at the railway station would end. 
 
 There has been so little extension of this double trolley 
 system that it cannot now be stated whether or not the pub- 
 lic authorities would generally allow this difference of con- 
 struction as between the two systems. It does not seem 
 probable. In Cincinnati, Ohio, where the Cincinnati Street 
 Railway Company has produced the most notable example 
 of double trolley service, guard wires were, however, not 
 erected. 
 
 The advantages of the double trolley system are two : First, 
 it causes little interference with the telephone circuits that use 
 ground return in the neighborhood of the railway lines. 
 This has been the cause of much litigation, urged by the 
 telephone interests, endeavoring to force the use of that sys- 
 tem of electric railways which would least interfere with the 
 established telephone service. Thus far the courts have 
 ruled, for the most part, against such requirements. 
 
 The case of the telephone labors under this disadvantage:
 
 I 44 THE ELECTRIC RAILWAY. 
 
 that it strives to destroy the most practical system of railway 
 circuits in order to continue a relatively poor system of tele- 
 phone circuits. The remedy for the evils brought upon the 
 telephone service by the disturbing earth currents of the 
 railway service is to be found in the use of complete metallic 
 circuits for the one or the other or both. To apply the rem- 
 edy to the telephone circuits is to produce the only installa- 
 tion which may be called first-class, with or without consid- 
 eration of railways. The currents required for telephones 
 are exceedingly small; the forces by which they may be 
 disturbed are correspondingly small. To insist that when 
 passing through the earth these currents shall not be percep- 
 tibly disturbed by other currents, is to insist upon a practical 
 monopoly of the earth as part of an electric circuit. The 
 best English and European telephone practice, uninfluenced 
 by any trouble from railway currents, tends decidedly toward 
 complete metallic circuits. To apply, on the other hand, the 
 same remedy to the railways, is to impose seri'ous difficulties 
 in the way of the practical success of the operation of cars 
 over the complexities of switches, turn-outs, cross-overs, and 
 the like. It will afford a pleasing exercise of ingenuity to 
 plan overhead apparatus suitable for the following conditions : 
 The double-tracked road of one company has single-tracked 
 branches, one to the right, one to the left, at a given street 
 crossing, while continuing its double-tracked course beyond the 
 point of junction. The double-tracked road of another com- 
 pany crosses that of the first at this point of junction, running 
 parallel to the branches. The tw T o companies refuse inter- 
 change of current and want to use the double trolley system. 
 
 We cannot do better, in setting forth more fully the merits 
 of this controversy, than to give the opinions of the Superior 
 Court of Cincinnati, and the Supreme Court in the State of 
 Ohio. These are found in Appendix A. 
 
 The second advantage to be noted is this: that effective 
 insulation of the motor windings may be more readily secured 
 than in the case of the rail-return circuit. In the earlier 
 stages of the art, this was indeed an important advantage ; 
 for on the single trolley roads no accident was more common 
 than the " grounding " of armature or field coils. Previous
 
 THE LINE. 145 
 
 practice in winding for comparatively high potentials had 
 rarely to deal with the case in which the metal of the machine 
 was in fact part of the circuit. To produce a short circuit 
 through the body of the dynamo, it was generally necessary 
 that the insulation of the wires should break in at least two 
 points of different potential. When, however, one brush of 
 the motor was connected, as was often the case, directly to the 
 motor frame (this latter being hung on the axle, whence 
 the current passed to the wheels) , it needed a break at but a 
 single point to cause trouble. In the violent fluctuations of 
 current strength and magnetic density incident to car service, 
 there were often produced very high electromotive forces of 
 induction, and these, if not the normal pressure of 500 volts, 
 often destroyed the insulation of armature or field. In some 
 designs this trouble has been practically met by attempting 
 to insulate the frame of the motor from the car truck. This 
 will be seen to impose considerable mechanical difficulty. 
 Generally, so much improvement has been made in the details 
 of armature and field winding, that the matter has ceased to 
 be big with misfortune, as was once the case. 
 
 There are now in the United States not more than ten 
 double trolley lines, only one of which, that at Cincinnati, is 
 of what may be called a modern construction that is, built 
 within the last two years. Of single trolley lines there are 
 several hundreds. 
 
 The third method of power supply, involving an extensive 
 conducting system, is that through underground conduits. 
 Thus far, all efforts, and they have been many and ingenious, 
 successfully to insulate the conductors in a conduit, have 
 failed, save when topographical conditions were very favor- 
 able. Some of these efforts will be described elsewhere. 
 The principles governing the calculation of copper, as already 
 set forth, will of course apply to conductors underground as 
 well as overhead. 
 
 It should be stated that in using the rail-and-earth return 
 it is desirable to avoid setting up a high current density in 
 the earth at points along the line or at the station. When the 
 line current is very great there may occur localized earth 
 currents of such magnitude as to seriously interfere with un-
 
 I4 <5 THE ELECTRIC RAILWAY. 
 
 dergrotmd circuits or even to produce considerable electric 
 effects on metallic pipes or cables. As this condition is un- 
 likely to exist except in roads with very heavy traffic, where 
 railroads are necessarily inadequate, it w T ould seem best to 
 use in such cases numerous ground plates along the line and 
 especially at the station, placing them below the level of water 
 and gas pipes and in moist earth. This will both raise the 
 general conductivity and avoid high local current densities in 
 the earth. 
 
 We have now treated the subject-matter of this chapter in. 
 its more general aspects, and in Appendix B will be found a 
 set of instructions for the erection of an overhead line. 
 
 There are now on the market many varieties of insulators, 
 frogs, pole clamps, etc. Simply for the sake of consistency, 
 throughout the instructions, some special types are referred 
 to. Modifications may readily be made in such of the lan- 
 guage as applies only to particular forms or dimensions. It 
 is intended simply to illustrate the general mechanical rules 
 which should be observed. 
 
 We go into considerable detail in these specifications be- 
 cause many railway companies now prefer to do this work 
 directly rather than through contractors. 
 
 For much of what is valuable in them we are indebted to 
 Mr. W. E. Baker, of Boston, who prepared similar instruc- 
 tions, which we have slightly modified for this publication.
 
 CHAPTER V. 
 
 TRACK CAR HOUSES SNOW MACHINES. 
 TRACK. 
 
 FOR half a century steam-railway engineers have studied 
 track construction. Great advances have been made in the 
 last ten years, and change is still considerable; this means 
 that the present railroads are not perfect, are not final ; yet, 
 withal, the steam-railway practice is the best guide in a gen- 
 eral way for street-railway men using electric cars to follow. 
 True, the conditions as to pavement and wagon traffic are 
 widely different in the two cases, and render the street- 
 railway problem for a given weight of car or locomotive 
 much the more difficult of the two. But it remains that the 
 best track, considered simply as to the company's own traffic, 
 is in both cases the approved "T" rail construction found 
 generally, with varying detail, on our steam railways. 
 
 To-day the problem of track-construction especially that 
 of joint construction is perhaps the most important one 
 pressing upon street-railway men for their attention. 
 
 It is beyond the scope of this work to give a treatise on 
 track construction; we desire, however, to present a few 
 examples of good practice at least, of that which is good 
 now. Much must yet be learned in this matter. 
 
 Broadly speaking, light track construction must give way 
 to heavier construction. Meanwhile we show in Fig. 75 the 
 construction at present largely used by the West End Rail- 
 way Company of Boston; in Fig. 76 the track used by the 
 Pittsburg, Allegheny and Manchester Traction Company, 
 and in Fig. 77, that toward which the West End Compan) T 
 of Boston now leans. This last is noteworthy, in that the 
 chairs are electrically welded to the rail, the only bolts used 
 being those at the fish plates. To reach these from time 
 to time a joint box has been advised, which is shown in 
 Fig. 78. This permits ready access to the nuts on the out-
 
 148 
 
 THE ELECTRIC RAILWAY. 
 
 side of- the rails, so that they may be inspected and set up 
 frequently and with little trouble. One side of the box 
 serves as a fish plate, the bottom as a chair. 
 
 Through the kindness of Mr. F. H. Monks, General Man- 
 
 FIG. 75. TRACK CONSTRUCTION OK WEST END STREET RAILWAY COMPANY. 
 
 ager of the West End Street Railway Company, we also show, 
 in Figs. 79 and 80, a large number of rail sections used in 
 various parts of the country. The lighter, especially the flat 
 sections, have seldom been laid originally for electric-car 
 service. There is, however, almost no type of track so 
 
 CONCRETE 
 
 FIG. 76. TRACK CONSTRUCTION OF P.. A. & M. TRACTION COMPANY. 
 
 poor but that some bold operator has been willing to use it 
 as a destroyer of electric apparatus and of dividends. 
 
 We will close this very short treatment of this subject by 
 the following quotation from a very excellent and succinct 
 letter recently addressed by Mr. F. H. Monks, General Man-
 
 TRACK CAR HOUSES SNOW MACHINES 
 
 149 
 
 ager of the West End Street Railway Company of Boston, 
 Mass., to the Street Railway Journal : 
 
 " The coming construction for electric roads will contain 
 but few jimcracks to break, wear out, and rust away. Strength 
 and durability are surely the qualities which should be sought. 
 Since the problem is so nearly akin to that which steam roads 
 have solved, why not follow closely on their lines, so far as 
 is possible, adding only such parts, forms, and appliances as 
 may be necessary to conform to the difference in the condi- 
 tions under which the two are laid and operated ? The writer 
 suggests the following: 
 
 " Ties. Sound oak or chestnut, six and one-half feet long, 
 six inches face, five inches thick, placed three feet on cen- 
 
 FIG. 77. PROPOSED RAIL FOR WEST END 
 COMPANY. 
 
 FIG. 7 8.- JOINT Box OF WEST END 
 COMPANY 
 
 ters, and two ties set eight inches apart under each rail 
 joint. 
 
 " Gravel. Sharp, clean, free from loam and clay. Tamp 
 thoroughly under and between ties, particularly in case of 
 those carrying rail joints. 
 
 " Rail. A girder of sufficient total height to allow two inches 
 of gravel between top of tie and bottom of paving stone; 
 head of rail two and one-eighth inches, flat flange two and 
 three-quarter inches, web one-half inch to nine-sixteenths of 
 an inch, lower flange five and one-half inches to six inches. 
 Rail to be laid directly on ties and spiked thereto by railroad 
 hook spikes four and one-half inches long. Rail to weigh
 
 1 5 o 
 
 THE ELECTRIC RAILWAY. 
 
 about one hundred pounds per yard, and to be drilled for 
 copper connections, tie bars, and splice bars. 
 
 FIG. 79 TYPES OF AMERICAN RAILS FOR ELECTRIC TRAMWAYS 
 
 " Tie bars. Flat iron bars six feet apart, round and threaded 
 ends, to pass through \veb of rails and set tight by proper 
 nuts.
 
 TRACK CAR HOUSES SNOW MACHINES. 
 
 " Splice bars. Thirty-inch steel channel bars, well fitted to 
 rail, and so drilled for six bolts as to allow for expansion and 
 
 FIG. SO.-TYPES OF AMERICAN RAILS FOR ELECTRIC TRAMWAYS. 
 
 contraction of rail. Best quality bolts and nuts to be used. 
 Head of bolt to be continually struck with twelve-pound 
 hammer when nut is being set up.
 
 I5 2 THE ELECTRIC RAILWAY. 
 
 " Joint boxes. Heavy cast-iron boxes with removable covers, 
 corrugated tops, to be placed on and spiked to ties, and set 
 outside track at each rail joint, thus permitting of frequent 
 and inexpensive joint inspection and repair. 
 
 " As it is intended in this paper to consider the question of 
 track construction alone, no mention need be made of elec- 
 trical connection of rails. Pave with block stone in the usual 
 manner, using only clean, dry, sharp gravel. As authorities 
 differ regarding the question of placing rails so that joints 
 will be opposite or otherwise, each company must decide for 
 itself. It will be noted by this mode of construction that 
 steam-railroad practice is closely followed. Here are no 
 chairs, stringers, or other supports. There are but seven 
 parts in construction, namely, ties, rails, tie bars, spikes, 
 splice bars, bolts and nuts, and boxes. Aside from excava- 
 tion, gravel, and paving, this track should be built for about 
 $10,500 per mile of single track. Many will doubtless say 
 that such cost of construction is prohibitive, but a fair answer 
 thereto is, that it will solve the vexing question now con- 
 fronting electric street-railway men in all large cities, of how- 
 to build a track which will stand the severe strain now put 
 upon it, which can be maintained and kept in repair at min- 
 imum cost. The first cost above named is entirely justifiable 
 and warranted." 
 
 CAR HOUSES. 
 
 The arrangement of car houses for electric railways must 
 differ from that familiar in horse-railway practice in that 
 provision must be made for readily inspecting, removing, 
 replacing, and repairing the motors. To accomplish this 
 inspection, removal, and replacement it has been customary 
 to excavate, between the rails of car-house tracks, pits of four 
 to six feet in depth. When an easy approach from the street 
 can be made the tracks may be elevated, leaving proper 
 working space between the level of the track and that of the 
 floor. Some question arises as to the required length of pit 
 room as compared with the total length of car-house track- 
 age. This ratio must depend upon the number of injuries 
 to mechanism, which in turn depends on the quality of the 
 machinery and the quality of the service maintenance. 

 
 TRACK CAR HOUSES SNOW MACHINES. 153 
 
 It may be roughly stated that pit room should be sup- 
 plied for at least one car out of ten in service. If the excava- 
 tion and protection of the pit space be not costly it will be 
 well considerably to increase this proportion. We show in 
 Fig. 8 1 a very convenient machine used by the East Cleveland 
 Street Railway Company, Cleveland, Ohio, for the expeditious 
 handling of motors in the pits.. An hydraulic hoisting appa- 
 ratus having a vertical range of a foot or more is mounted 
 in a low truck which runs on light rails laid in the pit. The 
 cut shows a motor on the platform of the hoist ready to be 
 
 FIG. 81. HYDRAULIC HOIST FOR HANDLING MOTORS. 
 
 lowered or raised. It is clear that a motor or any of its 
 heavier parts may thus be handled with much less time and 
 labor, and with more safety, than by the " slings" or " main 
 strength and awkwardness " not infrequently employed. It 
 is in the pit that the closest acquaintance with the motor 
 itself can be formed. We know of one large company which 
 requires division superintendents, inspectors, and motor-men 
 to serve a period in the pit, until the electrical department of 
 the company shall certify to the fitness of the employee for 
 the discharge of his duty. 
 
 In order to store cars at night quickly, and to get particular
 
 154 THE ELECTRIC RAILWAY. 
 
 cars out quickly when ready, most companies of considerable 
 size have used transfer tables usually running near the front 
 of the car house. Those that have been employed for the 
 ordinary horse cars may, of course, be used for electric cars. 
 We show in Fig. 82 a transfer table used by the West End 
 Company of Boston. It is propelled by a cable, which, in 
 turn, is operated by a stationary motor, the controlling appa- 
 ratus for which is placed at one end of the transfer table. 
 Another feature of this installation consists in the fact that a 
 car may cross the pit to the leading-out track just opposite 
 that on which the car may be standing. The tracks are 
 made continuous across the transfer pit by short inclined 
 rails covering the difference of level between the floor proper 
 and the bottom of the pit, whereon are laid also the rails on 
 which the car crosses the pit. The crossings of these rails 
 
 FIG. 82.-SECTION OF TRANSFER TABLE. 
 
 with the rails on which the table runs are of course quite 
 numerous, but the great facility of car movement thus gained 
 is worth much to a road working its cars on a close schedule. 
 
 In order to avoid the frequent climbing upon the car top 
 to inspect the trolley, the East Cleveland Street Railway 
 Company has arranged a raised platform by the side of an 
 entering track to the car house in such a manner that the 
 inspector is, when on the platform, at the proper level to 
 inspect the trolley. 
 
 Such a construction costs but little money, and it has been 
 already a source of much satisfaction to the management. 
 Fig. 83 * has been drawn from a photograph of this arrange- 
 ment. 
 
 Near the car house preferably under the same roof 
 
 should be placed the repair shop. If, as in a larger system, 
 there be several car houses, the principal repair shop should 
 
 " 011 ^ "^ kmdness f Mr ' C ' W ' Wason ' General Manager East
 
 TRACK CAR HOUSES SNOW MACHINES. 
 
 155 
 
 be near one of the largest car houses, or, if that is not con- 
 venient, near the station. The machinery of such a shop 
 may be run conveniently by a stationary electric motor. A 
 company operating cars even as few in number as ten will 
 find it not unprofitable to become the owners of a few good 
 
 FIG. 83. PLATFORM FOR READY ACCESS TO TOP OF CARS. 
 
 machine tools. A lathe and a drill press are, of course, the 
 most generally useful. We cannot pretend to give here any 
 valuable advice as to just how far a railway company should 
 go in the manufacture of repair parts. Local facilities and
 
 156 
 
 THE ELECTRIC RAILWAY. 
 
 the personal qualities of the management must determine 
 that question. We can only suggest that when either tools 
 or men in the repair shop are found to be idle half their 
 time, question should be raised as to the wisdom of keeping 
 them in service. It may well be cheaper to carry a liberal 
 stock of spare parts. In the repair shop, perhaps even more 
 than in other parts of the plant, is the value of order and 
 neatness displayed. Have a separate bin for each kind of 
 supplies ; keep the floor well swept ; especially have an eye 
 to metal filings; keep the winders and their supply of 
 wire well away from litter of any kind; keep armatures in 
 racks, not on the floor, where they may easily be bruised ; 
 have plenty of light, either natural or artificial ; keep a record 
 of every armature and field repair, giving to each its number, 
 date, and character of injury; if you manufacture supply 
 parts look well to their cost, that you may be sure as to the 
 
 FIG. 84. -HYDRAULIC JACK FOR REPAIR-SHOP USE 
 
 wisdom of making them at home or purchasing from dealers; 
 if you rewind armatures or fields manage to have at least a 
 5OO-volt current if not something higher from an adjacent 
 lighting circuit with which to test the parts in the shop 
 before replacing in the motor, remember that your winders 
 should have good will as well as skill their capacity to do harm 
 easily is great; repair injured parts as soon as possible after 
 discovery of the injury ; have a care to get good mechanics 
 not electricians for the work of the shop. A man who has 
 been "handy" with tools all his life will devise economical 
 ways of doing the work. If the repair shop is of considerable 
 magnitude he will design a few special tools that will quickly 
 pay for themselves. This should not be construed as mean- 
 ing that the number of special devices needed is considerable. 
 As a suggestion for effort in the right direction we show
 
 TRACK CAR HOUSES SNOW MACHINES. 
 
 157 
 
 in Fig. 84 an hydraulic press, valuable for pulling off or 
 pressing on commutators, pinions, or gears, or almost any 
 other use involving a straight pull or push. 
 
 This machine was- made for its own use by the East Cleve- 
 land Street Railway Company. 
 
 SNOW MACHINES. 
 
 To remove snow wholly from city streets seems beyond 
 the power or the will of most municipalities. Certainly, in 
 many cities of the United States such removal would be a 
 task beyond reason. The work of the electric-railway com- 
 
 FIG. 85. SNOW PLOW FOR ELECTRIC ROAD. 
 
 pany is, then, to displace the snow from its tracks; for this 
 purpose " plows" and " sweepers" have been used. Both were 
 already familiar in the service of horse railways. For perfect 
 cleaning of the track both seem necessary. 
 
 It has been thought by the management of one of the 
 largest electric-railway corporations that the sweeper does 
 its work too well in this, that other vehicles, traveling on 
 runners, are so much interfered with by the cleanly-swept
 
 158 THE ELECTRIC RAILWAY. 
 
 rails, and in turn so much interfere with cars that it is best 
 to leave an inch or two of snow on the track, as when cleared 
 by plows. Unless the snow is very much compacted, or has 
 become a sort of frozen slush, it is then possible to obtain 
 fair contact between wheel and rail, thus keeping up the car 
 service. It is intended to prevent packing and freezing by 
 prompt and constant use of the snow plows. Holding these 
 views, the corporation in question has supplied itself with 
 plows only, and these of simple type, such as are used in 
 connection with horses. The plow is driven by two 15 or 20 
 horse-power motors, one for each axle, the motors being on 
 a platform protected by a cab, and driving through the me- 
 dium of sprocket chains. It is shown in Fig. 85. Such a 
 
 FIG. 86,-SNOw SWEEPER FOR ELECTRIC ROAD. 
 
 machine is of course less expensive to maintain than a 
 sweeper. Other companies, on the contrary, use sweepers 
 only. If the question of sleighs be of small importance, and 
 the snows rather light, it would seem that, as between the 
 two machines, if only one can be had, the sweeper would be 
 We show in Fig. 86 a snow sweeper run by three 15 
 horse-power electric motors, two for the axles and one for 
 the brushes, or revolving brooms. 
 
 The sweeper brooms are both driven from a single counter-
 
 TRACK CAR HOUSES SNOW MACHINES. 159 
 
 shaft placed on the deck, this shaft being on the same angle 
 as the broom shafts, and consequently parallel to both of 
 them The sprocket wheels are on each end of the counter- 
 shaft, one for the front broom and one for the rear broom. 
 The object of using but one counter-shaft on the deck of the 
 sweeper is that it gives more room to the motor-men. The 
 sprocket chain which drives the brooms raises and lowers on 
 a radius, so that the chain is always kept tight and the possi- 
 bility of its ever dropping from the sprocket wheels is pre- 
 vented. 
 
 A powerful machine that claims to combine the functions 
 
 FIG. 87. COMBINED SNOW PLOW AND SWEEPER. 
 
 of plow and sweeper is shown in Fig. 87. The flier, or 
 rotating brush and plow, has a number of steel blades disposed 
 much as are the buckets of a steamboat paddle wheel. It 
 also carries steel brushes, one brush parallel to and touching 
 each steel blade. The brushes extend beyond the blades 
 about two inches, thus clearing away the snow or ice which 
 has not been cut away by the blades. Each flier is driven 
 at a high speed by a motor of 15 to 25 horse-power. By 
 vertical adjustment, this machine will leave either a perfectly 
 clean road-bed or a light covering of snow, the rails being 
 clean in both cases. 
 
 The combination is expensive, compared to the simple 
 plows and sweepers, and whether its value is in proportion 
 to its expense remains for the present winter to determine.
 
 160 THE ELECTRIC RAILWAY. 
 
 Salt should not be used for removal of snow unless nothing 
 else is at hand. Its use is likely to cause rapid corrosion of 
 the rail bonds. The lesson most distinctly taught during 
 the few winters whose snows have fallen on electric railways 
 is this: "Do not let the snow get ahead of you." A watch- 
 man should inform the proper authority of any considerable 
 fall occurring after service hours, steam in at least one en- 
 gine should be kept up for immediate use, and, generally, 
 everything should be in readiness at all times for putting the 
 snow gang at work within one hour, or less, from the start. 
 Such a course is hard to follow, perhaps, but it pays.
 
 CHAPTER VI. 
 
 THE STATION. 
 
 ONE of the first questions that arises in the construction of 
 an electric railway is, naturally, the position and character 
 of the power station. It is not at all a simple matter, for 
 there enter into it considerations of a rather complicated nat- 
 ure, all having a direct bearing on the economy of the future 
 system. It is needless to say that the line should first 
 be roughly located, in order to arrive at an intelligent un- 
 derstanding of the conditions that must be fulfilled. This 
 done, the subject of the proper location of the station can be 
 taken up. 
 
 So far as its' position with reference to the line is concerned, 
 it is perhaps sufficient to say that, other things being equal, 
 it should be as nearly central as possible. In the chapter on 
 line construction enough is said to impress upon the mind the 
 conditions imposed by the distribution of the copper, but it 
 should be remembered that the line is, comparatively speak- 
 ing, a permanent investment, and from the relatively small 
 interest charges and repairs it does not involve an im- 
 portant part of the expenses. 
 
 In any given station, whatever its position, a considerable 
 item of the running expenses will be labor, and this is virtu- 
 ally independent of position. The factor that should go 
 farthest in determining the proper location is the availability 
 of fuel. If it is possible, the power station should be so 
 placed that coal cars can be run up to its very door and the 
 fuel shoveled directly into the coal bins. In some few favored 
 t>laces an arrangement even nearer the ideal has been found 
 possible, as, for example, in Scranton, Pa., where the street 
 railway power station is located just at the foot of a gigantic 
 pile of culm, refuse from the neighboring coal mines, so that 
 the fireman could almost climb up the side of the pile and 
 kick the fuel under the boilers. Under such circumstances, 
 
 IT 161
 
 !6 2 THE ELECTRIC RAILWAY. 
 
 especially as the culm is secured at a nominal price, the cost 
 of fuel is almost negligible. Ordinarily, however, one must 
 depend on the railroads for the transportation of coal, and 
 hence it is most desirable to place the station close beside a, 
 railroad track, even if the site be rendered somewhat more 
 expensive by so doing. If the coal has to be carted at all, a 
 few hundred yards more or less distance makes very little 
 difference, as the expense is largely in the handling. 
 
 Aside from fuel, water is the next prime necessity to be 
 considered ; and it goes without saying that if a site can be 
 found close both to railroad tracks and water of suitable 
 quality for use in the boilers the station should be placed 
 there, even if at a considerable distance from the point that 
 would be indicated by consideration of the line alone. In 
 small stations, where condensing engines are not to be em- 
 ployed, the matter of water supply becomes somewhat less 
 important, but still deserves careful examination. If water 
 for condensation is required it is almost imperative to get 
 within pumping distance of a plentiful supply. 
 
 In locating a railway power station, then, the first thing" 
 to be considered is nearness to the supply of fuel, and, after 
 that, central position with reference to the line and availa- 
 bility of water. 
 
 In finally determining the best location for a station noth- 
 ing but actual expense estimates will enable a decision to be 
 formed. If the station is placed far away from the center of 
 the system additional copper will be needed in the line to 
 compensate for its increased distance, but to offset this there 
 may be the gain that comes from cheaper fuel and water. In 
 two places, one convenient to the supply of fuel, the other 
 in the center of the system, there will probably also be 
 differences in the cost of real estate. 
 
 If to obtain cheap fuel it becomes necessary to move the 
 station so far from the center of the system that the interest 
 on the increased amount of copper necessary is greater than 
 the probable annual saving in fuel, it is sufficiently obvious 
 that it will not pay to put the station near the fuel supply. 
 The relative cost of real estate in the two places must also 
 be taken into consideration ; it often happens, however, that
 
 THE STATION. 163 
 
 a position at some point near a railway track, where cheap 
 fuel can be obtained, is also a location where real estate is 
 not unreasonably expensive ; while the center of the system 
 is quite likely to be near the center of the town, where land 
 is decidedly costly. In choosing between two possible sites 
 for a station the interest on real estate at the two points 
 should be considered, as well as the interest on the invest- 
 ment in copper and the saving of coal. A few rough esti- 
 mates will show the more economical location. In the 
 majority of cases it is possible to get near coal and water 
 without getting very far away from a reasonably central 
 position on the line, but now and then the question becomes 
 more complicated- and recourse must be taken to the estimates 
 just mentioned. 
 
 It may here be well to take some notice of the use of water 
 power. It is not under all circumstances that the water- 
 wheel can successfully compete with the steam engine, and 
 the question should be determined on its merits in each par- 
 ticular case where it arises. Estimates should be made of the 
 cost of installation of the wheels and the necessary water 
 ways to feed them ; the cost of water rights should be found, 
 and the approximate amount of extra loss entailed by the 
 generally more inconvenient location of the power station. 
 
 With these data can be compared the probable cost of coal 
 necessary for producin y the equivalent horse-power, and the 
 difference in investment if engines are to be employed. The 
 constancy of the water power must be carefully considered, 
 with reference both to its failure at a critical moment and 
 the regularity of supply necessary to give the wheels sufficient 
 margin of po\ver for good regulation. 
 
 It must be borne in mind that, particularly in a small sta- 
 tion where the variations in load are excessive, water is by 
 no means easy to regulate ; and the greatest care should be 
 exercised in the installation for the purpose of securing uni- 
 formity of speed, even under the large changes of output 
 required. This irregular output, that is a prominent charac- 
 teristic of most electric-railway plants, in contradistinction 
 from every other sort of installation, is a thing which should 
 l>e most watchfully considered in designing future stations.
 
 164 
 
 THE ELECTRIC RAILWAY. 
 
 It is hard, even by diagrams, to give any adequate idea of 
 the changes of load to which a small railway plant may be 
 subjected; they are extraordinarily great and exceedingly 
 sudden. Fig. 88, which is a fac-simile of a record for ten 
 minutes on a recording ammeter, may give some faint idea 
 of the condition of things. It will be seen that at one point 
 the output jumped from zero to 150 horse-power and back in- 
 side of a single minute, and during the latter five minutes 
 shown in the diagram there were no less than twenty-five 
 sudden variations of 50 to 100 horse-power, each taking place 
 
 25 
 FIG. SBDIAGRAM OF VARIATION OF CURRENT IN AN ELECTRIC RAILWAY SYSTEM. 
 
 within a few seconds. The road from which this record was 
 obtained is about four miles in length, and was operating: 
 seven cars at the time of the test. 
 
 The effect of such a state of affairs is twofold : First, the 
 regulating power of the machinery is taxed to its utmost, 
 and even the best governed high-speed engines do not 
 
 espond quickly enough to keep the voltage on the line 
 constant under such circumstances, while water-wheels are 
 totally unable to keep pace with changes so sudden. Second, 
 
 wing both to poor governing and great variations in actual 
 ichanical strain, the machinery, whatever it may be, is all
 
 THE STATION. 165 
 
 subjected to severe and unusual tests of its strength and 
 stability. 
 
 It therefore becomes necessary that in a railway power 
 station the foundations, engines, shafting and dynamos 
 should be of the best mechanical construction. The use of 
 heavy fly-wheels on the engine or the shaft of the water- 
 wheel is to be recommended, especially in small stations; 
 and all engines designed for this class of work should be 
 exceptionally strong and solid in construction, and most 
 securely bolted to an unusually firm foundation. The same 
 is true of all shafting that is to be employed, and the dynamos 
 should be fixed in position as solidly as possible. On extend- 
 ed street railway systems, running a considerable number of 
 cars over a comparatively level track, the variations in load 
 are of course much less, and consequently such extraordinary 
 care in station construction becomes unnecessary. But in 
 these, as in all similar cases, it is best to err on the side of 
 safety. 
 
 Not only should this condition of variable load that we 
 have mentioned affect the mechanical design of the station, 
 but, to a certain extent, it should also modify the general 
 arrangement of the plant with reference to the location oi 
 the various portions of the machinery. Especially is this 
 true in small stations where only one engineer is em- 
 ployed. 
 
 It is most desirable, under such circumstances, so to place 
 the engines, switchboard and dynamos that they can be 
 readily seen and their performance watched from a single 
 point. If, for example, lightning enters a station, and one 
 of the dynamos begins to blaze at the commutator, the en- 
 gineer ought to be able to reach the switches without rushing 
 the length of the dynamo room and around various portions 
 of the machinery. If a fuse blows, it ought to be in such a 
 position as to be readily noticed without going around a 
 corner to look for it. This condition of easy accessibility is 
 not difficult to fulfill, but is sometimes neglected. Wherever 
 the entire equipment is subjected to often repeated and 
 severe strains particular care should be observed in getting 
 all the safety devices and switches where they can be easily
 
 !66 THE ELECTRIC RAILWAY. 
 
 reached. In large stations where several men are constantly 
 employed the necessity does not become so imperative. 
 
 With respect to the general design of a railway power sta- 
 tion marked differences must necessarily exist between small 
 and large installations ; of course this is true of electric-light 
 stations as well, but to a far less extent, for the same reasons 
 that we have endeavored to impress on the reader in the 
 preceding paragraphs. 
 
 The differences that should exist between plants of various 
 sizes will be pointed out later, by giving the sketches of three 
 designs for widely divergent station capacities. As a general 
 principle, it is safe to say that the subdivision of power 
 should be carried rather further in railway plants than is 
 ordinary in the construction of an electric-light installation, 
 for the reason that the various component parts generally 
 being subjected to far greater strains, and worked at a point 
 much further from their normal capacity, both security 
 against accidents and general economy demand rather 
 smaller 'units of power than in cases where the load is, rela- 
 tively speaking, uniform and somewhere near the full 
 capacity of the machinery. A fundamental law of economy 
 in installations of any kind is to work at all times as near 
 the full capacity of the machines employed as considerations 
 of safety will permit. Dynamos and engines are both de- 
 signed to carry their normal rated loads economically and 
 continuously, and while in most power stations for railway 
 work the great variations require a somewhat larger factor 
 of safety between the normal and maximum load than in the 
 case of electric-light work, the same broad idea must govern 
 the arrangement of the plans for either. 
 
 In beginning the task of designing a railway power sta- 
 tion, the first problem is to decide on the amount of power 
 to be required, both for present needs and future exigencies. 
 Neglect of looking forward to the possibilities of a few years 
 hence has caused inconvenience in many an electrical instal- 
 lation. On the other hand, construction for the distant period 
 when traffic may be several times its present amount is cer- 
 tain to produce an inefficient station. 
 
 The best policy is to build at first fully up to the immediate
 
 THE STATION. 167 
 
 capacity anticipated, leaving the design of the station in such 
 form that additional engines and dynamos can be installed 
 whenever needed. If a fairly good estimate is made for the 
 initial requirements the plant can be increased at a rate 
 quite sufficient to keep pace with any probable demands. It 
 is worth noticing, too, that if the number of cars on a given 
 line is doubled the maximum capacity of the plant is not 
 increased in anything like the same ratio, unless in cases 
 where very large numbers of cars are involved. If, for 
 example, an electric road starts with twenty cars and provides 
 ample equipment for them, the addition of twenty more will 
 certainly not require doubling the capacity of the engines 
 and dynamos ; for as the number of cars becomes greater the 
 average output demanded comes more nearly to approximate 
 the maximum load, without very much increasing the latter. 
 
 Where only three or four cars are in use it is quite possible 
 that all the motors may be making severe demands upon the 
 station at the same time; for example, two cars may be on 
 heavy grades at the moment that another is starting, but 
 with forty or fifty cars the varying outputs of the several 
 motors tend to balance each other, and it is safe to assume 
 that, even with a ten-car line, at no time will all the mo- 
 tors be worked simultaneously up to their full capacity. 
 
 So, in deciding on the proper capacity for the plant to be 
 constructed, a very important factor is not only the size of 
 the road, but its size with reference to the maximum output 
 that the conditions of service will require. 
 
 Not only do these conditions go far to determine the proper 
 capacity of the various prime movers employed, but, in the 
 case of steam engines at least, they afford good reasons for 
 selecting one or another type of machine. As has already 
 been mentioned, in the chapter on prime movers, various 
 forms of valve gear for steam engines permit of very different 
 ranges of cut-off; for example, some of the automatic high- 
 speed machines in frequent use to-day allow the steam, when 
 necessary, to follow the piston throughout nearly the full 
 stroke, while, as a rule, the Corliss valve gear than which 
 there is nothing more perfect as regards the results obtained 
 when working under favorable conditions usually permits
 
 !6S THE ELECTRIC RAILWAY. 
 
 a range of cut-off corresponding to only about two-fifths of 
 a whole stroke. 
 
 It therefore goes without saying that when extreme varia- 
 tions in load are to be expected, as is the case in compara- 
 tively small roads, the use of the Corliss engine would entail 
 a cylinder big enough to supply the maximum output often 
 several times the average output working at a cut-off of 
 about two-fifths the stroke. This, as has been mentioned in 
 the chapter on prime movers before referred to, virtually 
 compels working, on the average, at a cut-off too short for 
 anything like good efficiency. To certain types of compound 
 engines the same considerations apply, although it should be 
 said that, as a rule, these machines, particularly when used 
 condensing, stand underloading rather better than simple 
 engines. In the case of very small roads, the choice is lim- 
 ited to high-speed engines with a wide range of cut-off ; of 
 these the types on the market are almost too numerous to 
 mention, and most of them give excellent performances. As 
 the size of the road increases and the ratio between maxi- 
 mum and mean loads approaches unity, a point will be 
 reached where it will become possible to employ economi- 
 cally large compound engines, with Corliss or equivalent valve 
 gear. Condensation, even in small engines of all classes, 
 should be practiced wherever water is available, unless fuel 
 is extremely cheap. 
 
 With very large stations, triple, and even quadruple, ex- 
 pansion engines are coming into use and do admirable work. 
 It is hardly probable in any practical electric road, however, 
 that so good economy can be obtained with these complex 
 machines as is usual when they are employed for other pur- 
 poses where the load is far more regular. 
 
 Some striking examples of this have been brought to the 
 authors' attention ; in one case a triple-expansion engine of 
 an excellent modern type was put into use in the power sta- 
 tion of a comparatively small road, with the idea that great 
 economy of fuel would be obtained; the actual result was 
 very disappointing, for by careful tests it appeared that the 
 electrical horse-power hour, instead of being obtained by 
 the use of about two and one-half pounds of coal, or less,
 
 THE STATION. 169 
 
 actually required between six and seven pounds a result no 
 better than could have been reached by using a far less 
 expensive and complicated simple, non-condensing, high- 
 speed engine, of the kind generally employed for such serv- 
 ice. It was one of the cases where neglect of properly 
 considering the question of load led to unhappy results. 
 
 If we were, then, to attempt to obtain a general idea of 
 the kind of engine suitable for any given station, we must 
 be able to predict, roughly at least, the ratio between maxi- 
 mum and mean loads. In a general way, whenever the 
 former is three times, of thereabouts, the latter, nothing will 
 give better results than a plain, solid, high-speed engine 
 with a heavy fly-wheel ; with the maximum load about double 
 the mean load, the various forms of Corliss and similar en- 
 gines come into play with excellent results ; and where the 
 load ratio approaches even more nearly to unity and the 
 road is of considerable size, requiring 500 horse-power or 
 over, it is probable that the best results will be obtained by 
 using triple-expansion engines. These are only approximate 
 figures, but they at least give an idea of the circumstances 
 that should cause the selection of one sort of engine rather 
 than another. 
 
 Perhaps the best insight into the methods to be followed 
 in laying out a power station for an electric road may be 
 obtained by considering plans for several plants of particular 
 sizes and the special circumstances that lead to their adoption 
 in each case. We shall select, then, three specimen cases, 
 and investigate them in considerable detail. First, we shall 
 take up a five-car road of average character ; second, one with 
 twenty-five cars or thereabouts, and finally, a large city sys- 
 tem with one hundred cars in regular service. These three 
 cases may fairly serve as examples by which to show the 
 various conditions that have to be met in designing a per- 
 manent and efficient power station. 
 
 Let us suppose, in the first place, that the problem before 
 us is to install in a town of ten or fifteen thousand inhabitants 
 a small electric road, not, as will be the case in larger places, 
 for the purpose of enabling the business streets of the city to 
 be reached from the suburbs, but to facilitate transit from one
 
 I/O 
 
 THE ELECTRIC RAILWAY. 
 
 part of the town to another. The track will probably be less 
 than five miles in length and usually a single track with 
 turnouts. 
 
 The grades, of course, are liable to be of almost any 
 amount, depending on the configuration of the country; but, 
 as a rule, no continuous grades of more than six or seven 
 per cent, are likely to be encountered, although there may 
 be short pitches of slightly heavier gradient nothing, how- 
 ever, of sufficient length to require particular consideration. 
 On most small roads of this sort the service is rather infre- 
 quent, the cars being run on from fifteen to thirty minutes' 
 headway, and five cars will generally be enough to give 
 ample accommodations. Such a road is typical of a large 
 portion of those now in existence and of very many of those 
 to be built in the future; for the electric road, from its com- 
 paratively small cost of installation, can be utilized in a vast 
 number of places Avhere otherwise there would be no street 
 railroads at all, or at most miserable horse affairs, giving 
 very poor service. 
 
 Starting, then, with the assumption of our five cars oper- 
 ating over not more than five miles of track and over grades 
 of only moderate severity, the question that must be consid- 
 ered is the character and size of the power plant required. 
 In such situations the cars employed are, for the most part, 
 sixteen or eighteen foot bodies on six-foot wheel bases; 
 these have sufficient capacity for the work, can be employed 
 to draw trailers if necessary, and are of reasonable weight. 
 The same considerations that in Chapter IV. were used to 
 determine the distribution of copper will serve fairly well for 
 determining the capacity of the power station. 
 
 It may be useful here to give a brief table showing approx- 
 imately the power required to drive a sixteen-foot car weigh- 
 ing, with its equipment and a moderate load of passengers, 
 eight tons, up grades from one to ten per cent., at the uniform 
 rate of eight miles per hour, which is not very far from being 
 a fair average. On very light grades the determining factor 
 of the power required is the condition of the track, and on 
 very well laid track the so-called " coefficient of traction " 
 should be fifteen or sixteen pounds per ton; on ordinary
 
 THE STATION. 171 
 
 street-car track it is more frequently twenty, and rises from 
 that figure to twenty-five, thirty, and in some cases even to 
 forty pounds per ton ; twenty pounds is quite nearly correct 
 for the average conditions, and the table is founded on that 
 assumption. 
 
 Per cent, grade. 
 
 Power at wheels. 
 
 Per cent grade. 
 
 Power at wheels. 
 
 3-5 
 
 6 
 
 22.5 
 
 6.5 
 
 7 
 
 25.5 
 
 9.5 8 
 
 28.5 
 
 13.0 
 
 Q 
 
 32 
 
 16 
 
 IO 
 
 35 
 
 19 
 
 
 
 4 
 
 5 
 
 This table gives the mechanical power required at the car 
 axle, to the nearest half horse-power. 
 
 The average commercial efficiency of the motors is here 
 taken at the figure that is to-day true for most of those in 
 use, of about sixty per cent. To a very considerable extent 
 grades compensate themselves, so that their effect on the 
 average power required is not by any means so great as upon 
 the variations in the power. Heavy grades mean a widely 
 fluctuating maximum load, but they increase the average 
 daily output by only a comparatively moderate amount. 
 
 With the ordinary car equipment of two fifteen horse-power 
 motors, and the usual speeds, from eight to twelve miles per 
 hour, experience has shown that five to six electrical horse- 
 power per car is necessary on nearly level track. Ordinary 
 grades upon the road will not, in general, increase this 
 amount greatly; they will, however, cause more or less 
 extended periods of exceedingly heavy output, rising to 
 twenty-five or thirty horse-power for each car on the grade. 
 
 On large roads, where these times of abnormal output are 
 infrequent, one can safely count on the figure just given for 
 the power required per car; under such circumstances about 
 ten indicated horse-power, at the station, per car will there- 
 fore prove quite sufficient. On a road like the one we are 
 contemplating, with, possibly, severe grades and only five cars 
 in operation, it may easily happen, for example, that a couple 
 of the cars may be simultaneously upon the grade and a third 
 starting under somewhat unfavorable circumstances. The 
 amount of current ordinarily taken in starting a car is momen-
 
 jj 2 THE ELECTRIC RAILWAY. 
 
 tarily more than fifty amperes, which at the ordinary voltage 
 corresponds to about 2 5, ooo watts; we therefore might have 80 
 or 90 horse-power demanded for a minute or tw r o, and a 
 longer call for power of 50 to 75 horse-power, depending of 
 course on the length of the grades which are to be sur- 
 mounted. 
 
 In towns where there are small roads such as we are con- 
 sidering, it frequently happens, too, that inordinate demands 
 for power are made on the occasion of a somewhat infrequent 
 theatre night, a political demonstration, a base-ball game at 
 some point far out on the line, and other public gatherings. 
 This usually means bunching three or four heavily loaded 
 cars, sometimes all the cars on the line, at one point, and 
 starting them out within a minute or two of each other; for 
 a small road is dependent for its revenue largely on its 
 willingness and ability to accommodate just such unusual 
 demands. 
 
 It would therefore probably happen that on our five-car 
 road there would be times when the output would have to 
 reach nearly or quite 100 horse-power for a few minutes at a 
 time, although it would be strange if the average electrical 
 horse-power required throughout the day should exceed 30. 
 The reader will therefore readily understand that, under the 
 circumstances supposed, a far greater margin of power is 
 required than in case of a larger system. 
 
 In a comparatively small place it is fortunately not difficult 
 to obtain a favorable location for a power station, and the 
 expense of securing a site convenient to fuel is generally not 
 great. Our station should be located as centrally as feasible, 
 and close alongside a railway track or wharf, where coal can 
 be easily obtained. It is highly desirable, even in so small 
 an installation, to have a spare engine and dynamo, although 
 it frequently is impracticable ; it may be laid down, however, 
 as a rule of fundamental importance, that a road should. never 
 be trusted to a single dynamo for continuous running, and 
 even if the total output required be small, two dynamos should 
 be employed ; for nothing damages the reputation of a street- 
 railway system more permanently than a breakdown that 
 requires the suspension of traffic for a day or so, and such
 
 THE STATION. 173 
 
 breakdowns are sooner or later bound to occur where only a 
 single dynamo is used. Railway generators are peculiarly 
 susceptible to injury from lightning, and for this, if no other 
 reason, the above precaution is necessary. For our specimen 
 five-car road, a proper outfit of dynamos would be two of 
 about 40,000 watts rated capacity each; this means an ability 
 to supply 100 electrical horse-power, or more if necessary, 
 and sufficient capacity in either dynamo to keep up a toler- 
 able service on the road if its mate should unfortunately be 
 disabled. 
 
 As regards the type of engine that should be employed 
 there is little room for dispute, for the only thing that would 
 answer the purpose properly is a high-speed simple engine, 
 belted direct to the generators, and having as much weight 
 in frame and fly-wheel and as wide a range of cut-off as is 
 practicable. If it can be run condensing, so much the better, 
 as in most cases of employing steam engines. Its capacity 
 should be, approximately, 80 indicated horse-power at one- 
 quarter stroke cut-off and 90 or 100 pounds steam pressure. 
 
 It will be observed that the rated capacity of the engine of 
 this plant is much less than that of the dynamos; and this 
 is the correct arrangement, for, as will be shown in the 
 chapter on efficiency, a dynamo can be run at half its nom- 
 inal output without serious loss of efficiency, while under 
 similar circumstances an engine not only loses from its 
 friction becoming a large portion of the total output, but the 
 machine per sc becomes less economical of steam. Occasion- 
 ally compound engines of as small size as that mentioned are 
 used, but they are not to be recommended for such a small 
 and widely varying output, as the increase in economy of 
 fuel is not sufficient to counterbalance the increased first cost 
 and the slighter greater complication and consequent danger 
 of trivial accidents, which, though easily repaired, will, where 
 a single engine is employed, cripple the entire road. 
 
 The boilers for the plant should be quite capable of supply- 
 ing steam for 100 horse-power easily, and rather more if 
 pushed a little. Two boilers should be employed, to permit 
 giving one of them careful attention. A single boiler is 
 undesirable for the same reason as a single dynamo. It is
 
 !74 THE ELECTRIC RAILWAY. 
 
 even better to have two boilers either of them sufficiently 
 large to handle the plant if necessary and employ them 
 alternately; this principle, which, although more expensive 
 in first cost, is cheaper in the long run, will be dwelt upon 
 in speaking of the arrangement to be employed in large 
 stations. 
 
 We have, therefore, for this first specimen station, an 
 equipment consisting of two 4<D,ooo-watt dynamos, one 80 
 horse-power high-speed, simple engine belted directly to 
 them, and two boilers of about 50 nominal horse-power each. 
 
 As to the type of boiler to be employed, it is largely a 
 matter of taste and convenience ; probably, for general use, 
 nothing is much more satisfactory than the plain tubular or 
 return tube types, well set and provided with ample furnace 
 capacity. 
 
 Excellent results are obtained from the water-tube form of 
 boiler, of which there are now a number of meritorious pat- 
 terns. We should hesitate, however, to advocate their use to 
 the exclusion of the regular tubular boiler. 
 
 In any case it must be remembered that the intrinsic 
 differences in boilers are vastly less than the individual 
 differences introduced by general care and skillful firing. 
 Boilers should be periodically cleaned and overhauled in the 
 most thorough manner; hence the necessity for a spare 
 boiler, which will permit this to be properly done without 
 suspending operations, unless in part. It should be remem- 
 bered always, then, that a plain boiler with a skillful fireman 
 will give better results than the most expensive and elab- 
 orate form of patented boiler with a poor fireman. 
 
 A very convenient arrangement for such a small station 
 is shown in Fig. 89. Everything is under a single roof and 
 on one floor, with a stack at one side; the floor space is 
 divided into two nearly equal portions, one devoted to the 
 boilers and coal bins, the other to the engine and aynamos. 
 The engine is placed at the end of the dynamo room close 
 to the boilers, and belted directly to the two generators with 
 slightly different lengths of belt, so as to render both ma- 
 chines more accessible. 
 
 The switchboard is placed as shown, on the wall close to
 
 THE STATION. 175 
 
 both engines and dynamos, so that the engineer, sitting in 
 any convenient spot, can watch the operation of the entire 
 plant and be almost within reaching distance of the appa- 
 ratus. Of course circumstances will necessarily vary the 
 plans of the station, but the example given is convenient 
 in its arrangement. 
 
 As regards material, the station should preferably be a 
 substantial brick structure , if necessary, however, a frame 
 
 FIG. 89. ARRANGEMENT OF POWER STATION FOR A FIVE-CAR ROAD. 
 
 building answers the purpose fairly well. If, as is sometimes 
 the case in either construction, an iron roof is used, it should 
 be carefully sheathed, either by roofing felt or by a wooden 
 ceiling at the eaves; if this precaution be not taken, there 
 is likely to be constant trouble from moisture condensing on 
 the machines under variations of temperature, a contingency 
 which is carefully to be avoided. For a permanent station, 
 a brick building and substantial chimney should uniformly 
 be erected.
 
 !~6 THE ELECTRIC RAILWAY. 
 
 The foundations are a matter of prime importance ; both 
 engines and dynamos should be given a most solid bed con- 
 structed of rubble and cement, or brickwork, as convenience 
 dictates, but far more substantial than would be employed in 
 supporting ordinary machinery. The dynamos ought to be 
 carefully insulated, the wooden base on which they are usu- 
 ally placed being generally sufficient for this, provided proper 
 foundations are employed. The interior fittings and switch- 
 board ought to be the subject of careful, thorough construc- 
 tion : for no money is saved by cheap and hasty work about 
 a station. Lightning arresters and the like should be given 
 non -combustible bases of a size large enough to avoid danger 
 of any woodwork catching fire in case of accident. The usual 
 insurance rules for electric installations will serve as a suffi- 
 cient guide for putting up the subsidiary apparatus. 
 
 Passing now to the next case to be discussed, let us con- 
 sider a road employing in the neighborhood of twenty-five 
 cars for regular service, and situated in a thriving city of a 
 size sufficient to give a reasonably heavy traffic. Let the 
 grades be about as in the former case and the cars be of 
 similar type. 
 
 For roads operating from five to twenty-five cars, probably 
 no better arrangement could be devised than an amplification 
 of the system just suggested, employing as the system in- 
 creases two or three engines and four or six dynamos, and 
 allowing from twelve to eighteen normal indicated horse- 
 power at the engines, according to the size of the system 
 and the severity of grades. Compound engines maybe used 
 with great advantage for machines of 100 horse-power and 
 upward, especially if it is possible to obtain water for con- 
 densation. As for a twenty-five-car road, it is on debatable 
 ground, where a question necessarily arises between the 
 slight economy to be gained by direct belting and the great 
 convenience of being able to operate any and all the dynamos 
 from either of the engines. At about the same size of plant, 
 too, the low-speed engine begins to come into play. 
 
 For a road of the size we are considering, an allowance of ten 
 or twelve indicated horse-power per car would nearly always 
 be sufficient, and a proper equipment for twenty-five cars will
 
 THE STATION. 
 
 consist of four 6o-kilowatt dynamos. When long cars or 
 snow sweepers are used, each should be reckoned as equal to 
 two ordinary cars. It is probably preferable to have recourse 
 to a countershaft, and this should be situated at one end of 
 the dynamo room and on very substantial foundations. The 
 dynamos, arranged as shown in Fig. 90, should each be driven 
 from a pulley provided with a friction clutch, enabling any 
 of the machines to be employed. The shaft itself should be 
 
 FIG. 90. ARRANGEMENT OF POWER STATION FOR A 25-CAR ROAD. 
 
 divided into two sections, connected to each other and to the 
 engines by friction clutches. 
 
 With an installation of this size two engines should always 
 be employed, belted direct to driving pulleys at the ends of 
 the line shaft ; these engines should be of similar or identical 
 pattern, and may be either simple or compound, as the con- 
 ditions of fuel expense may dictate. Corliss or similar 
 low-speed engines may very well be used under these cir- 
 cumstances, as, except in certain cases, the fluctuations of 
 load are not likely to be great enough to put such an engine 
 at any considerable disadvantage. With slow-speed engines, 
 either simple or compound, a rated capacity of 150 horse- 
 power is desirable for each. If high-speed engines having a
 
 i;8 THE ELECTRIC RAILWAY. 
 
 wide range of cut-off are employed, 125 horse-power nominal 
 capacity for each will probably be sufficient. The boilers 
 should be three or four in number, aggregating about 300 
 horse-power. Fig. 90 shows the general arrangement of such 
 a station as has just been described. In operating it, a little 
 judgment will enable excellent results to be attained. Dur- 
 ing a large part of the day one engine operating two, or pos- 
 sibly three, of the dynamos will handle the load easily and. 
 evenly; in the morning and the evening, and at times when 
 especially heavy work is required, all the dynamos and both 
 engines can be put into use. Under these circumstances, 
 repairs on the engines or dynamos are easily carried out 
 without interfering seriously with the regular service of the 
 road. If compound engines are used, they should be worked 
 condensing if possible. The cross-compound type is prob- 
 ably preferable, in the matter of ease of repairs and accessi- 
 bility, to the tandem patterns, although some of the latter 
 do excellent work. The fly-wheels in any case should be of 
 unusual weight at least 50 per cent, heavier than would be 
 employed in driving ordinary machinery. For roads oper- 
 ating over twenty-five cars, an amplification of the plans just 
 given works admirably, the size of engines and dynamos 
 being increased to meet the larger output required. 
 
 Taking up the case of a one-hundred-car road, something 
 of the same line of equipment may be followed, but the 
 dynamo units should of course be larger. For one hundred 
 cars six iso-kilowatt machines are desirable, allowing a 
 sufficient surplus of power to reserve one dynamo as spare 
 equipment. The same arrangement of the line shaft may be 
 advantageously followed. It is probably advisable, however, 
 to divide it into three sections instead of two, providing, as 
 before, friction clutches for each d/namo pulley and each 
 driving pulley. Usually two or tnree ngines may be used, 
 preferably three, as this number enables i better adjustment 
 of the load and permits one of the engines 10 be tnrown out 
 of use for repairs without causing serious inconvenience. 
 For so large a plant, triple-expansion condensing engines 
 are very strongly to be recommended ; three of 400 nominal 
 horse-power each would answer the purpose admirably.
 
 THE STATION. 
 
 1/9 
 
 Under these conditions the mean load can be kept reasonably 
 near the full output of the machines in use all the time. 
 
 Except for a few hours each day, two engines will do the 
 work and do it well, and one dynamo could be, and should 
 be, reserved to be thrown in only when it is absolutely 
 necessary. Slow-speed engines, with Corliss or equivalent 
 valve gear, are in their element in an installation of this size ; 
 and all the great advantage of their unusual efficiency can be 
 
 FIG. 91. ARRANGEMENT OF POWER STATION FOR A IOO-CAR ROAD. 
 
 enjoyed. The boiler capacity, aggregating say 1,200 horse- 
 power, should always be divided into five or six units, so 
 that the boilers can be thoroughly overhauled, one at a time, 
 without causing any special inconvenience. A good arrange- 
 ment for a one-hundred-car station is shown in Fig. 91. 
 
 In a road of this size it is decidedly advantageous to employ 
 some double-truck long cars, with two motors of 20 horse- 
 power or thereabouts ; and these should be counted as the
 
 !80 THE ELECTRIC RAILWAY. 
 
 equivalent of two standard cars in making up the total equip- 
 ment. On small roads such cars are not advisable, as they 
 are very heavy and would have to be run very lightly loaded 
 a large part of the time ; but as the size of the system and 
 the traffic increase, their use becomes desirable, as in every 
 case where high speed is to be attempted. 
 
 In unusually large electric-railway installations, operating 
 several hundred cars, it is often advisable to employ an 
 arrangement radically different from the one just mentioned 
 
 that is, very large low-speed dynamos, coupled or belted 
 
 directly to triple-expansion engines, forming a combination 
 plant quite similar to that now generally used for ship light- 
 ing, but on an enormously larger scale. The advantages to 
 be found in this description of plant are, first, a saving of 
 perhaps five per cent., owing to dispensing with the counter- 
 shaft and belting; and, second, compactness, an advantage 
 not to be despised in large cities, where land is exceedingly 
 expensive. The disadvantages are those that always attend 
 the use of a comparatively small number of machines ; for, 
 in case of accident, a considerable fraction of the plant might 
 be rendered for the time being unfit for use, thus interrupt- 
 ing the service. Unless the size of the installation is so- 
 great that five or six 300 or 400 horse-power dynamos are 
 required, the arrangement of countershaft that has just been 
 described will generally be found preferable, for it enables 
 any or all of the dynamos to be run from any engine or 
 combination of engines. This secures not only immunity 
 against breakdown, but also enables the load to be readily 
 adjusted so as to keep the engines and dynamos actually in 
 use nearly up to their full output, and consequently operating 
 at their best efficiency. 
 
 A number of our present electric roads have employed 
 for service aggregating as high as one hundred cars merely 
 an exaggeration of the plans just laid down for the roads of 
 the smallest size, some few designing engineers having car- 
 ried the subdivision of power to a very needless extreme. If 
 the conditions of high average load are fulfilled, such sys- 
 tems as those just mentioned may- work very economically; 
 but it is believed that with equal care the designs we have
 
 THE STATION. l8l 
 
 shown will produce somewhat better results in economy of 
 fuel, continuity of service, and small repair bills. As regards 
 the subdivision of power, one general rule may safely be 
 laid down, deviation from which is very likely to lead to high 
 running expenses or disastrous uncertainty of service ; it is 
 this: The number of power units should be just such that the dis- 
 abling of one of them ivill not interfere with the successful operation 
 of the system. It will very seldom happen that more than a 
 single unit, composed of an engine and dynamo, or engine 
 and two dynamos, will be disabled at the same time. Owing 
 to the fact that accidents are rather more likely to happen to 
 dynamos than to engines, the engine units may be larger 
 than the dynamo units. In very small plants it becomes 
 necessary sometimes to take one's chances of being crippled 
 by an accident, btit every station should be so designed that 
 an engine or dynamo can be stopped for repairs at any time, 
 without seriously interfering with the car service. 
 
 Since writing the above, there has appeared a very valu- 
 able statistical article by Mr. J. S. Badger* on the operation 
 of electric roads ; and from this we cull some data regarding 
 station operation that are of the utmost value as emphasizing 
 the general principles of station design here laid down. The 
 very best performance as yet recorded for an electric-railway 
 power station is furnished by the East Cleveland road No. 
 4 of Mr. Badger's paper. 
 
 The station equipment consists of horizontal return tubular 
 boilers with Murphy furnaces, furnishing steam at 100 pounds 
 pressure to three 200 and three 125 horse-power Armington 
 & Sims simple, high-speed engines, belted directly to Edison 
 generators. Of these there are six No. 32 and six No. 16, 
 the former being of 80 kilowatts and the latter 40 kilowatts 
 rated capacity. 
 
 The road is in general level, with, however, one five per 
 cent, grade 400 feet in length ; seventy motor cars and 
 seventy trailers are on the average in daily use. One 
 electrical horse-power is obtained for every five pounds of 
 slack or four pounds of nut coal consumed, the evaporation 
 being 7^ pounds of water per pound of slack. The cars run 
 
 * The Electrical World-. October 31, 1891.
 
 !82 THE ELECTRIC RAILWAY. 
 
 on an average 91.1 miles per day each, and the consumption 
 of fuel per car mile is 4. 3 pounds of slack or a corresponding 
 proportional quantity of nut. This is a very excellent exam- 
 ple of correct proportioning of station capacity to work, 
 although even a better performance would have be'en at- 
 tained had compound condensing engines been used. 
 
 It will be noted that the aggregate capacity of the dynamos 
 at the full rated load was 980 horse-power, a little less than 
 half the rated capacity of the motors, thus following closely 
 the proportion we have suggested elsewhere. Further, the 
 engines were of a trifle less capacity than the dynamos, 975 
 horse-power, and at full output the dynamos would give \2 l / 2 
 electrical horse-power for each motor car. Trailers were 
 steadily used, a practice the advantage of which has often 
 been pointed out, so that the actual electrical horse-power 
 provided per car at the rated output of the dynamos would 
 probably be about eight at all events, somewhat less than 
 10; and this also is in accordance with the estimates we have 
 given. 
 
 In the actual operation of the station, the coal required 
 was but five pounds of slack for every electrical horse-power 
 hour, which corresponds very well with the evaporation given, 
 showing a consumption of nearly 35 pounds of water per 
 indicated horse-power, a thoroughly consistent figure, and 
 one not by any means unusually great for high-speed engines 
 running under variable load. 
 
 Had compound condensing engines been substituted for 
 the simple machines, preserving practically the same propor- 
 tion of engines to dynamos, even this excellent performance 
 could have been improved; since five pounds of slack, at an 
 evaporation of 7^, was required per electrical horse-power 
 hour certainly much more than would m have been needed 
 with large condensing engines. 
 
 Were any evidence needed to show that the good result 
 was due to proper proportioning of the station, it can be 
 readily found from the details of Mr. Badger's road Xo. 2. 
 In this case the engine power aggregated 675 horse-power, 
 while the dynamos were five in number, of 80 kilowatts each, 
 total about 535 horse-power; the engines were all high-speed
 
 THE STATION. 183 
 
 machines of standard make, running on a steam pressure 
 of 80 pounds ; the grades of the road were decidedly heavier 
 than in the case previously mentioned, the steepest being g l / 2 
 per cent, and about 400 feet in total length. Sixteen motor 
 cars only were in regular use, and although the company 
 had trail cars, they were seldom used. The full dynamo 
 capacity was a trifle over 35 horse-power per car, and the 
 fuel used was nut and slack, of which the consumption was 
 1:0 less than 1 1 pounds per car mile as against 4.3 on the East 
 Cleveland road. Although the conditions mentioned were 
 somewhat more severe in road No. 2, so far as economy is 
 concerned, the direful effect of the preposterously large en- 
 gine and dynamo capacity is only too evident in the enor- 
 mous consumption of fuel. 
 
 It will be noticed, in planning the model stations, that in 
 every case the whole installation has been shown on a single 
 floor, and that the ground floor; and whenever it is practi- 
 cable this is the best arrangement, because thereby immunity 
 from vibration and ready access to all parts of the plant are 
 secured. Sometimes, though rarely, in city stations, ground 
 space becomes so expensive that it must be economized so 
 far as possible ; it then becomes necessary either to use the 
 direct-coupled dynamos just mentioned, or to place the engines 
 on the ground floor and belt upward to the dynamos, either 
 directly or through the medium of a line shaft. This simple 
 procedure is likely to cause a great deal of trouble from hot 
 bearings. A number of such plants are running, usually for 
 electric-light apparatus, and in many cases they will be found 
 in operation with the caps removed from the bearings and 
 convenient means at hand for cooling the journals with water. 
 In any such case the dynamos should be provided with a 
 substantial support directly under them, and should not be 
 trusted to the supporting floor beams. It is far better to use 
 the direct coupled engine and dynamos where circumstances 
 compel contracted quarters for the station. 
 
 A word may well be said here on the subject of belting. 
 It is advisable to use leather belting of the very best quality, 
 special attention being paid in selecting it to closeness of 
 grain and pliability. There are now in the market a consid-
 
 !84 THE ELECTRIC RAILWAY. 
 
 erable number of patented beltings of various descriptions 
 canvas belting, perforated belting, and link belts of different 
 kinds; these various sorts are frequently used, and, as a rule, 
 give satisfaction ; but we have yet to learn that they are pro- 
 ductive of better general results than first quality plain leather 
 belting kept in good order 
 
 With regard to the minor arrangements of a station, there 
 are many details which can hardly find a proper place in this 
 volume, unless it be sufficient merely to suggest them. The 
 switchboard in any station should be placed where it is very 
 easily accessible and in plain view of the dynamos ; the other 
 apparatus should be easily visible from the regulating devices, 
 and these latter should be grouped together for convenience 
 and placed where they can be reached in a moment if any- 
 thing happens. Plenty of room should be allowed for 
 switches and cut-outs, and both should be of the most sub- 
 stantial description. Whether the cut-outs are magnetic or 
 merely fuses, they should be looked after at frequent inter- 
 vals and kept in good order, and never should be set to act at 
 the nominal rated capacity of the dynamos. It is an unneces- 
 sary refinement of cautiousness and will result in stoppage of 
 the cars without any good results, for any well-constructed 
 dynamo will carry for a few minutes nearly 50 per cent, over 
 its rated capacity without doing any damage whatsoever. 
 
 In case, for example, of a short circuit on the line* it is far 
 better to overload the plant by 25 or 30 per cent, and keep 
 the cars running, than to save the fuses at the expense of 
 stoppage of traffic; for fuse wire is cheap, but the loss of 
 public confidence that results from frequent cessation of cur- 
 rent on a line is a very serious matter, particularly when the 
 railway company has to ask any favors of the municipality. 
 The dynamo itself gives plenty of warning of an overload 
 before it is in any way injured ; and a judicious and watchful 
 engineer will very often save blocking the cars by keeping 
 his hand on the switch and his eye on the ammeter. Of 
 course it should be remembered that with a large station a 
 given short circuit on the line may cause no serious results, 
 while in a smaller station it might necessitate cutting out 
 the dynamos to save the armatures.
 
 THE STATION. 1 8$ 
 
 Every station should be equipped with a good set of testing- 
 instruments, consisting of a standard ammeter, voltmeter, 
 one or more magnetos, and a portable testing set for the 
 measurement of resistance. It is desirable that the ammeter 
 and voltmeter should be portable, so that they can be used 
 on the cars if necessary ; and in large stations special instru- 
 ments should be provided for this purpose. In addition, at 
 least one engine indicator should be provided better still, 
 two and for stations where there are compound or triple- 
 expansion engines, enough to use simultaneously on each 
 cylinder of a given engine. The expense is not great, and the 
 instrument is more than likely to save many times its cost in 
 fuel. A standard steam gauge for testing the gauges in the 
 boiler room is a useful addition to the equipment, also some 
 form of portable tachometer for getting the speed of engines 
 and dynamos. 
 
 It is worth while stating something here about lightning 
 arresters. Enough has been said in the chapter on motors 
 to give some-idea of the importance -of providing such appa- 
 ratus, but everything said about the danger to motors is 
 thrice true of dynamos. A railway generator, being a shunt 
 or compound wound machine of high voltage, is peculiarly 
 sensitive to the attacks of lightning; for there is direct com- 
 munication between the line and the armature, and a lightning 
 stroke entering the station is very likely to do mischief. 
 
 The direct damage done to the machine by the lightning 
 itself is usually small, and a powerful stroke is not at all 
 necessary to very disastrous results. The action is usually as 
 follows : the lightning enters the station and reaches the 
 machine in sufficient amount to puncture the insulation of 
 the armature; the current follows the spark, and the machine 
 promptly destroys itself. The extent of the damage may 
 go all the way from a few wires melted off to an almost 
 complete destruction of the armature windings and the 
 commutator. 
 
 It therefore becomes necessary in protecting railway ma- 
 chines against lightning to ward off if possible every particle 
 of the discharge. Ordinary lightning arresters, the function 
 of which is simply to provide an alternative course to earth,
 
 lS6 THE ELECTRIC RAILWAY. 
 
 are practically useless under these circumstances ; it is not 
 sufficient that the lightning may choose between going 
 through the machine and to ground through the arresters ; it 
 must if possible be compelled to take the latter course. It is 
 a well-known fact that both arc dynamos and railway motors, 
 in fact, all series-wound machines, are much less liable to 
 damage from lightning than shunt or compound wound 
 machines ; this is owing to the fact that before any lightning 
 can enter the armature from the line it must pass through 
 the series magnetizing coils of the machine, which possess 
 quite high self-induction and choke back the entering dis- 
 charge in a more or less efficient manner. 
 
 Bearing this in mind, the use of powerful impedance coils 
 situated between the lightning arresters and dynamos, as 
 
 FIG. 92. ARRANGEMENT OF IMPEDANCE COIL WITH LIGHTNING ARRESTER. 
 
 shown in Fig. 92 , is strongly to be recommended ; by offering 
 a powerful resistance to the passage of a sudden discharge 
 through them in the direction of the armature, they are 
 likely to drive the lightning to ground and save the machines. 
 Without such assistance it is quite possible for the lightning 
 to jump to ground through the lightning arrester,, and also 
 partially to discharge into the dynamo and destroy it. It is 
 quite impossible to give explicit advice as to the necessary 
 power of the impedance coil required, but one having ap- 
 proximately the magnetic dimensions of the field magnets of 
 the motors will probably answer the purpose. 
 
 As lightning usually enters a station from the line 
 rather than from the ground, it may be a useful precaution
 
 THE STATION. l8/ 
 
 so to connect the series coils on compound-wound dynamos 
 that they shall be between the armature and the line, and by 
 their considerable impedance help protect the former.* 
 
 On account of lightning, if for no other reason, a railway 
 dynamo ought to be carefully insulated from the ground ; for 
 under these circumstances the armature runs less risk, inas- 
 much as the lightning discharge is then not so likely to 
 jump to and through the armature core. Aside from this, 
 the frame ought not to be grounded, on the score of safety 
 to the men who are handling the dynamo. As to the pro- 
 tection of the station itself from lightning, it may be well 
 to suggest that in case an iron roof is used, as is not infre- 
 quent, grounding it at one or more points converts it into a 
 most efficient lightning rod. (See, further, Appendix E). 
 
 In the boiler room accessory apparatus should not be for- 
 gotten. In large stations, mechanical stokers are a highly 
 desirable addition to the plant, as they enable an economical 
 use of grades of coal that will much reduce the fuel bill.. 
 Separators and feed- water heaters should be regularly em- 
 ployed, unless the arrangement of the boilers with reference 
 to the engines is such as to render the former unnecessary. 
 For feeding the boilers, duplicate or triplicate means should 
 be provided, so that no easily supposable series of accidents 
 can prevent the proper feeding of water. If condensing en- 
 gines are used, care should be taken to have the supply of 
 water for condensation ample ; and in every case the engineer 
 should be watchful both of the quantity and quality of the 
 water for the boilers. Where good, clear river or pond 
 water can be obtained, nothing better is to be desired ; where 
 this is not available driven wells may answer the purpose. 
 If one is compelled to depend on the city water supply, the 
 bill is apt to be unpleasantly large, particularly if condensa- 
 tion is to be attempted, and arrangements should be made 
 for procuring an independent supply of water. 
 
 In the dynamo room there are divers devices that may be 
 profitably introduced. One of these is a permanent traveling 
 crane, supplied with differential tackle of sufficient power to 
 
 * This proceeding is useless where the series coils are shunted by a resistance.
 
 !88 THE ELECTRIC RAILWAY. 
 
 handle the component parts of any of the dynamos used. 
 With this apparatus the replacing of an armature or field 
 becomes a very simple and easy matter, while without it 
 much time is consumed and considerable danger of injuring 
 the machinery is incurred. It is advisable to fit up in con- 
 nection with the dynamo room a little workshop where 
 minor repairs can be executed ; for an engineer who is handy 
 with tools can oftentimes save his company a good deal of 
 money, and he should certainly be provided with a sufficient 
 kit of machinists' tools to encourage him in their use. Large 
 stations usually should have a repair shop connected with 
 them, and many companies even go so far as to rewind 
 armatures sometimes making an excellent job and saving 
 a very considerable amount of money thereby. 
 
 In constructing a station of any size whatever, room should 
 be left for an increase of plant not to be put in, however, 
 until it is needed. A few feet extra length in line shafting 
 .will easily permit the introduction of a couple of extra 
 dynamos in a station of some size, and in a very small sta- 
 tion ground space is usually cheap enough to permit leaving 
 room for an additional engine and pair of dynamos when they 
 shall become necessary. 
 
 As regards the general care of dynamos, little special in- 
 struction need here be given. As railway dynamos run for 
 the most part at five or six hundred volts, they are somewhat 
 more liable to short circuits, both in armature and commu- 
 tator, than are ordinary incandescent or arc machines. As a 
 result of this, they should be kept scrupulously clean ; the 
 commutators should be cleaned with great care, especially if 
 carbon brushes are used ; the brushes should be kept nicely 
 trimmed, and, in fact, extra precautions should be taken 
 throughout the station. 
 
 The variations in load on railway machines are apt, as we 
 have many times before stated, to be very violent and very 
 sudden, so that there may be more sparking at the commu- 
 tator than is usual in other dynamos. A sharp lookout 
 should be kept for this, and the brushes shifted whenever 
 necessary. 
 
 The carbon brush has come into very considerable use,
 
 THE STATION. 189 
 
 and a word regarding its employment is appropriate.* 
 Where carbon brushes are to be used a little extra work 
 should be spe'nt on the commutator, in keeping it perfectly 
 clean, otherwise it is likely to heat from short circuits caused 
 by carbon dust, and frequently exhibits the symptom of mi- 
 nute sparks flashing from under the brushes from segment 
 to segment. When carbon brushes are working at their best 
 they simply impart a grayish tinge to the entire commutator, 
 without causing either heating or flashing. If there is a no- 
 ticeably unequal blackening of the commutator it is likely 
 that trouble will follow, and the commutator should be rubbed 
 clean at once. 
 
 Dynamos of different makes and sizes cannot always be 
 run in parallel on a railway circuit, for the reason that they 
 may be magnetically unlike, and will not then respond with 
 qual promptness to magnetic changes; therefore, when the 
 load varies, as it may often and violently, the machines will 
 not respond equally, and the load will be unevenly distrib- 
 uted, causing unnecessary strains on one or more of the 
 machines. It is likewise very undesirable to run in parallel 
 dynamos driven by engines of different makes and speeds, 
 for when a sudden load is thrown upon the system the engine 
 governors will not respond with equal promptitude, throwing 
 the system out of balance and causing certain machines to 
 <Io far more than their share of the work . Engine makers will 
 often point with pride to the fact that their machines are so 
 well governed that they will not vary in speed more than one 
 or two revolutions per minute, when running empty and 
 when running loaded. This boast is sometimes true, but 
 nevertheless the speed of these very engines at the time of 
 throwing on the load may momentarily vary 10 or even 20 
 per cent. 
 
 A slow-running engine necessarily governs with less 
 promptitude than a high-speed engine, and in most cases the 
 
 * Probably the handiest way of trimming a carbon brush is to strap a piece of sand- 
 paper, face out, around the commutator after the current has been shut off for the night, 
 and drop the brushes into the brush-holders, letting the machine run slowly until the 
 sandpaper has smoothed off the end of the brushes into the proper form. A supply of 
 brushes thus prepared can be put in whenever necessary, and will run smoothly and 
 quietly from the very moment they touch the commutator. Take care not to let the car- 
 ibon dust get to the commutator connections, as it may make trouble.
 
 190 
 
 THE ELECTRIC RAILWAY. 
 
 practiced eye can detect at the moment of sudden load a 
 very marked slowing down of the fly-wheel, unless it be of 
 unusual weight. It is for this reason that heavy fly-wheels 
 are so strongly recommended, for they enable the engine to 
 tide over the period of sudden load with greater ease than 
 would otherwise be the case; it must not be forgotten, how- 
 ever, that no engine governor in general use to-day begins 
 to act until the speed has actually begun to fall, so the fly- 
 wheel is of particular service only in helping the engine 
 over very sudden changes. 
 
 It is worth remembering that from the high voltage of 
 railway dynamos and the consequent violence of any short 
 circuits that occur, the machines are subject to serious acci- 
 dents under circumstances which would lead to but trivial 
 results in low-voltage stations. It is therefore advisable to 
 cut out the machine as soon as it shows signs of trouble that 
 cannot be immediately removed by caring for the commutator 
 a little. The insulation should be frequently tested after 
 shutting down for the night, and every precaution taken to 
 avoid die beginning of difficulties, for when once begun they 
 grow with extraordinary rapidity. 
 
 In case of a short circuit on the line the engineer should 
 stand by the machines, watching them and the ammeter, and 
 t ready to cut them out if it becomes necessary ; it is a good, 
 safe rule, however, to hold on until the fuses go, in order to 
 avoid shutting down the line. With copper brushes a short 
 circuit on the line was frequently followed by a very consid- 
 erable injury to the commutator, but now that the carbon 
 brush has come into general use the machines will stand a 
 pretty severe ground on the trolley wire without danger. 
 
 Finally, the engineer of a power station should co-operate 
 with the superintendent of the road in keeping watch of the 
 performance of the entire system. The engines should be 
 indicated at least once a day, and the ammeter and voltmeter 
 readings should be taken simultaneously with the indications. 
 The amount and time of abnormally large currents, the 
 occurrence and severity of short circuits in particular, ought 
 to be noted in an engineer's log-book, as well as the effect of 
 any unusual service occurring on the line, such as the em-
 
 THE STATION. 19! 
 
 ployment of a snow plow, the effect of trailers, if they are 
 used, and the result of any particularly heavy service. If 
 possible, the amount of coal burned each day should be noted, 
 and also a rough check kept of the amount of water used. . 
 The former amount can sometimes very conveniently be 
 ascertained, if not every day, at least occasionally, by taking 
 a time when the coal pile can be easily cleared out and meas- 
 uring in barrels the coal used. 
 
 If, as is usually the case, particularly in stations of any 
 size, feed-pumps are used for supplying the boilers, the 
 weight of water delivered per stroke of pump can be very 
 readily determined; and by attaching to the pump a simple 
 stroke counter like the registering mechanism of a gas meter, 
 the total amount of water pumped per day may be at once 
 found; then, by bringing the water at the close of the day's 
 work in each boiler back to the point where it was at starting 
 up in the morning, the total amount of water evaporated per 
 day can be roughly ascertained. By taking the coal con- 
 sumption and water consumption in connection with the 
 indications of the engine and the simultaneous volt and 
 ammeter readings, the whole performance of the station can 
 be periodically determined without much trouble, and sources 
 of loss detected and estimated. 
 
 An engineer's log-book in which can be entered the data 
 mentioned is highly desirable, and in Appendix C is given 
 a convenient form of page for noting these various facts. 
 It is filled in so as to show in a general way its application. 
 The report can be and should be made as full as practicable, 
 and in case of special tests it is only necessary to fill the log- 
 book more completely than is shown in the example, taking 
 indicator and ammeter readings at more frequent intervals. 
 Now and then, at least, voltmeter and ammeter readings, at 
 intervals of a few minutes throughout the time of running, 
 should be made; and, in fact, the engineer of the power sta- 
 tion ought to take pride in knowing all the details of the 
 performance of the machines, for not only are these of im- 
 portance in helping him to operate the station successfully, 
 but they possess a money value to the company that employs 
 him.
 
 I9 2 THE ELECTRIC RAILWAY. 
 
 In closing this somewhat brief consideration of railway 
 power stations, we can do no better than to impress again on 
 the mind of the reader the necessity for solid and substantial 
 work throughout, the importance of using only the most 
 carefully made engines of extra weight and strength, and 
 the prime necessity of so arranging the installation that it 
 can be kept loaded as 'nearly as possible to its full capacity 
 throughout the hours of running. And finally, the last place 
 to economize is in the engineer's wages ; for a cheap engineer 
 can do enough mischief in one day, through sheer ignorance, 
 to eat up the dividends for an entire year. It requires an 
 unusually intelligent and skillful man to handle a railway 
 power station properly, and such an one must be obtained, 
 even at a cost considerably above that familiar to those whose 
 experience has been in other lines. 
 
 HINTS FOR THE CARE OF RAILWAY POWER STATIONS. 
 
 It may be well to append here, for the benefit of those 
 who have to do with the running of stations, a brief series 
 of hints on the special care that must be given to station 
 apparatus ; although what has previously been said is quite 
 sufficient for those who are already somewhat familiar with 
 the operation of dynamos such as are used for lighting. 
 
 BOILERS. 
 
 To begin at the boiler room, it is almost unnecessary to 
 state that scale is the fireman's worst enemy. Boiler scale 
 is the natural and logical result of insufficient attention to 
 the boilers. All water holds in solution a certain amount of 
 mineral matter, varying from the merest traces next to none 
 at all to quite a perceptible fraction of an ounce per gallon, 
 in mineral waters. Hard water, so called, is that containing 
 an unusual proportion of mineral matter, most frequently 
 salts of lime. 
 
 During the process of evaporation that goes on in the 
 boiler all this material is left behind, and unless the boiler is 
 periodically cleaned out, a deposit, calcareous in nature and 
 sometimes almost as hard as stone, will be formed wherever 
 the water touches the iron shell or tubes of the boiler.
 
 THE STATION. 193 
 
 In addition, there is likely to be a certain amount of sus- 
 pended matter in water particularly river water which is 
 .added to the general collection of foreign matter anql is likely 
 to coat the interior of the boiler with mud. We have seen 
 boiler scale a quarter of an inch thick and nearly as hard as 
 flint taken from the boilers used in an electric-light station ; 
 and, in fact, it is only too common to find scale allowed to 
 accumulate through carelessness, although seldom to so great 
 an extent. 
 
 The result is not only to diminish very much the evapora- 
 tive power of the boiler, but to expose the iron, unshielded 
 by circulation of water from the effects of direct heating, in 
 such a way as to rapidly deteriorate it and very likely to 
 cause a boiler explosion. The remedy for scale is to period- 
 ically clean out the boilers thoroughly and carefully, and to 
 use from time to time suitable scale preventers, which usually 
 act mechanically by hindering the scale from sticking to 
 the iron. A large number of these are in use, and perhaps 
 the simplest of any is crude petroleum, a small amount of 
 \vhich is placed in the bottom of the boiler after it has been 
 thoroughly cleaned and before the water is turned on ; as the 
 boiler fills, its interior is coated with a film of oil which 
 does not allow the hard deposit of scale to cling. 
 
 No general rule can be laid down as to the frequency of 
 boiler cleaning necessary, for it depends entirely on the 
 quality of the water used. A very little experience will 
 enable an engineer, under any particular circumstances, to 
 tell in what quantity deposit is forming and to apply the v 
 proper remedies at once. As mentioned previously, it is 
 best in installing any station so to arrange the boiler units 
 that one or more can be thrown out of use and thoroughly 
 overhauled without interfering seriously with the working 
 capacity of the plant. 
 
 It should be the engineer's business to see that the boiler- 
 room accessories are kept in proper condition, that the safety 
 valve is all right, the steam gauge registering correctly, the 
 joints tight, and the pumps in working order. Duplicate 
 means should invariably be provided for feeding the boilers, - 
 for one of the most serious accidents that can happen is the 
 13
 
 I94 THE ELECTRIC RAILWAY. 
 
 derangement of the feeding apparatus so that water cannot 
 be properly fed into the boilers. If injectors of the Korting; 
 or any other type are employed, a steam pump also should 
 invariably be supplied; and even if but seldom used, it 
 should be tested at frequent intervals to see that it is kept in 
 perfect working order against the time, which must inevitably 
 come, when the injector gets clogged and refuses to do its 
 duty. All steam pipes of any length should be jacketed 
 with some one of the many non-conducting coatings now in 
 use magnesia, asbestos, mineral wool, or the like. 
 
 ENGINES. 
 
 With respect to the engines, one general rule may be laid 
 down: watch their performance and overhaul them at the 
 first sign of trouble, for trouble goes on from bad to worse 
 with the greatest speed. 
 
 A well-made, well-set and well-cared-for engine is as reli- 
 able a piece of machinery as the ingenuity of man has yet 
 devised, but if ill-treated, even the best engine will go on 
 strike with extraordinary persistence. As previously men- 
 tioned, the foundations for engines to be used in rail- 
 way power stations should be extra heavy, and the engine 
 itself thoroughly well balanced. There should be very little 
 or no vibration when the machine is running, either in the 
 engine itself or in the steam pipes. If the engine is regu- 
 larly indicated, as it should be, the valves can be put in their 
 proper order very readily, and any abnormal features about 
 the engine noted at once. Any thumping, rattling, or unu- 
 sual irregularities in the running of an engine should be 
 investigated at the earliest possible moment and the proper 
 remedies applied. We cannot go here into the details of 
 running an engine, but can only give these general cautions 
 rendered all the more necessary in railway power stations 
 on account of the severe strains to which the machinery is 
 sometimes subjected. 
 
 DYNAMOS. 
 
 Some general hints have already been given as to the 
 proper care of this most important part of a power-station
 
 THE STATION. 195 
 
 equipment, but it may be worth while to go somewhat further 
 into detail, and to give an idea of the troubles likely to be 
 encountered and the ways of reaching them. 
 
 The list of faults to which a dynamo may be subject varies 
 somewhat with the size and character of the machine ; and 
 what applies to small motors, or to arc-light dynamos, is not 
 likely to apply with equal force to railway generators. These 
 latter are usually very well made and efficient machines, as, 
 indeed, they have to be to perform the severe service exacted 
 from them. 
 
 As regards their general arrangement, they should be on 
 firm foundations, readily accessible, and provided with means 
 for loosening, tightening, and aligning the belts. First of 
 all, they should be kept clean and dry, and usually, if well 
 cared for, will give very little trouble. However, every 
 dynamo is liable sooner or later, from one cause or another, 
 to operate in a somewhat unsatisfactory manner. Whatever 
 the cause of such abnormal performance may be, it must be 
 hunted up and the proper remedy applied at once. 
 
 Speaking in a general way, the troubles most likely to be 
 met are sparking, heating of the commutator, armature, or 
 field magnets, heating of bearings, and failure to generate 
 current at all; this last is unusual, and the cause may be 
 either very serious or very trifling. Of course, several of 
 these " bugs " often develop at once, as, for example, a pro- 
 longed overload is likely to cause sparking, and heating of 
 the commutator, armature and bearings. 
 
 Taking, however, the causes mentioned in their order, we 
 may note that sparking at the commutator may be produced 
 by a rather wide variety of causes. The commonest one 
 so common as to hardly require comment is a combination 
 of overload and wrong position of the brushes ; the former, 
 a glance at the ammeter will tell, and in addition it becomes 
 manifest by overheating of the armature, severe strains, and 
 sometimes even slipping of the belt ; the latter cause can be 
 remedied by moving the rocker arm to the point of minimum 
 sparking. 
 
 In the earlier railway dynamos it was quite common to find 
 a, sudden variation of load requiring a considerable change
 
 I9 6 THE ELECTRIC RAILWAY. 
 
 in the position of the brushes; in the later types, however,, 
 there is little or no change of lead with load, and whether 
 the current is one ampere or several hundred, the brushes 
 can be left very nearly in the same position. If you chance 
 to be operating a dynamo where the non-sparking position 
 of the brushes does shift, eternal vigilance is the only way 
 to avoid sparking at the commutator. 
 
 The cause of sparking just mentioned presupposes that the 
 machine is in good condition, but it is not altogether uncom- 
 mon to find sparking produced by faults either in the com- 
 mutator or in the armature. If the former is eccentric, 
 irregularly worn, or has one or more bars loose or set irreg- 
 ularly with respect to the others, sparking is sure to ensue; 
 for the tension on the brushes is irregular, they are thrown 
 into vibration, and the result makes itself evident at once. 
 An examination of the commutator will readily detect these 
 defects. A rough commutator needs no description ; and if 
 it is eccentric, the fact can be readily detected, either by a 
 visible wabbling or by holding a stick lightly against its sur- 
 face, when any irregularity in motion will make itself appar- 
 ent to the sense of touch. Loose or irregular bars are found 
 with equal ease. 
 
 In these days of carbon brushes, worn commutators are 
 not so common as they once were ; if from using copper 
 brushes and keeping them in the same position too long the 
 commutator should become untrue, it should be smoothed 
 very cautiously with a fine file, or very fine sand-paper 
 never emery-paper great care being taken to brush the 
 minute particles of metal out of the insulation between the 
 commutator segments before starting up the machine. In 
 very bad cases, a thin cut taken off the commutator in a lathe 
 will remedy the difficulty ; but this is considerable trouble, 
 and on some large machines a tool holder is arranged to be 
 fitted alongside the commutator, so as to turn it down very 
 easily by simply running at low speed and feeding the tool 
 by hand. Similar in its results to a rough commutator is a 
 rough brush. A little examination will show when the brush. 
 is irregularly worn, so as to make poor contact, and if so, it 
 should be trimmed or ground down.
 
 THE STATION. 
 
 I 97 
 
 Sparking" due to faults in the armature may be very severe, 
 but is almost invariably local in its character ; that is, affect- 
 ing- only a few of the commutator segments. This peculiarity 
 is very readily noticeable, and such an effect may be due 
 either to a short-circuited coil in the armature or a broken 
 coil, either one of which will produce violent sparking at the 
 commutator bar or bars most immediately concerned with 
 that particular coil. When a coil is short-circuited it heats 
 very violently, may burn out entirely, and is quite likely to 
 make itself obvious by a smell of overheated insulation. On 
 running the machine slowly the current may show pulsations, 
 and on stopping, running the fingers over the armature will 
 generally locate the short-circuited coil immediately by the 
 heating, particularly if one is guided to it by the visible 
 burning at the edges of the corresponding commutator seg- 
 ments. 
 
 Short circuits may occasionally be produced by stray par- 
 ticles of metal getting between the segments of the commu- 
 tator or among the armature connections. Under these 
 circumstances the trouble can be found by inspection and 
 easily removed. More often the short circuit is in the coils 
 themselves, and may usually be looked for at the head of the 
 armature. 
 
 Perhaps the most elusive of all faults in armatures is what 
 may best be described as a flying short circuit that is, a 
 short circuit between two or more of the armature coils of 
 such a character that it does not appear when the machine is 
 at rest, but becomes noticeable as soon as the revolution of 
 the armature, by centrifugal force or magnetic drag, presses 
 the offending \vires together. Under these conditions the 
 machine may fail to excite, as the shunt winding is completely 
 short-circuited ; there will be little or no heating, because of 
 the small excitation and little current flowing in the coils ; 
 and oftentimes even a careful measurement of the resistance 
 of the armature coils all around the commutator will fail to dis- 
 close the trouble, for the simple reason that when the arma- 
 ture stops the short circuit no longer exists. Such a fault is 
 perhaps best found by separately exciting the fields and then 
 running the machine rather slowly, when all the character-
 
 j^8 THE ELECTRIC RAILWAY. 
 
 istic signs of short-circuited coils will appear and can be 
 located by the consequent heating. 
 
 If there is a broken circuit in the armature, as sometimes 
 happens, there will be very serious flashing at the commu- 
 tator localized as before but there will be no special heat- 
 ing of a single coil. The defect may be sought at the 
 commutator end of the armature, as breaks in the wire are 
 most frequent where the connections are made with the 
 commutator segments. In either of the cases we have been 
 discussing it is best not to try any makeshifts, but to cut the 
 machine out until it can be properly repaired. A temporary 
 remedy sometimes applied in such a case is to cut off the 
 defective coil from the commutator bars with which it is 
 joined, and then temporarily to connect the disconnected 
 bars to those next succeeding ; but it is not advisable to do 
 this unless driven to it. 
 
 In addition to the causes of sparking already enumerated, 
 there is a tendency for sparks to flash around the commutator 
 over several segments. This is most often observed where 
 carbon brushes are used, and is usually a result of carbon 
 dust getting into the insulation between the segments and 
 over the surface of the commutator generally. It is often 
 accompanied by heating of the commutator, and never occurs 
 where proper care is taken. 
 
 Heating in either the field-magnet coil or the armature of 
 a dynamo is generally due to one of two causes overloading 
 or short circuits ; overloading can be told by the condition of 
 the ammeter, and most often results in railway power stations 
 from a ground on the line, although the fuses will usually 
 take care of the machines. The heating due to short circuits 
 in the armature we have already mentioned. In addition 
 there may be particularly when the machine is first set 
 up moisture in the armature coils, producing a general 
 case of short circuit through the insulation. Where this 
 is present the armature usually feels moist, and may even 
 steam. It is a rather unusual condition, and can be best 
 remedied by drying the armature very gently for a consider- 
 able time. 
 
 Passing a current through it is the easiest way to do this,
 
 THE STATION. 199 
 
 although the amount used should not exceed the regular cur- 
 rent for which the armature is intended. 
 
 Heating in field-magnet coils may arise from forcing the 
 voltage, thus sending through the coils a greater current 
 than that for which they were designed ; or from a genuine 
 short circuit in the coils, not a usual occurrence, and easily 
 discovered by measuring the resistance of the two or four 
 coils with which the machine is wound and comparing them 
 each to each. If the difference in resistance rises to more 
 than a few per cent., there is probably a short circuit. 
 Moisture may get into the field coils just as in the armature, 
 and is expelled in the same way. 
 
 Heating of the commutator may be due to general overload, 
 or to short circuits between the segments, due usually to 
 particles of metal or carbon from the brushes. The first 
 case may be detected by the ammeter, the second by the 
 localized heating and the sparking that ensues. Cleaning 
 very carefully is the only suitable remedy. 
 
 One of the commonest troubles encountered in an electric 
 station of any kind is the heating of the bearings in one or 
 more of the dynamos ; this is generally due either to lack of 
 oil or to excessive pressure coming from an overload. With 
 the ordinary types of railway generators, in which the bear- 
 ings are self -oiling, there is little likelihood of the first cause 
 producing any serious results, unless the oil well leaks or is 
 extraordinarily neglected ; and a glance at it will tell whether 
 the trouble is there to be sought. Occasionally the armature 
 shaft may be sprung, or the bearings may be out of line, 
 although neither of these conditions is common. In the 
 former case, the armature turns hard and is likely to stick at 
 a particular point; in the latter, the shaft still turns with 
 difficulty, but with nearly equal difficulty all the way round, 
 and the revolution becomes much easier if the bearings are 
 slightly loosened from their foundations. There is no help 
 for a sprung shaft, while the bearings, if out of place, can 
 be aligned. Neither of these difficulties is as common, how- 
 ever, as a hot bearing, due either to the pressure of the 
 .shoulder of the pulley side wise against the bearing or too 
 great belt tension.
 
 200 THE ELECTRIC RAILWAY. 
 
 In eithei case the bearing on the pulley end will be heated 
 more than the other. If the trouble is due to lateral thrust 
 as can be easily told by trying to push the armature back and 
 forward in its bearings while the machine is running the belt 
 pull should be aligned. Lateral play of perhaps an inch is 
 allowed on almost all dynamos, so that with ordinary care 
 there need be no trouble from this source. 
 
 If the heating is simply a case of too great belt tension, 
 which can generally be told by inspection, the only thing 
 to be done is to ease the belt ; the trouble is unlikely to occur 
 at all if the bearings are looked after and kept in proper con- 
 dition. The greater the area of contact on the pulley can be 
 made, either by using a larger pulley on the machine or a 
 smaller one on the driving shaft, or 'by using wider pulleys 
 and belting, the more easily the same power can be trans- 
 mitted, with less pressure. 
 
 It is not a good plan to try to cool dynamo bearings with 
 water, for water is not a pleasant thing to have in the vicinity 
 of an armature ; better cut out the machine until it can be 
 fixed, or, if it be absolutely necessary to run, the very careful 
 application of ice may sometimes relieve the difficulty. 
 
 Occasionally a dynamo will positively refuse to do its work ; 
 and perhaps the most frequent causes are a flying short circuit 
 in the armature, such as has just been mentioned, or a break 
 or bad connection in the field coils. If the former is the 
 case, separate excitation by one of the other machines will 
 soon locate the trouble ; if the latter, measuring the resistance 
 of the field coils will generally disclose the difficulty. Some- 
 times in setting up the machine an accidental bad connection 
 may pass unnoticed until an attempt is made to run. If a 
 dynamo has been assembled recently, there may be a bad 
 magnetic joint between the yoke and the magnet cores, 
 when the machine will fail to excite properly. This will be 
 found by inspection, and it is remedied without difficulty by 
 doing the work over again more carefully. 
 
 Setting up or taking down a dynamo is a rather difficult 
 matter, and in a permanent station it is a very good idea to 
 have a simple traveling crane for the purpose of doing such 
 work as may be required. Where this does not exist, tempo-
 
 THE STATION. 2OI 
 
 rary supports can be erected and a differential tackle will do 
 the work. If the traveling crane is at hand, the removal of 
 an armature becomes a simple matter, for its weight can 
 be taken by the tackle and it can then be slipped out of posi- 
 tion by a single movement of the' crane ; the yoke or other 
 attachments which may be in the way being previously re- 
 moved. Such a course is only necessary in very large arma- 
 tures, for small ones can be easily handled by a gang of 
 men, the shaft always being so blocked up as not to bring 
 the armature down on the pole pieces. In shifting an arma- 
 ture with the crane or temporary tackle the ropes and chains 
 should never be allowed to touch the windings; the grip 
 should be entirely on the shaft. 
 
 MINOR APPARATUS. 
 
 As regards the subsidiary apparatus required to run a sta- 
 tion, every switch, cut-out and lightning arrester should be 
 kept in thoroughly good working order, with no bad contacts, 
 and no dirt allowed to accumulate. Lightning arresters 
 should be set with the plates or points across which the 
 lightning is expected to jump very close together, from a 
 sixteenth to an eighth of an inch, and should be kept free 
 from dust, which might otherwise cause a short circuit. Gen- 
 erally only those forms should be used in which after one 
 lightning discharge the arrester is automatically reset and 
 ready for another. 
 
 The measuring instruments of the station should be com- 
 pared now and then with each other and with standard 
 instruments, and the engineer or superintendent of the sta- 
 tion ought to take pains to understand the function of each 
 piece of subsidiary apparatus, of whatever kind, with which 
 he has to do. Specific information in these matters can best 
 be obtained from the company which furnishes the particular 
 forms of apparatus employed, as there are frequently minor, 
 though important, details peculiar to each especial make.
 
 CHAPTER VII. 
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 
 
 WHATEVER may be the advantages of electric traction, 
 whatever its convenience as a means of rapid transit, it is on 
 its efficiency that its ultimate importance must depend. We 
 must realize at the start that the electric motor is not a prime 
 mover, a fundamental source of energy ; it only furnishes a 
 very perfect and elegant means of utilizing electrical en- 
 ergy, already generated by some prime mover, at the point 
 where it may be most convenient to employ it, whether that 
 point be fixed, as in the case of stationary motors, or moving, 
 as in the case of street railways. Advantages may be and are 
 gained by employing motors, sufficient to offset considerable 
 losses in the necessary transmission and transformation of 
 electrical energy, but if these losses rise above a certain 
 amount the system must inevitably be a commercial failure. 
 
 Let us look, then, deliberately at the series of transmissions 
 and transformations necessary in electric traction, and form 
 as close estimates as possible of the losses, their magnitudes, 
 and the most practicable means for reducing them to a more 
 satisfactory figure. The first transformation of energy is 
 from the pressure of steam generated in the boilers to the 
 rotary motion produced by the engine and employed in driv- 
 ing the dynamos. 
 
 Then the mechanical energy obtained is first transferred, 
 through the medium of shafting or belting, to the dynamo, 
 where it is again transformed and appears as electrical cur- 
 rent on the line. In this convenient shape it is transferred, 
 with little loss, to the point on the line where the motor or 
 motors may happen to be. There it undergoes another trans- 
 formation in the motor back to mechanical energy, which 
 is then transferred, through the medium of gearing of one 
 sort or another, from the armature shaft to the car wheels.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 2O3 
 
 Fortunately, the losses at several points in this somewhat 
 complicated system of transmutations are comparatively 
 small ; and for practical purposes we may consider the losses 
 to be substantially as follows : first, the losses in the engine 
 and attachments ; second, those in the dynamo ; third, those 
 on the line; fourth, those in the motor, fifth, those in the 
 gearing. Luckily, not all these are serious. In the art of 
 electric traction as to-day practiced, their relative magnitudes 
 are about as follows : the most formidable are the first and 
 the last; they are of about the same magnitude, varying 
 enough in different cases to render it quite impossible to say 
 offhand which is the larger. Then come the losses in the 
 dynamo and motor, generally smaller than either of the 
 former, and that in the motor being somewhat the larger. 
 Finally, the loss on the line, which in many cases is the least 
 of all. Reduction of gearing has in some cases made that 
 source of loss relatively small, but too often at the expense 
 of motor efficiency. 
 
 Let us take up these several causes of inefficiency in, 
 order : 
 
 ENGINE EFFICIENCY. 
 
 Of the total indicated horse-power developed by ordinary 
 engines running at full load, nearly 10 per cent, is consumed 
 in friction of the moving parts. Occasionally the loss may 
 be less than 5 per cent. ; not infrequently it rises to 1 5 
 per cent. Its amount is nearly a constant, not a constant per 
 cent. It may increase or diminish with the load, but gener- 
 ally remains somewhere nearly uniform. 
 
 It is probable that every engine has a particular load for 
 which the friction is a minimum, owing to the fact that the 
 strains to which the machine is subjected produce slight 
 flexures that vary the resistance at the bearing surfaces ; for 
 the purpose in hand a safe estimate for the friction is 10 per 
 cent, of the rated capacity of the machine, and this, by the 
 convention usually adopted among engine builders, is its 
 output for boiler pressure of from 80 to 100 pounds and a 
 cut-off of about one-fourth stroke. A few concrete cases 
 may be of interest to the reader.
 
 204 THE ELECTRIC RAILWAY. 
 
 TYPE OF ENGINE. SPEED. I. H. P. FRICTION IN H. P. 
 
 Buckeye 280 23 5 
 
 Westinghouse 3 4 
 
 Armington & Sims 290 80 9 
 
 Corliss 88 120 10 
 
 Compound condensing 347 44 
 
 These are merely examples to show the general magnitude 
 of the friction. The differences in friction between two 
 engines of the same make and the same engine at differ- 
 ent times are as great or greater than any difference that is 
 likely to exist between two similar types of engine of like 
 size. Relatively, friction is less in large engines. 
 
 In railway power stations it is the almost invariable rule, 
 at present, that the driving engines are used not at their full 
 output, but at something considerably below it. The exceed- 
 ingly irregular demands for power on small roads make the 
 average load throughout the day decidedly less than the full 
 output of the engine. The result of this state of things is 
 that the friction, not a serious loss when working at full load, 
 becomes an important cause of inefficiency. Take, for ex- 
 ample, an engine rated at 100 horse-power, driving a dynamo 
 of similar capacity: the loss by friction in the engine will be 
 about 10 horse-power, irrespective of load; and used as such 
 engines generally would be on roads of comparatively small 
 size, the average load seldom would be more than one-half 
 the figure above named. Consequently, 20 per cent, of the 
 indicated horse-power would be wasted in worse than useless 
 friction. If the average load should fall to 30 horse-power, 
 a condition by no means infrequent on small roads, about 30 
 per cent, of the indicated horse-power is lost, and so on. 
 
 In two cases that have fallen under the writers' observation 
 the efficiencies of the engines, owing to underloading, were 
 reduced to 70 and 68 per cent., respectively; while even at 
 the greatest output noted during the day's run the efficiencies 
 were only 85 and 84 per cent. On another road, under very 
 different .conditions, the efficiency of the engine at mean load 
 was 88 per cent. ; at maximum load for the day, 92 per cent. 
 
 The differences between these three cases are exceedingly 
 instructive. In the first-mentioned one the engine was a 
 125-horse-power Corliss, driving two 30,000- watt dynamos 

 
 THE EFFICIENCY OF ELECTRIC TRACTION. 20$ 
 
 through the medium of a counter-shaft. The mean load was 
 about 40 horse-power and the maximum load 80 horse-power 
 six cars being in regular service on the line. 
 
 In the second case two Armington & Sims engines were 
 employed, belted directly to two 6o,ooo-watt generators run- 
 ning in parallel. Seven cars were used on the line. The 
 average indicated horse-power was 51.4 and the maximum 
 indicated horse-power 108. In the third case the engine in 
 question was a Ball machine of 1 50 rated horse-power, driving 
 two 5o,ooo-watt dynamos. Fourteen cars were in regular use. 
 The mean indicated horse-power was 86.4 and the maximum 
 horse-power 124. The first two cases are examples of how 
 not to design a railroad power plant. The third is a specimen 
 of fairly good engineering practice, although the dynamos 
 might have been somewhat larger to good advantage. 
 
 The more nearly the average load on the engine approxi- 
 mates to its full rated capacity, the greater the efficiency that 
 can be secured. A properly constructed steam engine can be 
 worked for short periods considerably above its nominal horse- 
 power without injury ; and it is more desirable to do this 
 than to allow its average load to become low. 
 
 For example, the engine of the third road mentioned was 
 worked occasionally as high as 215 horse-power for a few 
 moments, to supply a sudden and unusual demand for power 
 on the line. It was quite capable of doing this without dam- 
 age, and it was only by this means that the unusual efficiency 
 mentioned could have been secured. To work an engine, 
 then, most efficiently, so far as loss by friction is concerned, 
 it is necessary to keep it loaded as nearly up to its rated 
 capacity as possible. On small roads, where the maximum 
 electrical output may be three or four times the average, this 
 can only be done by occasionally working the engine up to 
 its extreme capacity in other words, by employing an en- 
 gine with a wide range of cut-off, so that the steam may 
 follow almost up to full stroke if necessary. In addition 
 to losses by friction in the engine there are also losses in 
 shafting, if it is employed. 
 
 It almost goes without saying that the best engine efficiency 
 will be obtained by belting or coupling the dynamos directly
 
 20 6 THE ELECTRIC RAILWAY. 
 
 to the engines; and for this purpose, unless the machines are 
 very large, a high-speed engine is preferable, since it not 
 only regulates better under extreme variations of load, but 
 usually allows a wide range of cut-off. 
 
 And this brings us to another very important consideration 
 in efficiency. We have already seen that underloading an 
 engine produces a needlessly large loss in friction and re- 
 duces the efficiency of the system; but it also does worse 
 
 than this it lessens the efficiency of the engine directly and 
 
 considerably. If the steam be permitted to expand to many 
 times its original volume in the cylinder, there will ensue 
 an amount of condensation that very materially increases the 
 amount of steam necessary to produce a given effort. In 
 other words, if the steam is cut off too early in the stroke it 
 is employed at a great disadvantage, and the amount of coal 
 that must be burned under the boilers to furnish the required 
 number of indicated horse-power in the engine will be greatly 
 increased. On the other hand, if the cut-off be too late, as 
 would happen if an engine were regularly overloaded, the 
 steam passes into the exhaust before it can be employed to 
 the best advantage ; and there is a corresponding increase in 
 the amount of fuel required. 
 
 For any given pressure of steam there is a particular ratio 
 of expansion that will use the steam in the most economical 
 way, and consequently will require the minimum amount of 
 coal per horse-power hour. A considerable amount of inves- 
 tigation has been spent in finding this point ; and the best 
 formula that has yet been produced is probably the following, 
 due to Emery : 
 
 i i 
 
 r ~ I -\-p_ 
 
 22 
 
 Where r is the ratio of expansion, that is, the ratio of the 
 volume of steam admitted to its final volume, and / is 
 the absolute pressure in pounds per square inch, that is, the 
 pressure of the steam plus the atmospheric pressure. 
 
 For the boiler pressures generally employed, 90 to 100 
 pounds to the square inch, the most economical point of cut- 
 off is between one-fourth and one-fifth of the stroke. This
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 2O/ 
 
 is about the point generally taken by engine builders in 
 designating the nominal horse-power of their engines, as has 
 been previously explained. There is, then, a double reason 
 for working an engine up to its full rated horse-powder. 
 First, it reduces the friction to as insignificant a factor as 
 possible ; and, second, it employs the steam to the best advan- 
 tage. As a mattet of fact, overloading is less injurious in 
 the latter respect than underloading; so that we have an 
 additional reason for not employing engines larger than are 
 necessary to do the w r ork required. 
 
 Compound or triple-expansion engines, in which the total 
 expansion of steam is carried through several cylinders 
 instead of being confined to one, do not suffer as severely 
 from variations in the expansion ratio as simple engines ; so 
 that they are not only more economical per se, but are better 
 adapted for w r ork in which the variations of load are likely to 
 be large. It is not our purpose to go again here into the de- 
 tails of the question of engine economy, but it may be stated 
 that compound engines require for a given output about two- 
 thirds as much coal as simple engines, and triple-expansion 
 condensing engines only about two-fifths as much- the ma- 
 chine in each case being worked at its full normal load. 
 
 For maximum engine efficiency and economy, the facts 
 would, then, indicate the use of the compound or triple- 
 expansion condensing engine, loaded as nearly as possible to 
 its full rated capacity under ordinary circumstances, and 
 allowed now and then to work up to nearly its extreme power, 
 in case of .an emergency. If simple engines are used, the 
 same requirements as to load follow. Engineers are often 
 called upon to decide between the use of a high-speed engine 
 with slide or piston valves and the low-speed engine with 
 Corliss or equivalent gear. For most kinds of service the 
 latter has the advantage in economy; in street-railway work, 
 however, the conditions except in quite large stations are 
 reversed; since, in the first place, the Corliss engine has a 
 smaller range of cut-off than the high-speed types, and con- 
 sequently cannot be regularly worked at so near its full 
 power, if the variations in load are to be large. Secondly, 
 w^hen worked considerably below its capacity it suffers more
 
 208 
 
 THE ELECTRIC RAILWAY 
 
 severely from cylinder condensation, on account of the large 
 exposed surfaces. In cases where the maximum load does 
 not exceed twice the average load the Corliss valve gear will 
 
 FIG. 93.-AVERAGE LOAD CARD. CORLISS ENGINE. 
 
 give excellent results."- We cannot emphasize this question 
 of underloaded engines more forcibly than by showing the 
 annexed indicator cards, taken from the first two engines 
 referred to at the beginning of this chapter. 
 
 Figs. 93 and 94 are respectively the diagrams for mean and 
 
 FIG. 94. MAXIMUM LOAD CARD. CORLISS ENGINE. 
 
 maximum loads of the Corliss engine in the power station at 
 Lafayette, Ind.f Figs. 95 and 96 are the same for one of the 
 
 * We speak of Corliss engines in this treatise as embodying the important principle of 
 independent admission and exhaust valves. All that is said of them applies with equal 
 force to some other engines possessing this same characteristic. Some of these have a 
 "positive" valve gear (i.e., with all the valves opened and closed by the engine itself) 
 and can be run at a higher speed than the regular Corliss type. 
 
 t 'fhc Electrical World, June 22, i88g.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 209 
 
 engines in the Syracuse, N. Y., power station, tested by Mr. 
 Hulett.* A glance will show the advantages that would exist 
 in a case where the mean load is at an economical cut-off 
 instead of a somewhat inefficient one, as at Syracuse, or the 
 even worse state of things at Lafayette. 
 
 Nevertheless, the card given by the Corliss engine is 
 noticeably better in configuration than the other; and were 
 
 FIG. 95. AVERAGE LOAD CARD. HIGH-SPEED ENGINE. 
 
 it not for the exceedingly short cut-off and consequent heavy 
 cylinder condensation, the superiority of the valve gearing 
 w r ould have asserted itself in a diminished amount of fuel. 
 As the case stood, however, the Syracuse engine gave the 
 indicated horse-power hour on 5.3 pounds of coal, as against 
 7.3 in the other. 
 
 To sum up, then, the golden rule for efficiency in power 
 stations is to use the engine at its most economical point of 
 cut-off when delivering its average output; and this will be 
 in nearly every case at the full nominal capacity of the en- 
 
 FIG. 96. MAXIMUM LOAD CARD. HIGH-SPEED ENGINE. 
 
 gine. Employ compound or triple-expansion condensing 
 engines wherever the cost of fuel renders it desirable. If 
 simple engines are used, employ heavily built high-speed 
 engines belted directly to the generators, unless the conditions 
 are such that the variations in load are not much more than 
 50 per cent, of the average, when an engine of the Corliss or 
 similar type equipped with a heavy fly-wheel will econo- 
 
 * The Electrical World, August 16, 1890.
 
 2IO THE ELECTRIC RAILWAY. 
 
 mize fuel: With a thoroughly well-designed power station, 
 the efficiency of the engine, so far as frictional losses are 
 concerned, maybe raised to between 85 and 95 per cent., 
 with direct belting to the dynamos or good counter-shafting; 
 while under less favorable circumstances it is likely to fail as 
 low as 70 per cent.* 
 
 DYNAMO EFFICIENCY. 
 
 Much that has been said regarding the evil effects of 
 underloading engines applies also to dynamos, but there are 
 important differences between the two cases. There is noth- 
 ing in the dynamo corresponding to the inefficient utilization 
 of fuel that we find in an underloaded engine, and whereas 
 the losses by friction in an engine remain nearly constant, 
 quite irrespective of load, in the dynamo there are certain 
 losses that decrease very rapidly as the load diminishes. 
 Taking the dynamo as we find it, it may be worth while to 
 glance for a moment at the various causes that tend to dimin- 
 ish its efficiency. 
 
 There is, of course, a certain amount of friction in the 
 bearings, and besides this there are electrical and magnetic 
 losses due to the reversals of magnetization in the armature 
 core (hysteresis), losses by eddy currents in the metallic parts 
 of the machine, .energy spent in keeping up the magnet- 
 ism of the field magnets, and, finally, a very considerable 
 loss due to heating the armature. 
 
 The true frictional loss is generally rather small and re- 
 mains virtually constant. In constant-potential machines, 
 
 * It may be well here to note the practical efficiency of the steam engine as a mechanism 
 for transforming the heat energy of coal into mechanical power. As has been explained 
 in Chapter II.. the possible efficiency is limited by the practical range of temperature 
 attainable, but ivithin this limitation the best engines are tolerably efficient. With 
 compound or triple-expansion condensing engines the actual efficiency may be from one- 
 half to two-thirds of that theoretically possible between the given temperature limits. 
 With most engines three or four tenths of the possible efficiency would be an estimate 
 more nearly in accordance with fact. To take a concrete case, the 150 H. P. pumping 
 engine at Pawtucket, R. I. cards from which are shown in Figs. 23 and 24, Chapter II., 
 during a careful test gave the following results Total heat received by the engine per 
 hour, 2,427,198 units; indicated work, 143.49 H P. that is, 368,769 heat units; actual 
 efficiency, 15 per cent. Upper limit of temperature, 356 F.; lower limit, 100 F.; 
 
 ~ ? = .31. Therefore the engine in question gave very nearly half the possible 
 
 efficiency. The coal consumed was 1.54 pounds per indicated H. P. hour, the steam used 
 13.64 pounds.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 211 
 
 losses from hysteresis and eddy currents and the loss in the 
 field magnets are also practically constant ; but the energy 
 spent in heating the armature increases very rapidly as the 
 current increases. It is, in fact, equal to the resistance of 
 the armature multiplied by the square of the current. Thus, 
 with an armature resistance of one-tenth of an ohm and a cur- 
 rent of fifty amperes flowing, the loss in the armature will be 
 250 watts; while with 100 amperes it will rise to 1,000 
 watts. Consequently, when the load on the dynamo is di- 
 minished, its efficiency does not fall anywhere nearly so 
 rapidly as the efficiency of the driving engine falls, because, 
 while the losses in the latter remain constant, those in the 
 armature are much diminished by the lessened heating of 
 the armature. 
 
 With machines having a comparatively high armature 
 resistance the total efficiency may fall off but slightly as the 
 current diminishes. We frequently hear the electrical effi- 
 ciency of a dynamo mentioned, but the phrase is apt to mislead ; 
 for in practice we do not deal with the electrical, but with 
 the commercial, efficiency, and this includes not only losses 
 in the armature and field coils, but also all the others to which 
 we have alluded. 
 
 With the best forms of constant-potential dynamo now 
 constructed, the commercial efficiency, that is, the ratio of the 
 power supplied at the pulley to the electrical energy devel- 
 oped on the line, may rise slightly above 90 per cent., but 
 more often falls two or three per cent, below that figure. 
 
 Thus a dynamo giving 60,000 watts (about 80 horse-power) 
 at full load will require nearly 90 horse-power to be sup- 
 plied at the pulley and nearly 100 indicated horse-power at 
 the engine. As we have just seen, on lighter loads the 
 commercial efficiency of the dynamo will diminish, though 
 not very rapidly, until very light loads are reached ; in other 
 words, until the loss in the armature becomes quite small in 
 comparison with the losses from friction, hysteresis, and eddy 
 currents. Fig. 97 shows the efficiency curves of several types 
 of machine at various loads. The horizontal scale shows the 
 proportionate load on the machines, the vertical scale the 
 efficiency obtained at those loads. The curve numbered I is
 
 212 
 
 THE ELECTRIC RAILWAY. 
 
 for an English incandescent machine with magnets of the 
 inverted horseshoe type, intended for an output of 45,000 
 watts. Curve II is from a Wenstrom incandescent dynamo 
 of about the same size, and Curve III is from a Thomson- 
 Houston railway generator of 60,000 watts capacity. It will 
 be seen that from half load to full load the efficiency varies 
 but little, while below half load it falls off quite rapidly. It 
 is very apparent, then, that to secure the best efficiency in a 
 
 PER CENT OF FULL LOAD 
 FIG. 97. CURVES OF DYNAMO EFFICIENCY AT VARYING LOAD. 
 
 railway power station it is necessary to operate the dynamos 
 at somewhere nearly their full output, in which case the 
 capacity of the machine should nearly correspond to the 
 maximum continued output required. 
 
 If the variations in load are likely to be great it may be- 
 come necessary to work the dynamos above their intended 
 capacity, for a few minutes at a time, as occasion demands ; 
 but while- this overloading is quite unnecessary on large
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 213 
 
 roads, it is not to be recommended even on small ones, 
 unless in cases where the overload is not likely to continue 
 for more than a few moments. A safe rule is to provide a 
 dynamo of a rated capacity equal to about half the aggregate 
 rated capacity of the motors. This will be ample, unless in 
 the case of very small roads with heavy grades. The average 
 applied power required is not far from 6 to 8 horse-power 
 for each 1 6-foot car, supposing the latter to be equipped, as 
 usual, with two 15 -horse-power motors; it is unlikely to rise 
 above 10 horse-power. This subject, however, has been 
 already fully treated in the chapter on station design. 
 
 With a dynamo equipment thus designed the average out- 
 put would be generally a little over one-half the full capacity 
 of the machines; and their efficiency should consequently be 
 high, something like 85 or 90 per cent. Under very favor- 
 able circumstances, when the number of cars in operation is 
 so large that the maximum current required at any time upon 
 the line does not greatly exceed the average current, this 
 efficiency may be somewhat increased; though it cannot be 
 expected to reach, or to rise above, 90 per cent., except in 
 rare instances. Taking, now, these figures with regard to 
 dynamo efficiency in connection with what we have previously 
 stated about engines, we may be able to form a fairly accurate 
 estimate of the probable station efficiency that is, the ratio 
 between the indicated horse-power at the engine and the 
 electrical horse-power available upon the line. 
 
 There is certainly room for considerable improvement in 
 the usual practice, particularly in the earlier stations. The 
 power station of the Lafayette, Ind., street railway, tested 
 by one of the authors, gave a station efficiency of but 40 per 
 cent., owing to the fact that the engine was worked at less 
 than a third its full output and the dynamos at less than a 
 fifth. The power station at Syracuse, N. Y., tested by Mr. 
 Hulett,* is an example of decidedly better practice. Its en- 
 gines are belted directly to the generators, and were run on 
 an average at about a third of their full output ; the dynamos, 
 however, of twice the capacity of those used at Lafayette, 
 were under considerably more than twice the average load ; 
 
 * Tlie Electrical World, August 16, 1890.
 
 214 
 
 THE ELECTRIC RAILWAY. 
 
 .and consequently, since at these small outputs the dynamo 
 efficiency varies rapidly with the load, the Syracuse dynamos 
 must have given altogether better commercial efficiency. In 
 addition, by avoiding the use of a counter-shaft, the power 
 wasted in friction was relatively less than in the case first 
 mentioned ; so that the total average efficiency of the station 
 rose to 62.8 per cent. 
 
 Another road, instanced by Mr. G. W. Mansfield,* gave 
 
 PER CENT OF FULL LOAD 
 
 FIG. 98. -CURVES OF PLANT EFFICIENCY AT VARYING LOAD. 
 
 an average station efficiency of 54.6 per cent. The number 
 complete power-station tests published is comparatively 
 small, and the number of accurate ones smaller still. The 
 three above mentioned bear internal evidence of approximate 
 accuracy, which is more than can be said of some others that 
 have been reported. 
 
 With oui^presen^ dynamos and engines, a station efficiency 
 
 + The Electrical World. August , 7 , 1889.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 21$ 
 
 of 75 per cent, is about the best that can be hoped for on 
 ordinary roads ; and this can only be obtained by the greatest 
 care in designing the station. This statement is amply illus- 
 trated by Fig. 98, which shows the commercial efficiency of 
 the combined engine and dynamo plant in three cases. 
 
 The first curve, i , represents the efficiency, at various 
 loads, of a Goolden incandescent dynamo coupled direct to 
 a Willans engine ; the dynamo alone had a commercial effi- 
 ciency at full load of 91.2 per cent., and the engine showed 
 about the amount of friction usual in machines of its size. 
 It will be seen that the highest commercial efficiency reached 
 was 8 1 per cent., while at half load it fell to 72.2 per cent. 
 Curve 2 shows the commercial efficiency of an Armington & 
 Sims engine coupled directly to an Edison dynamo ; the half- 
 load efficiency is seen to be about 70 per cent., and the maxi- 
 mum load efficiency was 71.4 per cent. Curve 3 shows the 
 efficiency of the Lafayette, Ind., power station a Corliss 
 engine driving three Edison dynamos from a counter-shaft, 
 two of them railway dynamos of 30,000 watts capacity each, 
 and the other a smaller power dynamo. The maximum 
 commercial efficiency of the combination was 65 per cent., 
 and the half-load efficiency less than 60 per cent. 
 
 This third case very plainly shows the effect of the counter- 
 shaft, and the use of two railway dynamos instead of one of 
 double the capacity. 
 
 The obvious moral is to belt or couple the dynamos directly 
 to the engine, whenever possible. Cases often arise where it 
 becomes desirable to drive more than two dynamos from a 
 single large engine, but the advantage of convenience thus 
 secured is obtained at the expense of a certain small amount 
 of efficiency.* 
 
 Passing now from the consideration of the power station 
 where the electrical energy is generated, let us take up the 
 problem of its efficient transmission to the point where it is 
 needed. 
 
 * Recent developments of the steam turbine may affect engineering practice consider- 
 ably. A late test of a loo-kilowatt combined turbine and dynamo gave at full load the 
 electrical horse-power hour for 27.6 pounds of steam, and at half load for 29 pounds This 
 is a better result than has ever been obtained from any except compound engines, and 
 renders possible the economical use of direct-coupled dynamo and engine plants of a 
 much smaller size and lower first cost than have been generally contemplated.
 
 2i6 THE ELECTRIC RAILWAY. 
 
 The losses encountered in transmitting electrical energy 
 over a system of conductors are two in number. First in 
 -importance is the loss due to the resistance of the line. This 
 is equal, in watts, to the resistance multiplied by the square of 
 the current, and can readily be determined and allowed for. 
 Second, and of but very little importance under ordinary 
 circumstances, is the loss due to leakage usually, but not 
 always, imperceptible. In overhead systems of distribution 
 as now constructed the insulation resistance, by actual meas- 
 urement, can be readily brought to over a megohm per mile 
 of line. Therefore, even on very extensive systems, the 
 insulation will be amply high to prevent any appreciable loss 
 of power. It should be noted that doubling the length of a 
 line halves the insulation resistance, so that a line of the 
 insulation above mentioned and ten miles long would have 
 a total insulation resistance of 100,000 ohms. 
 
 EFFICIENCY OF THE LINE. 
 
 The loss due to line resistance may evidently be made as 
 small as may be desired by increasing the cross-section of 
 the conductors. Of course it is not economical to do this 
 beyond a certain point, for the cost of copper would then 
 more than offset the cost of the small extra amount of power 
 demanded by a line of higher resistance. At great distances 
 from the power station the line is comparatively inefficient; 
 near the power station there is almost no loss, and as lines 
 are constructed to-day the average line efficiency is generally 
 from 90 to 95 per cent. For example, on the Lafayette 
 street railway previously mentioned the average loss on the 
 line is about 7 per cent., on the Syracuse railway about 9 
 per cent., and in other cases even smaller. Where the sys- 
 tem is a compact one with plenty of feeders, 5 per cent, will 
 readily cover the losses in the line. If a wide district is 
 reached from a single power station, twice this figure is a 
 fair allowance. In certain cases even greater loss may prove 
 economical in the long run ; as by locating the power station 
 at a point where coal is cheap and water for condensation is 
 available, the operating expenses may be lessened in spite 
 of the slightly increased output. The details of this matter
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 21^ 
 
 are, however, not suitable for treatment here, as they have 
 been discussed at length in the chapter on the line. 
 
 Taking into account, then, the losses that occur in the 
 transmission of the electrical energy from the point of gen- 
 eration to the motors on the cars, and combining with it the 
 station efficiency, we find that in the most favorable case a 
 trifle over 70 per cent, of the indicated horse-power at the 
 engine may be applied to the motors. In the roads now 
 operated the proportion is much more likely to be between 
 55 and 60 per cent. 
 
 It is instructive .thus to summarize the losses as each suc- 
 cessive step in utilizing the power of the engines at the cars 
 is passed, for it helps one to recognize the real difficulties 
 that have been met in electric traction and the success that 
 has been obtained in spite of them. On the whole, the result 
 up to this stage of the operations is fairly satisfactory. The 
 figures thus far deduced apply with equal force to any case 
 of electric traction, except where storage batteries are em- 
 ployed a method which will be treated by itself in a follow- 
 ing chapter. 
 
 Where distribution is by means of underground conduits 
 instead of aerial lines, the conditions are generally the same 
 that we have been discussing. The losses by leakage are 
 likely to be considerably greater, but this can in part be com- 
 pensated by a slight increase in the cross-section of the con- 
 ductors. The difficulty thus far with conduit systems has 
 been not leakage properly so called, but a condition of things 
 that has frequently approximated a short circuit. With 
 improved conduit construction this may be remedied. Ex- 
 perience has shown thus far that the danger here is not so 
 much low general insulation as complete breaking down of 
 the insulation at one or many points. When it becomes pos- 
 sible to save conductors from actual short circuit, it will 
 probably also be possible to avert any disastrous amount of 
 leakage. Third-rail and similar systems of supply are often, 
 though to a less extent, open to the same objections that apply 
 to the conduit ; but if the insulation can be kept up at all, it 
 can probably be brought to a point where the losses in the line 
 will not greatly exceed those now found in overhead systems.
 
 2I g THE ELECTRIC RAILWAY. 
 
 The next subject for consideration is the efficiency of the 
 electric motor as a converter of electrical into mechanical 
 energy. 
 
 EFFICIENCY OF MOTORS. 
 
 So far as losses in the machine proper are concerned, the 
 railway motor in most of its forms gives a very satisfactory 
 efficiency. Other things being equal, a given dynamo elec- 
 tric machine is somewhat more efficient as a motor than when 
 used as a generator, provided the conditions are nearly the 
 same. One reason for this is that in the motor whatever 
 magnetizing power the armature coils possess is added to the 
 effective magnetization of the machine, while in the gener- 
 ator it is subtracted from it. For this reason motors are fre- 
 quently given an armature relatively more powerful than 
 would be adopted in a generator, particularly if it is desirable 
 to construct a motor of high output compared with the weight. 
 As an example of the effect of armature reaction in helping 
 the efficiency of a motor, we may give the following cases 
 derived from the experiments of Hopkinson. Two large 
 dynamos precisely similar in construction and dimensions 
 were used together, one as a generator driving the other as 
 a motor. The efficiency of the first machine was 87.1 per 
 cent, as a generator and 92 per cent, as a motor. The sec- 
 ond machine had an efficiency of 84 per cent, as a generator 
 and 89 per cent, as a motor. 
 
 Street-railway motors as now generally built are series- 
 wound machines ; but for the necessity of using a high e 1 actro- 
 motive force and reducing the weight of the machine to the 
 smallest practicable figures, they could be made exceedingly 
 efficient; as it is, there is little to complain of in this respect, 
 although the losses from the resistance of the field coils and 
 from hysteresis, in rather large armatures, are generally 
 more considerable than they would be in dynamos of a simi- 
 lar output. 
 
 The principal sources of loss in our present street-railway 
 motors are the regulating devices and the gearing. In start- 
 ing a soo-volt motor under heavy load, it is necessary in 
 some way to interpose resistance to prevent a flow of current
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 219 
 
 so great as to endanger the armature. It would evidently 
 be bad practice to have a normal field resistance high enough 
 to accomplish this end, for after the motor has gathered 
 headway this resistance becomes quite unnecessary and is 
 distinctly harmful. So, in the motor systems most generally 
 employed, special regulating coils are thrown into series with 
 the motor at the moment of starting, and then more or less 
 gradually cut out, so that by the time the armature has 
 reached its full speed the resistance of the field coils is little 
 more than that necessary to secure the requisite magnet- 
 izing force. In the old Sprague motors each limb of the field 
 magnet was wound with three separate coils, the aggregate 
 resistance of which was sufficient to serve the purpose men- 
 tioned. After starting, the coils at the first in series were 
 thrown into various combinations, and were all finally par- 
 allel with each other; so that the final resistance was thus re- 
 duced to a reasonable amount. 
 
 In the old type 7 ^-horse-power motors the initial resist- 
 ance, with the field coils all in series, was 20 ohms for each 
 motor, the final resistance 3^ ohms. The later type, of 15 
 horse-power, was wound with much heavier wire, with 
 greater precautions against heating; and the resistance 
 varied from 8.6 at the start to about 1.6 ohms when all the 
 coils were in parallel. 
 
 Later, since the Sprague system has passed into the con- 
 trol of the Edison Company, a starting coil of about 6 ohms 
 resistance has been placed in series with the motors, but is 
 immediately cut out after starting, and the combinations 
 then proceed as before, although the relative resistances 
 have been slightly changed. The Thomson-Houston motor 
 system, pursuing substantially the same course, employs a 
 rheostat, the resistance of which can be varied within quite 
 wide limits, and also two coils on each limb of the machine, 
 which are first used in series, but finally one of them is cut 
 out, leaving a single coil of low resistance in circuit. 
 
 The YVestinghouse device for regulation is like the Thom- 
 son-Houston, except that the rheostat is made in a few sec- 
 tions instead of a large number. 
 
 The uncertainty regarding the real efficiency of street-rail-
 
 220 THE ELECTRIC RAILWAY. 
 
 way motors as they are now used in practice comes from 
 the varying resistance of the regulating devices, and only a 
 rough estimate can therefore be formed of the average losses 
 that occur from heating. In most forms of motor the hyster- 
 esis losses are quite heavy, being, at the average speeds, not 
 far from 200 watts for each motor. Some careful dynamom- 
 eter tests made by one of the authors* upon Sprague motors 
 of the later type give an average commercial efficiency for a 
 single motor of 83.6 per cent. ; with two motors the hysteresis 
 losses are doubled, the power is generally not increased in the 
 same ratio, and the final result for the average efficiency of 
 the pair was 77.6 per cent. ; this does not include, however, 
 the losses in gearing. The figures for all the various sys- 
 tems are not easily attainable, but it is safe to say that 75 per 
 cent, would be a fair average, taking into account the fact 
 that some of the systems use one motor and some two, and 
 that the electrical efficiencies and various sources of loss vary 
 considerably, according to the type of motor used. 
 
 The figures thus given bring us at once face to face with 
 
 the question of employing a single large motor in the place 
 
 of the two generally used. That the efficiency would be 
 
 raised by such a substitution is well-nigh self-evident, and is 
 
 amply borne out by experience. The causes of this are, 
 
 \ first, the advantage in efficiency of a single large machine 
 
 \ over two small ones ; the electrical efficiency may be made 
 
 I somewhat higher, owing to the advantage of substituting 
 
 one magnetic circuit for two ; and the losses from hysteresis 
 
 / and friction are perceptibly lessened. 
 
 As the practical aspect of this problem is inextricably in- 
 volved with the efficiency of the gearing, the latter will be 
 discussed before proceeding to further consideration of the 
 question in hand. With regard to the possible average effi- 
 ciency of street car motors, it is quite within bounds to say 
 that with a properly designed single motor 80 per cent, is 
 quite attainable ; and even this efficiency may be somewhat 
 raised, though it is hardly likely to reach 90 per cent., on 
 account of the present necessity for regulating devices of 
 some kind or other and the need for a comparatively light 
 
 *The Electrical World, May 31, 1890.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 221 
 
 weight machine. In most street car systems, even to-day, 
 the armature speed is somewhere nearly i.ooo revolutions 
 per minute, at usual car speeds; and consequently there 
 must be a reduction, by gearing, to one-tenth or one-twelfth 
 the armature speed, between the armature and the car axle. 
 This has been until recently generally accomplished by a 
 double set of spur gears. The efficiency of such gearing is 
 usually given in books on engineering at over 90 per cent. 
 for each set of gears, and with reasonable precautions against 
 Cirt this figure may be reached readily enough. 
 
 Boxed-in gearing is obviously more advantageous in this 
 respect than open gears exposed to the weather. In the test 
 of Sprague cars just mentioned, the efficiency of the gearing 
 belonging to a single motor was found to be a trifle over 85 
 per cent. Even the best spur gearing, protected from dirt, 
 is not likely to give 90 per cent, for the double reduction 
 required. The joint efficiency of the gearing of both motors, 
 in the same series of experiments was found to be about 72 
 per cent. ; showing both that the two sets of gears do not run 
 in perfect harmony, and the effect of doubling the fixed losses. 
 
 It is very doubtful if a pair of motors can ever be made to 
 run perfectly together; for even if the two machines were 
 originally designed to be as nearly alike as practicable, there 
 will probably be slight differences between their respective 
 speeds, sufficient to throw them out of harmony with each 
 other and increase both the losses in the motor and those in 
 the gearing ; and this is more emphatically true if, as gener- 
 ally happens, the motors are now and then taken apart for 
 repairs and the magnetic circuit thereby impaired. Lack of 
 general cleanliness about the motors, and penetration of oil in- 
 to the joints of the magnetic circuit, have a distinctly deleteri- 
 ous effect, and of themselves are sufficient to produce injurious 
 differences between the two motors. It is not probable that a 
 pair of motors can ever be brought to work in harmony 
 for any great length of time; and there is, therefore, a con- 
 stant strain upon the gearing that makes itself evident ex- 
 perimentally, as has just been indicated. 
 
 During the last six months all the prominent manufact- 
 urers of electric street railway motors have brought out low-
 
 22 2 THE ELECTRIC RAILWAY. 
 
 speed machines with but a single speed reduction between 
 the armature and the axle. This result is reached either 
 by unusual arrangements of the magnetic circuit, or, more 
 often, by multipolar motors. The armatures of these ma- 
 chines have, generally speaking, a speed of between 350 and 
 450 revolutions per minute, at ordinary car speeds. The 
 motors themselves are of about the same weight as the older 
 double-reduction motors, have a gear efficiency much higher 
 
 reaching probably fully 90 per cent, in most cases and the 
 
 only uncertain factor is the electrical efficiency. The best in- 
 formation now obtainable indicates that it is hardly as high as 
 in the higher speed machines, although experience in design- 
 ing has taught makers to obtain a better average efficiency at 
 varying loads. These single-reduction gear motors unques- 
 tionably may gain enough in the abolition of one set of gear- 
 ing to more than compensate for the loss in electrical effi- 
 ciency resulting from low armature speed. This gain in total 
 commercial efficiency is probably trifling, seldom over 5 per 
 cent., but the reduction of noise and of wear and tear on the 
 gearing makes this type of motor a favorite.* 
 
 Methods of reducing the axle speed other than spur gearing 
 are not now generally employed. In this connection should 
 also be mentioned the use of connecting-rods driving di- 
 rectly both sets of wheels from a single very low-speed 
 motor, a case which has already been discussed in Chapter 
 III. Belts and sprocket chains are now and then used, but 
 have been for the most part abandoned by reason of their 
 uncertainty. Worm gearing is employed in a few cases 
 principally storage systems and has probably about the 
 same efficiency as the double-spur gearing which has already 
 been discussed. 
 
 * Some recent careful tests indicate that there has been in some cases too great a sac- 
 rifice of electrical efficiency in obtaining low- armature speed. In a trial of three types of 
 single reduction gear motors, the speed and power required for a round trip over a long and 
 nearly level track being accurately determined in each case, the co-efficient of traction 
 being assumed at 25 pounds per ton, an unusually high figure, the commercial efficiencies 
 were as follows: Motor I., 37 per cent.; motor II., 5 8 per cent.; motor III., 75 per cent. 
 Motor I. was undoubtedly underloaded, but the figures disclose no material gain, to say 
 the least, over the older types of motor.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 
 
 223 
 
 GENERAL EFFICIENCY WITH ONE AND WITH TWO MOTORS. 
 
 The experiments of Hale* indicate an even greater discrep- 
 ancy between the efficiency of a single motor and of a pair 
 than the experiments already mentioned. Fig. 99 shows the 
 losses from the various parts of an electric street railway 
 system as given by Hale. The diagram explains itself in 
 part; it is only necessary, in addition, to state that the differ- 
 
 MILES PER HOUR 
 
 ElK. HorW,A'. F. 
 
 FIG. 99. PROPERTIES OF MOTORS AND GEARING. 
 
 ence between the curves G and E represents the loss in 
 engine and dynamo friction ; that between E and D, the 
 electrical loss in the dynamo; that between D and C, the 
 loss in the line : that between C and B, the electrical loss in 
 the motors; and, finally, that between B and A, the loss in 
 friction of motors and gearing. The figures must be taken 
 
 The Electrical World, May 17, 1890.
 
 -4 
 
 THE ELECTRIC RAILWAY. 
 
 as merely approximate; but they are certainly instructive, 
 and are sufficiently near the truth to emphasize the facts 
 already mentioned. 
 
 From what has already been said, however, it is clear that 
 quite different results might be expected according to the 
 particular form of motors investigated. The total efficiency 
 of the combined motor and gearing, as found by one of the 
 authors, was for one motor 71.2, and for the pair 55.9 
 per cent. This is about a fair average. It is very certain, 
 at all events, that a single motor with its gearing has a com- 
 mercial efficiency more than 10 per cent, greater than in the 
 case where two motors are employed. 
 
 This amount is certainly worth saving, and there is no 
 reason why such a loss should be incurred. In the early 
 days of electric traction, when the conditions of work were 
 not sc thoroughly understood as now, two motors were sup- 
 posed to be necessary ; first, for security in case of accident 
 to one; second, for better driving where heavy grades were 
 to be mounted. As regards the former question we are of 
 opinion that with the latest apparatus these precautions are 
 unnecessary, as greater care in construction, and precautions 
 against injury to motors, have in a large measure obviated 
 the dangers that were at first feared; as regards the second 
 count, it is altogether probable that any grade which should 
 be attempted by electric street cars is practicable with a 
 single motor connected to both axles. It should be stated, 
 however, that there is one case in which the employment of 
 two motors instead of one possesses such considerable advan- 
 tages as to point distinctly to the advisability of its introduc- 
 tion into practice. It frequently happens that for suburban 
 service a decidedly high speed of the car is necessary for 
 example, between fifteen and twenty miles per hour. Such 
 speed is undesirable and would not be permitted in the 
 crowded part of the city through which the same cars may 
 have to pass. These conditions are beautifully met by em- 
 ploying two motors, and operating them in parallel for the 
 high-speed work and in series during the necessarily slower 
 progress through the city. The average efficiency of the 
 apparatus can be very perceptibly raised by this device, which
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 225 
 
 is already in use at several points. The general features of 
 this case have been discussed in Chapter III. 
 
 As regards gearing, it is doubtful if any device for trans- 
 mitting the power from the armature to the axle with a 
 reduction of speed can be made more efficient and more 
 thoroughly reliable than a single set of spur gearing running 
 in oil. Mention should here be made of the practice of 
 placing the armature directly upon the axle, so as to avoid 
 the use of gearing altogether. This obviates at once the 
 present losses in gearing, and thus saves a very perceptible 
 amount of power. Were the speeds employed on street 
 railway lines considerably higher than they are at present, it 
 would be a perfectly simple matter to put the armature upon 
 the axle; and for railway work and railway speeds this is 
 undoubtedly the best practice, and has been followed with 
 success in the City and South London Railway inaugurated 
 about a year since. 
 
 Three types of gearless motor for street car service have 
 been brought out during the past year, and have been em- 
 ployed on a rather extensive experimental scale. They are 
 described elsewhere, in Chapter III., and illustrate three rad- 
 ically different solutions of the low-speed problem. 
 
 The Westinghouse gearless motor has its armature core 
 keyed directly to the axle, without any attempt at cushioning 
 of any kind. . 
 
 The Short gearless motor is provided with a cushioned 
 support quite independent of the car axle, with which, how- 
 ever, the armature is concentric, but mounted on a hollow 
 shaft with about an inch clearance all around the axle. The 
 power is transmitted to the wheels by driving spiders heavily 
 cushioned with rubber. 
 
 The Eickemeyer motor is flexibly supported midway of 
 the truck, and drives both sets of wheels by outside connect- 
 ing-rods. 
 
 No tests on any of these machines have been published, 
 but the best data attainable indicate that the total commercial 
 efficiency obtained is no greater than with ordinary double 
 and single reduction gear machines in other words, every- 
 thing that is gained in abolishing the gearing is lost in the 
 15
 
 22 6 THE ELECTRIC RAILWAY. 
 
 purely electrical efficiency. The reduction of the armature- 
 speed to as low as from 100 to 1 50 revolutions per minute 
 means very serious difficulties in getting up a counter electro- 
 motive force to insure good efficiency and torque enough to 
 handle the cars readily upon grades. 
 
 In none of the forms of gearless motor mentioned is it 
 entirely evident on examination how these two difficulties 
 are met. All three run well in practice, and possess the 
 common advantage of very smooth and noiseless operation. 
 Eickemeyer employs one large motor instead of two small 
 ones, probably gaining thereby electrically, and losing me- 
 chanically through the connecting-rods. It is doubtful if 
 any of the forms mentioned can show as high an average 
 efficiency as the best single-reduction gear motors that are 
 now rapidly coming into use. On the contrary, it appears 
 that the efficiency of the gearless machines is less under the 
 present conditions than that of their competitors.* 
 
 They do, however, enjoy the advantage of very quiet run- 
 ning and freedom from repairs to any gearing and its connec- 
 tions, and are capable of doing very good service, provided 
 they run under conditions that insure a rather high average 
 speed, as for instance in suburban work. Where two gearless 
 motors are employed on a single car, their operation in series 
 is decidedly advisable, at ordinary car speeds; with this 
 modification, they have even now a considerable field for 
 usefulness. After the design of suclr low-speed machines 
 becomes more familiar through experience, it is probable 
 that their commercial efficiency can be raised nearly to the 
 value now possessed by the single-reduction gear motors, 
 but up to the time of writing the gearless motors have not 
 come into sufficiently extensive use to show properly by 
 experience their vices and virtues. 
 
 Serious difficulties to be met in this case are those involved 
 in regulating devices for the purpose of starting and varying 
 the speed. In the City and South London Railway a rheostat 
 is employed, and this would also be fairly successful on street 
 cars as, indeed, experiment has shown. 
 
 * A rather careful approximate test of one of the gearless motors indicated a commer- 
 c.al efficiency of about 40 per cent, at a car speed of between 8 and 9 miles per hour.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 22/ 
 
 Another solution of the same difficulty may be found in 
 the various plans that have been devised for connecting the 
 armature spindle flexibly with the axle, so that the motor 
 may run almost free at the moment of starting and the load 
 be assumed gradually. Several varieties of clutch, epicyclic 
 and hydraulic gearing, have been used for this purpose. Any 
 of them would dispense with the steady use of the rheostat, 
 and with the consequent loss of efficiency ; but at the expense, 
 perhaps, of serious mechanical difficulties. These cannot be 
 told a priori, and the use of such mechanisms has been up to 
 the present hardly more than experimental. Their advan- 
 tages cannot be doubted ; and were a thoroughly practical 
 apparatus devised, it would probably find its way into very 
 considerable use. 
 
 Summing up, then, the subject of motor and gearing effi- 
 ciency, we find that from 80 to 90 per cent, electrical efficiency 
 is attainable, though not generally attained in street railway 
 motors; that a little over 90 per cent, is quite within reach 
 in gearing efficiency, and that all losses due to gearing may 
 be dispensed with by the use of a motor placed directly upon 
 the axle. In the actual practice of to-day these efficiencies 
 together are generally reduced to less than 75 per cent., and 
 often to between 60 and 70 per cent. 
 
 The gains that may well be made are, first, a gain in motor 
 efficiency, especially in the low-speed machines; second, a 
 considerable gain from the use of a single motor ; and third, 
 another important gain in the use of only one set of gearing, 
 or its complete abolition. If a single motor geared to both 
 axles be employed, about 80 per cent, is the highest probable 
 average commercial efficiency, but if high-speed work over 
 a long line is to be attempted the maximum efficiency men- 
 tioned is within reach. The difficulties inherent in slow 
 speeds on street railway work, however, are so great that 
 only under extraordinary conditions can high efficiency be 
 obtained. Better tracks, allowing the free use of a single 
 motor, and better all-around construction in the driving 
 machinery, are the means by which improvements are to be 
 secured. 
 
 We are now in a position to sum up the total commercial
 
 22 g THE ELECTRIC RAILWAY. 
 
 efficiency of electric traction, that is, the ratio between the 
 indicated horse-power at the station and the power actually 
 delivered at the car axle. The two complete tests referred 
 to frequently in this chapter, at Lafayette, Ind., and Syra- 
 cuse, N. Y., give respectively 25 and 37 per cent, as this final 
 efficiency. We may, however, unite the several efficiencies 
 of station, line, and car machinery to obtain an approximate 
 figure. Sixty-five per cent, is probably a high average for 
 station efficiency in the roads now built. The line efficiency 
 may be taken at about 92 per cent. , giving a total efficiency 
 up to the motor of approximately 60 per cent. 
 
 With the motors and the gearing generally employed, 
 the average commercial efficiency of the combination is 
 probably not often in excess of 65 per cent., giving a total 
 commercial efficiency for the system, from engine to car 
 wheel, of 39 per cent. ; this, of course, is but an estimate. 
 But taking all the factors into consideration it is probable 
 that the average of the roads now in operation would fall 
 quite nearly to the point indicated. In very few cases would it 
 fall below 30 per cent. ; in still fewer would it rise much 
 above 40 per cent. 
 
 With regard to the efficiency that may be reached by more 
 careful station designing and better motors, we can give but 
 an approximation. From the figures that precede we may take 
 the practicable station efficiency at 75 per cent., the line at 
 95, the motors at possibly as high as 90, and the gearing, if 
 used, as averaging about 90 ; giving a possible average effi- 
 ciency of between 55 and 60 per cent, for the entire system. 
 Anything over 50 per cent, can be attained only by the 
 utmost care in design and construction , and it is very doubt- 
 ful if this point is passed by any line now operated. 
 
 It will be thus seen that there is plenty of room for im- 
 provement in commercial efficiency, and that while to-day 
 most electric railroads are decidedly inefficient, it is practi- 
 cable to improve them to a very considerable extent. Fifty 
 per cent, in total efficiency places the electric roads at least 
 on a par with most cable roads, as regards the question of 
 efficiency alone, with the additional great advantages of inde r 
 pendent motor units, variable speed, and the ability to back.
 
 THE EFFICIENCY OF ELECTRIC TRACTION. 229 
 
 Much improvement can readily be made, but it is in the 
 mechanical rather than the electrical parts of the system ; 
 the chief losses, as we have already seen, are in the station 
 and in the car machinery. The former may be obviated by 
 careful designing of the station equipment, the latter by 
 improvement in the mechanical means for transmitting the 
 power from the motor armature to the car axle. 
 
 No stronger argument can be adduced in support of elec- 
 tricity as a motive power than the fact that, taking the electric 
 railway as it is to-day, even though nearly two-thirds of the 
 power generated at the station is often lost on the way, 
 though the expenses for repair are sometimes heavy, and 
 the first cost of equipment great, it still stands quite unap- 
 proached as a method of cheap and effective rapid transit. 
 Every improvement that is made in the apparatus means an 
 additional advantage for electric traction. It should be noted, 
 too, that the very improvements in motors and gearing 
 which will raise the general efficiency also tend to diminish 
 the repair bills. 
 
 In closing this chapter it should properly be stated that 
 although increase of general efficiency is highly to be desired 
 and will result in a considerable saving, it must not be sup- 
 posed that the power station expenses, with which efficiency 
 has chiefly to do, are a very prominent factor in the total 
 operating expenses of an electric road. On the contrary, the 
 cost of fuel is in nearly every case less than 10 per cent, of 
 the total expenses of the road, so that the whole amount that 
 may be directly saved by improvements of various sorts tend- 
 ing to increase the efficiency probably does not exceed 5 per 
 cent, of the total operating expenses. This is, however, 
 enough to make the difference between losing money and pay- 
 ing a fair return on the investment.
 
 CHAPTER VIII. 
 
 STORAGE-BATTERY TRACTION. 
 
 A CHAPTER on the applicaton of the storage battery to elec- 
 trical traction must of sad necessity resemble that famous 
 treatise written by a returned sailor on the manners and 
 customs of the Fiji islanders: "Manners, none; customs, 
 beastly." 
 
 When in 1880 the voltaic accumulator invented by Plante 
 about twenty years previously was modified into a more 
 practical form and brought to public notice, it was received 
 with every symptom of unwonted joy ; for its advent seemed 
 to open a magnificent vista of electrical energy in portable 
 form, ready always and anywhere to supply power for all 
 sorts of industrial purposes. It was puffed by scientific 
 authorities, lauded by the press, and vigorously exploited by 
 the trade. 
 
 More than ten years have now passed, and the promised 
 land overflowing with volts and amperes is still just beyond 
 our reach ; not that the accumulator has been a failure ; on 
 the contrary, it has been a partial success ; for, under favor- 
 able conditions and for certain purposes, it has done, and 
 will continue to do, admirable work. 
 
 To define the storage battery is not altogether easy ; and 
 the distinctions between secondary and storage cells purely 
 fanciful, and introduced into the art principally for the pur- 
 pose of befogging the mind with legal quibbles have been 
 a fruitful source of misunderstanding. Speaking in a broad, 
 general way, we may define the storage battery to be a voltaic 
 couple of such materials that the current-producing chemical re- 
 actions may be more or Jess completely reversed clectrolytically. 
 It differs from the primary battery only in that the materials 
 consumed in the latter, and of necessity replaced to continue 
 the electrical action, may in the former be brought back 
 nearly to their initial state by electrolysis. 
 
 230
 
 STORAGE-BATTERY TRACTION. 231 
 
 As a matter of fact, certain primary batteries are almost 
 identical in their chemical properties with certain storage 
 batteries; although the usual form of accumulator is com- 
 posed of materials that cannot be economically employed in 
 forming a primary cell. The accumulator, therefore, is sub- 
 ject to many of the same general laws as the ordinary voltaic 
 cell ; it is coupled up in precisely the same way, and gives 
 current and electromotive force not widely different from 
 those of a primary cell of similar area of plates. 
 
 It has, however, the immense advantage that its component 
 materials after discharge can be brought back to nearly their 
 original state on passing a current through the cell ; and it is 
 this feature that has given the apparatus the name of storage 
 battery. The most familiar form of accumulator is that in 
 which the voltaic element is composed of two lead plates 
 coated with oxides of lead and plunged into dilute sulphuric 
 acid. Originally the necessary materials were formed on 
 the plates by electrolytic action; Faure, however, introduced 
 a so-called pasted cell, in which two plates of lead, perforated 
 or roughened so as to permit of more readily retaining the 
 active material, were coated artificially with oxides of lead 
 one plate with peroxide, the other with protoxide. These 
 oxides mixed with dilute sulphuric acid are plastered upon 
 the plates, and, on drying, the paste sets firmly ; and the plates 
 are then immersed in the electrolyte ready for charging, or, 
 rather, the process of forming, which name is given to the 
 preliminary charging and discharging that is necessary to 
 reduce the active material to the proper chemical condition 
 before it is put to real service. The chemical actions that 
 take place in the accumulator are very complex and vary 
 somewhat with circumstances. Speaking in a general way, 
 during the discharge the active material is largely reduced to 
 sulphate of lead ; and during a charge this is decomposed, 
 forming on the one plate peroxide of lead and on the other 
 reduced lead. This charging and discharging can be repeated 
 over and over; and experience has shown that of the elec- 
 trical energy spent in charging the cell, between 80 and 90 
 per cent, can, under favorable conditions, be regained as 
 electrical energy on the discharge. The electromotive force
 
 232 
 
 THE ELECTRIC RAILWAY. 
 
 of lead accumulators such as have just been mentioned is in 
 most cases when the cell is actively beginning its voltaic 
 action about 2 y volts, and during the action of the battery 
 fallsslowly at first, very rapidly after a time. On charg- 
 ing, the potential difference rises rather rapidly until it 
 
 FIG. 100. -TYPICAL STORAGE CELL. 
 
 reaches about two volts, and then more slowly until it reaches 
 2.3 or 2.4 volts; the change in voltage being similar to that 
 which appears during the discharge. 
 
 The ordinary storage battery shown in Fig. 100 consists of a 
 number of lead plates formed as just described, connected in 
 parallel into two sets, one of positive and the other of nega- 
 tive plates, and plunged in alternating order in a vessel of 
 dilute acid. The plates that contain the active material, that 
 is, oxides of lead, are usually reticulated as shown in Fig. 101, 
 so that they are full of small square or polygonal cells, in 
 which the paste that forms the active material may conven- 
 iently stick. When the cell is in working order, the process 
 of charging converts the salts of lead on the positive plate 
 into oxide of lead and reduces those on the negative plate, for 
 the most part, to a somewhat spongy mass of metallic lead. 
 On discharging, most of the material on both plates goes
 
 STORAGE-BATTERY TRACTION. 
 
 23S 
 
 back into lead sulphate. If it were possible to prepare 
 cheaply and in available form plates of peroxide of lead 
 and spongy lead, \ve should be possessed of a primary battery 
 in every respect similar to the secondary battery during the 
 process of discharging ; as a matter of fact, such plates would 
 be inconvenient to prepare mechanically; whereas it is very 
 handy to form them electrolytically, by passing a current 
 through the cell after it is discharged. 
 
 This, in the rough, is the principle of the secondary battery 
 most frequently in use to-day. A large number of other 
 combinations of materials possess in a greater or less degree 
 the same power of forming a voltaic cell the actions of which 
 during the process of charging may be reversed electrolytic- 
 ally ; and such combinations form accumulators of greater or 
 
 E 
 
 c 
 
 C 
 
 B 
 
 n 
 
 r 
 
 r 
 
 r 
 
 B 
 
 r 
 
 B 
 
 r ' 
 
 r 
 
 n 
 
 n 
 
 B 
 
 r. 
 
 n 
 
 a 
 
 c 
 
 B 
 
 r 
 
 r 
 
 r 
 
 B 
 
 p 
 
 p 
 
 r 
 
 p 
 
 p 
 
 JZ[ 
 
 6 
 
 r 
 
 r 
 
 c 
 
 r 
 
 r 
 
 r. 
 
 r 
 
 c 
 
 a 
 
 p 
 
 r 
 
 r 
 
 n 
 
 i 
 
 r. 
 
 c 
 
 r 
 
 a 
 
 B 
 
 r 
 
 r 
 
 r 
 
 r 
 
 c 
 
 c 
 
 c 
 
 p 
 
 p 
 
 n 
 
 r 
 
 c 
 
 r: 
 
 r 
 
 r 
 
 r 
 
 r 
 
 r 
 
 r 
 
 B 
 
 p 
 
 r 
 
 r 
 
 n 
 
 a 
 
 p 
 
 r. 
 
 r. 
 
 r 
 
 p 
 
 r 
 
 r. 
 
 r: 
 
 p 
 
 r 
 
 n 
 
 r 
 
 r 
 
 n 
 
 p 
 
 c 
 
 p 
 
 P 
 
 r 
 
 n 
 
 c 
 
 r 
 
 r 
 
 r 
 
 r 
 
 p 
 
 r 
 
 r: 
 
 n 
 
 P 
 
 r 
 
 P 
 
 B 
 
 a 
 
 n 
 
 E 
 
 r. 
 
 B 
 
 r 
 
 P 
 
 P 
 
 r 
 
 c 
 
 n 
 
 n 
 
 r 
 
 c 
 
 B 
 
 n 
 
 B 
 
 r 
 
 r 
 
 c 
 
 n 
 
 B 
 
 3 
 
 5 
 
 r 
 
 n 
 
 p 
 
 n 
 
 a 
 
 IT 
 
 nr 
 
 D 
 
 r^ 
 
 r~ 
 
 rr 
 
 c 
 
 n 
 
 nn 
 
 tr 
 
 d 
 
 n 
 
 FIG 101. -TYPICAL "GRID." 
 
 less usefulness, of which several will be mentioned further 
 along in this chapter. 
 
 Having thus obtained some idea of the general character 
 of the accumulator, we may appropriately pass to a special 
 consideration of its application to practical apparatus, partic- 
 ulary in electrical traction. Experience has unfortunately 
 shown that while these applications are in the abstract very- 
 simple and beautiful, in practice there are very serious inher- 
 ent defects in the accumulators of to-day so formidable, 
 indeed, that the ingenuity of inventors has been sorely taxed
 
 234 THE ELECTRIC RAILWAY. 
 
 to obtain an accumulator that shall be reasonably efficient, 
 moderately reliable, and tolerably durable. 
 
 The accumulator gives an opportunity of utilizing a dynamo 
 for storing up chemical energy at any convenient time or 
 place, and permits it to be transformed back again into elec- 
 trical energy whenever and wherever desired. The possi- 
 bility is a most tempting one, and immediately the Faure 
 cell was brought out in 1880 it was put to use for all sorts of 
 industrial purposes. When, as frequently happens, it is 
 desirable to obtain light at times when no power is available, 
 the accumulator can be charged whenever an engine and 
 dynamo chance to be running, and then employed to supply 
 electric lamps as they are needed. In installations for the 
 lighting of private houses, as well as in some central stations, 
 it is very inconvenient to keep the dynamos in action all 
 night ; and for the small amount of service required during 
 the late evening or after midnight, accumulators can be, and 
 are, worked with a considerable degree of success. Into this 
 field of work the accumulator was put at once, and has con- 
 tinued in gradually increasing use until the present. It is 
 also to a certain extent employed for running small motors, 
 electro-medical apparatus, and similar very light work, under 
 circumstances where the cell must be charged at one place 
 and discharged at another. Eight or ten years ago it was 
 frequently proposed to have a central station occupied ex- 
 clusively with charging accumulators to be distributed for 
 furnishing light and power to regular customers. It was 
 soon found, however, that the great weight of the batteries 
 precluded this otherwise very beautiful possibility. 
 
 Hardly had electrical traction been proposed, than it was 
 seen that with a durable and efficient accumulator most 
 ideal conditions could be realized; it would be possible to 
 charge the batteries at a point on the line where a power 
 station could be very economically maintained, put them 
 into the cars, and keep the road in steady operation, replacing 
 the batteries with freshly charged ones at every trip, or 
 whenever the original set should become nearly exhausted. 
 
 Experiments tending to this end were carried out as early 
 as 1880, and in 1883 a car was put into service at Kew Bridge,
 
 STORAGE-BATTERY TRACTION. 235 
 
 London ; this car was equipped with a Siemens dynamo set 
 to run as a motor, and under the seats were stored fifty large 
 storage ceils weighing in the aggregate about 4,000 pounds. 
 The car could carry over forty persons, and ran over its level 
 track very smoothly at the rate of about six miles an hour. 
 
 Encouraged by this success, a considerable number of ex- 
 perimenters went actively to work ; and during the next two 
 or three years cars were put into operation in London, Brus- 
 sels, Paris, and elsewhere. The motors were as a rule of 
 from 4 to 8 horse-power, and the weight of storage battery 
 varied, according to the power supplied, from 2,000 to 5,000 
 pounds. The energy stored was 'capable of running the car 
 from thirty-five to fifty miles without changing the batteries ; 
 at the end of this period fresh cells had to be substituted, 
 and as these were of considerable weight the operation was 
 not easy. In Brussels a number of accumulator cars were 
 put into regular service about five years ago. In this country 
 the work has been taken up by several different companies ; 
 and during the last three years experiments 6n a commercial 
 scale have been carried on in New York under the auspices 
 of the Julien. Company, which for a considerable time oper- 
 ated ten or twelve cars on the Fourth Avenue road ; in Phil- 
 adelphia, in New Orleans, in Washington, and elsewhere. 
 
 At the present time the accumulator cars in New York, 
 Philadelphia, New Orleans, and Dubuque have been aban- 
 doned, and not a single road is to-day commercially operated 
 in the United States by storage batteries. The nearest ap- 
 proach to it is an honest experiment in Washington, D. C., 
 the results of which are as yet uncertain.* There are besides 
 a number of experimental cars of all sorts and descriptions 
 not in regular service. It will thus be seen that in spite of 
 the work that has be"en done, the results have not been alto- 
 gether satisfactory. It is worth while looking in detail into 
 the matter for the purpose of seeing what the trouble has 
 
 * We make this statement advisedly, since not even one road, so far as reported, is 
 operated independently of the storage-battery companies. No safe commercial conclu- 
 sions can be drawn from cars that are continually under the supervision of skilled elec- 
 tricians employed by the installing and not by the operating company.
 
 THE ELECTRIC RAILWAY. 
 
 been, and if possible forming some idea as to the outlook for 
 better results in the future. 
 
 Much ingenuity has been expended in working out the de- 
 tails of storage-battery traction; almost every conceivable 
 form has been given to the lead plates intended to hold the 
 active material; several of these are shown in Figs. 102 and 
 103. The intent has been to- secure a light supporting struct- 
 ure capable of holding the plugs of active material very 
 firmly. * The cells used on cars are almost universally placed 
 in trays under the seats, and either slid out endwise or taken 
 out sidewise by lifting up doors at the sides of the car. One 
 
 \ 
 
 FIG. 102. SPECIMEN GRIDS. 
 
 very neat and successful arrangement for accomplishing this 
 has been in use in connection with the Julien cars on the 
 Fourth Avenue road in New York, and is shown in Fig. 104. 
 It consists, as will be seen, of something very like a dumb 
 waiter with several shelves placed on either side of a space 
 just wide enough to permit the car to run in. The car is 
 brought between the two series of shelves, the exhausted bat- 
 
 * It is unfortunate that lead, the only cheap metal capable of resisting the attacks of 
 sulphuric acid, is so undesirable from a mechanical point of view, and so heavy. If the 
 plugs of active material are in intimate contact with the grid they are almost certain by 
 change in volume to warp the weak lead support, while it is difficult to "take up" 
 expansion without encountering electrical difficulties.
 
 STORAGE-BATTERY TRACTION. 
 
 237 
 
 teries shoved out on the empty shelves, and the dumb waiters 
 lowered so that fresh trays of charged accumulators can be 
 
 i nn nHH ffiffi 
 
 ^/^ 
 
 " o 
 
 
 
 
 
 
 o 
 
 ? 
 
 
 
 
 o 
 
 
 ' 
 
 
 
 - 
 
 
 
 
 o 
 
 
 - 
 
 
 
 - 
 
 
 : 
 
 o 
 
 
 : 
 
 
 
 : 
 
 
 : 
 
 
 
 
 : 
 
 
 
 -, 
 
 
 - ; 
 
 ; 
 
 
 : 
 
 
 
 
 
 
 
 
 
 
 3 : 
 
 
 
 3 
 
 
 i O 
 
 o 
 
 
 
 
 
 Q 
 
 
 
 o 
 
 
 x^ 
 
 o o 
 
 3 / 
 
 / 
 
 
 _L 
 
 O O 
 
 o 
 
 :;.* : :::::: : .**.:j 
 
 . 
 
 o O + o 
 
 ::::::::::. :::<> 
 
 
 FIG. 103. SPECIMEN GRIDS. 
 
 thrust into place. The operation can be very expeditiously 
 performed by this means. Some very ingenious special reg- 
 ulating devices are employed, both in the station and on the
 
 238 
 
 THE ELECTRIC RAILWAY. 
 
 cars ; and in the latter case the power possessed in storage- 
 battery traction, of using more or less cells on the motor, as 
 required, enables the latter to be used in a quite economical 
 way rather more efficiently, doubtless, than the ordinary 
 machines where regulation is effected by a rheostat or by field 
 commutation. 
 
 The motors for storage-battery use are almost always de- 
 cidedly less powerful than those generally employed in the 
 direct systems of supply, for the very good reason that to 
 
 FIG. I04APPARATUS FOR REPLACING THE BATTERIES ON A CAR. 
 
 carry batteries enough to supply continuously an amount of 
 power equivalent to that used on the ordinary electric rail- 
 way is at present almost out of the question ; and, furthermore, 
 is unnecessary, because storage-battery cars are not well fitted 
 for work on severe grades, for reasons which will be ex- 
 plained presently. 
 
 There is little that is unusual in the appearance of electric 
 cars fitted with storage batteries ; the only thing to be re- 
 marked is the absence of the trolley or any visible collecting 
 device, and sometimes a slight widening of the lower body 
 of the car for the sake of receiving the batteries. Altogether
 
 STORAGE-BATTERY TRACTION. 239 
 
 the experiments with storage-battery traction have met with 
 only partial success ; for reasons not for the most part de- 
 pendent on any lack of care or skill on the part of those who 
 have engineered the various attempts, but on account of 
 defects inherent in the batteries not confined, as a rule, to 
 any particular type, but common to the entire class. As 
 long as an accumulator car is favored with expert care and 
 handled with great judgment and discretion it is likely to 
 operate well, and, so far as successful running is concerned, 
 to compare favorably with any other form of automobile car. 
 
 When these conditions are not fulfilled, the time comes 
 when the storage batteries exhibit deterioration, they are not 
 properly replaced, and failure is the uniform result. To 
 investigate the difficulties that have interfered with the de- 
 velopment of storage-battery traction we must begin with 
 the root of all the evils that befall them, that is, with the 
 accumulator itself. 
 
 Following the classification adopted by M. Reynier we may 
 divide accumulators into four genera, first, the lead and sul- 
 phuric acid genus, including the original Plante cell which, 
 by the way, is not to be despised, even to-day and the 
 Faure cell with all its modifications, including the great bulk 
 of accumulators that are in use. Second, the lead-copper 
 genus, consisting practically of plates coated with lead oxide 
 immersed together with copper negative electrodes in a solu- 
 tion of sulphate of copper; these are not used in anything 
 more than an experimental way, and are simply interesting 
 theoretically. Third, a very similar cell composed of lead 
 positive plates, zinc, and sulphate of zinc, which possesses 
 an electromotive force slightly higher than the ordinary lead 
 cell and has a high capacity in proportion to its weight, but 
 is open to the serious fault of losing part of its charge on 
 open circuit, and shares many of the inherent defects of the 
 ordinary Plante type. Fourth, the alkaline zincate genus r 
 the analogue of the Lalande-Chaperon battery, better known 
 in this country in its modified form of the so-called Edison- 
 Lalande cell. 
 
 In this case the positive plate is of porous copper, the 
 negative plate iron often gauze and the liquid sodium or
 
 240 
 
 THE ELECTRIC RAILWAY. 
 
 potassium zincate. This type of cell has some admirable 
 properties and has been applied to traction with tolerable 
 results, although not yet in use long enough to justify any- 
 thing like a final judgment as to its merits. The first genus 
 contains nearly all the species in practical use, their differ- 
 ences being mainly constructional. The reactions are about 
 the same in all, and the defects the same in kind, differing 
 only in degree. If it were possible to make the chemical 
 action in an accumulator perfectly reversible, we should be 
 possessed of an invaluable adjunct in all electrical operations, 
 but unfortunately this is not the case. The fundamental 
 difficulty is one not confined to the accumulator, but one 
 inherent in nearly all chemical reactions that are at all com- 
 plicated. It is generally true that all except the simplest 
 reactions are not quantitative in character; that is, more or 
 less material is converted into a form that cannot be utilized, 
 or by-products are produced in greater or less quantity. 
 
 The amount and character of the by-products is very largely 
 regulated by the speed or the temperature at which the reac- 
 tion is carried on ; it often happens that a slow reaction and 
 a rapid one, albeit of the same general character, lead to 
 different results. Theoretically, the chemistry of the storage 
 battery is that which has been already stated a reduction 
 during discharge of oxides of lead on the positive plate, with 
 the formation of lead sulphate on both plates. During the 
 charge, the .lead sulphate is decomposed electrolytically, 
 forming lead oxide on the positive plate and spongy lead on 
 the negative plate ; unfortunately this is not what is known 
 to chemists as a "clean" reaction. On the discharge, free 
 oxygen and hydrogen are formed in the solution, together 
 with ozone and hydrogen peroxide, that attack the plates 
 without materially assisting in the production of the cur- 
 rent. 
 
 There is local action of a useless and injurious character 
 between the support of the active material and the different 
 parts of the active material itself. Other and more compli- 
 cated* substances basic sulphates of lead, and the like are 
 also produced in the cell ; besides these there is the inevitable 
 heating owing to ohmic resistance, and losses due to over-
 
 STORAGE-BATTERY TRACTION. 
 
 2 4 I 
 
 charging; and among all these from. 10 to 15, or even as 
 high as 30 to 40, per cent, of the energy may be lost in the 
 process of charging and discharging, the amount depending 
 
 2357 1O HOURS 10 2O 25 
 
 FIG. 105. CURVES SHOWING VARIATION OF CAPACITY WITH DISCHARGE RATE. 
 
 largely on the rate at which the reactions go on the faster 
 the charge and discharge, the more by-products and the 
 lower efficiency. 
 16
 
 242 
 
 THE ELECTRIC RAILWAY. 
 
 In fact, the capacity of the battery, that is, the amount o* 
 electrical energy it is capable of giving out, is a function 
 of the rate of discharge, as well, of course, as of the rate of 
 charge. In some forms of storage cell this relation is almost 
 linear, so that if a given accumulator has a storage capacity of 
 60 ampere hours at a discharge rate of 10 amperes, it will have 
 a capacity of but 30 when 20 amperes are continuously drawn 
 from it. Fig. 105 shows a curve taken from a particular type 
 of cell that has been used extensively, showing admirably 
 the disastrous effect of attempting to force the discharge rate. 
 
 The losses incurred may be divided into four groups : first, 
 the direct losses due to heating ; second, losses due to local 
 action between the active material and the supporting grids ; 
 third, the losses due to local action in the active material ; 
 and, fourth, losses due to unreversed chemical action. These 
 various factors possess, in different types of cells, and in the 
 same cell under different conditions, very different relative 
 values. The last two sources of loss are intimately con- 
 nected with each other and are, on the whole, generally the 
 most formidable. Of the irreversible action a portion is due 
 to the formation of irreversible compounds, and another por- 
 tion to the electrolytic action producing free hydrogen and 
 oxygen, ozone, and hydrogen peroxide. The formation of 
 the latter substances does not tend permanently to deteriorate 
 the cell, while a portion at least of the former compounds 
 are of such a character as to damage the cell in a way that 
 cannot well be repaired. Thick grids, with plugs of active 
 material of corresponding thickness, are particularly likely 
 to suffer from the various difficulties mentioned with the 
 exception of the pure heating effect because the chemical 
 action in a large and dense mass of material is anything but 
 uniform throughout its volume, and potential differences are 
 undoubtedly set up between different portions of the same 
 plug. 
 
 Aside from these chemical defects of the lead storage bat- 
 tery there are others of a mechanical nature. Chief among 
 these may be mentioned the well-known and very serious 
 buckling, to which the positive plates in particular are very 
 subject; the cause is the expansion and contraction of the
 
 STORAGE-BATTERY TRACTION. 243 
 
 plugs during charge and discharge. This is to a considerable 
 extent unavoidable, although efforts have been made with 
 considerable success to eliminate its ill effects by so forming 
 the plug as to permit of expansion in its own plane, thereby 
 partially, at least, avoiding any tendency toward lateral dis- 
 tortion. The final result of buckling is a short circuit and 
 rapid destruction of the battery. Aside from this, some of 
 the lead sulphate formed during discharge is very frequently 
 lost, falling down into the bottom of the jar and thereby being 
 put out of useful action. At very rapid discharge rates, too, 
 there is a strong tendency toward disintegration of the plugs 
 of active material. Sometimes this action is very violent, 
 producing radiating cracks from the center of the plug, al- 
 most as if explosive force were at work from a point within. 
 
 Aside from all this, the lead accumulator is heavy; the 
 total weight per horse-power hour of energy stored is in most 
 types from 100 to 125 pounds. By employing thin plates of 
 active material upon unusually light supporting grids consid- 
 erably lighter cells have been formed, giving i horse-power 
 hour of output on approximately 50 to 75 pounds total weight, 
 but these light cells have by experience been found to be 
 too fragile to stand the wear and tear of continued use, and 
 have consequently been generally abandoned except for ex- 
 perimental purposes. From what has already been said it will 
 be easily understood that a very high discharge rate means 
 decreased capacity, and hence a greater necessary weight 
 of accumulator per horse-power hour stored. The figures 
 given are for the so-called normal discharge rates, running 
 usually, in the sizes of cells ordinarily employed, from 10 to 
 30 amperes. Of course the limiting factor in the weight of 
 battery necessary is the practicable discharge rate. So much 
 for the general properties of the lead accumulator. 
 
 The second and third genera of accumulators mentioned 
 previously have not come into commercial use, and there are 
 no signs of their development. The alkaline zincate cell, 
 however, has very recently come to the front, and has accom- 
 plished some excellent work. The starting point of this cell, 
 as previously mentioned, was the Lalande-Chaperon primary 
 battery, which was composed of an agglomerated plate of oxide
 
 244 THE ELECTRIC RAILWAY. 
 
 of copper, a zinc plate, and a solution of caustic potash. The 
 action in this cell is the gradual consumption of the zinc, 
 and the reduction of the copper oxide to metallic copper 
 through the agency of the oxygen set free. The oxide of 
 copper serves, then, as the depolarizing agent. Lalande and 
 Chaperon discovered the fact that this form of cell was 
 reversible, although they did very little to elaborate it. 
 
 In the hands of MM. Commelin, Desmazures and Bailhache 
 the Lalande-Chaperon battery was developed into a quite 
 efficient accumulator. The positive electrodes of this battery 
 are composed of porous copper plates formed by the com- 
 pression of finely divided electrolytic copper upon a nucleus 
 of copper gauze ; the negative electrodes are made of amal- 
 gamated, tinned, iron wire gauze; the positive electrodes are 
 surrounded by parchment-paper cells, the office of which is 
 supposed to be to prevent any cupric oxide which is slightly 
 soluble in caustic alkalies, but does not dialyze readily from 
 becoming mixed with the potassium zincate ; the resulting 
 accumulator is converted by charging into what is virtually a 
 Lalande-Chaperon primary cell. It possesses under discharge 
 the low electromotive force of only about .8 of a volt per 
 couple, but nevertheless has a high weight efficiency, 50 to 75 
 pounds being the total weight required for I horse-power 
 hour of output. 
 
 These batteries were employed with success in some ex- 
 periments on submarine torpedo boats carried out under the 
 auspices of the French Government, noticeably in the exceed- 
 ingly interesting trials of the torpedo boat La Gymnotc, in 
 the harbor of Toulon. The type, however, has not come into 
 general use as yet. 
 
 The efficiency appears to be just about the same as that 
 of lead accumulators; but practical trials like the one just 
 mentioned showed that on forcing the output, there was, 
 as there is in lead accumulators, considerable loss due to- 
 irreversible action, and a certain tendency to disintegration. 
 The chemistry of the cell, however, has not been sufficiently 
 studied to enable the exact character of these losses to be 
 definitely ascertained. The weight is about the same as that 
 of the lightest types of lead battery. The alkaline zincate
 
 STORAGE-BATTERY TRACTION. 
 
 24$ 
 
 cell, however, is of somewhat greater volume for equal out- 
 put, though doubtless of greater strength and somew r hat 
 greater duration of life than lead batteries of the same 
 weight efficiency. It has been desirable to go somewhat into 
 detail in regard to this matter, inasmuch as this type of cell 
 has, in the hands of three American inventors, Messrs. Wad- 
 dell, Entz,and Phillips, undergone some modifications, and has 
 been applied by them with fairly good results to electric trac- 
 tion. The form of electrodes they have employed is shown in 
 Fig. 1 06. The positive plates are composed of a folded porous 
 copper wire, having a solid nucleus and surrounded with a 
 woven envelope to serve a purpose similar to the function of 
 the parchment-paper cell in the French form of battery. 
 The negative plate is of iron, and the weight efficiency and 
 general efficiency of the battery seem 
 to be about the same as those of its 
 prototype ; but how far the difficulties 
 of unsatisfactory life of the plates, 
 irreversible action at high discharge 
 rates, running down on open circuit, 
 and short circuits from a growth of 
 "zinc tree, "have been avoided, only 
 time can show. 
 
 Taking up, now, the application of 
 the storage battery to the specific 
 problem of electric traction, one is at 
 once confronted by the very consider- 
 able weight of accumulators abso- 
 lutely necessary in the present state 
 of the art to secure continued ser- 
 vice. The experience of a very large number of electric- 
 roads extending over several years has shown that the work 
 required per car mile on ordinary 16 or 18 foot cars is a 
 little below i horse-power hour; at the car perhaps three- 
 fourths of a horse-power hour is generally sufficient ; this 
 for ordinary grades and the usual rates of speed of from eight 
 to ten miles per hour. We can therefore form a very good 
 idea of the amount of stored energy that must be carried for 
 ordinary service on an accumulator road. Assuming the 
 
 FIG. 106. POSITIVE PLATE OF 
 ALKALINE ACCUMULATOR.
 
 2 4 6 THE ELECTRIC RAILWAY. 
 
 .average car mileage per day at 100, which is very nearly the 
 mean of the roads now running, one sees immediately that 
 the weight of batteries carried for an all-day run would have 
 to be in the vicinity of /, 500 pounds. Ordinarily it has been 
 the practice to change the batteries two or three times a day, 
 but this smaller battery power carried means, as has already 
 been seen, on attempting to force the output for certain emer- 
 gencies, a considerably diminished storage capacity, so that 
 it has been found practically necessary to provide from 3,500 
 to 4,500 pounds of battery for the regular service of a car. 
 
 This dead weight is carried about continuously. The single 
 car that has been experimentally operated with alkaline 
 ^incate batteries has considerably reduced this figure, carry- 
 ing only about 2,500 pounds of battery, as might be antici- 
 pated from the greater weight efficiency ; but it has not at 
 the time of writing been in service long enough to permit a 
 fair judgment of the results. As a matter of fact, the effect 
 of forced output on batteries in such service is at times so 
 great that even the weights of battery mentioned are not 
 designed to furnish the power ordinarily used on overhead 
 electric roads; in other words, most of the storage-battery 
 cars that have been operated in this country and elsewhere 
 have secured reduced weight of battery by skimping on power. 
 
 A very brief computation founded on a catalogue of stor- 
 age batteries will show that to produce the 10 or 15 horse- 
 power necessary on even quite moderate grades, either a 
 weight of battery considerably in excess of that mentioned 
 muct be carried, or the discharge rate must reach a point 
 very considerably above that for which the rated capacity of 
 the cells is given. There is really no occasion for carrying 
 battery power enough for long runs, for batteries can be 
 changed just as horses are ; the difficulty is, however, that a 
 battery, even if charged every trip, must be capable of a cer- 
 tain maximum rate of output; and this at present entails the 
 use of heavy batteries, however often they are changed. 
 The weight of accumulators to be carried can only be mate- 
 rially decreased by securing a higher efficient discharge rate. 
 
 It is readily understood that, in electric traction, in starting 
 a car, and on curves and grades, there is always violent de-
 
 STORAGE-BATTERY TRACTION. 2/1/ 
 
 mand for current, lasting anywhere from a few seconds to 
 several minutes, and demanding from the battery, if one is 
 used, an especially heavy output. Every such call for a 
 large current tends to lessen the efficiency and life of the cell, 
 so that while in the laboratory the storage battery can be 
 worked to nearly 90 per cent, efficiency a figure which can 
 be approximated in electric-light service for traction pur- 
 poses this percentage must be much reduced. Naturally, 
 those most immediately interested are very chary about giv- 
 ing details as to commercial efficiency ; but from the best 
 sources of information obtainable, the results of experiments 
 on a considerable scale with storage-battery cars, one is jus- 
 tified in concluding that in every-day traction work the mean 
 efficiency of the cell is probably between 60 and 70 per cent. 
 in some cases even lower. Seventy per cent, is certainly 
 a favorable estimate for the battery; and this lowered effi- 
 ciency means also a lowered weight efficiency, hence more 
 frequent changes of battery or a greater amount carried. 
 
 If it were not for the great weight of battery the low 
 efficiency could be tolerated, but the result is a considerable 
 increase in the dead weight that must be carried about; and 
 dead weight in an automobile street car of any pattern is an 
 unmitigated objection. The two tons of battery placed in a 
 car require a very much stronger structure than would other- 
 wise be necessary; and although the motors employed in 
 storage-battery work are usually lighter than those used in 
 the systems of direct supply, for the reason that it is intended 
 to supply less power, the total weight of the car equipment 
 is brought up to a figure unpleasantly high. A standard 16- 
 foot car, with its two 15 -horse-power motors, weighs about 
 five tons; while the storage-battery cars of the ordinary pat- 
 tern such as have been used on the Fourth Avenue road in 
 New York weighed about seven tons, of which 3,600 to 3,800 
 pounds was battery. 
 
 The carrying capacities of the two, however, are practi- 
 cally equal; so that, per unit of useful work, the necessary 
 supply of power required in the direct and the storage sys- 
 tems of supply is in the ratio of five to seven. With a sim- 
 ilar motor equipment, that is, the same capacity for doing
 
 248 THE ELECTRIC RAILWAY. 
 
 work, the ratio would be about four to seven. That the two 
 tons of extra weight carried about by storage-battery cars 
 means a considerable disadvantage in the matter of power, 
 is a question that admits of but one answer. With the 
 ordinarv co-efficient of traction of about twenty pounds per 
 ton and a car speed of eight miles per hour, the additional 
 power continuously needed to drag around the storage battery 
 amounts to nearly i horse-power; this figure is. for a level 
 track ; if grades are to be attempted in addition, the additional 
 power required would be a little less than i horse-power for 
 each per cent, of grade. 
 
 As regards the actual commercial efficiency of traction by 
 storage batteries, it is not unfair to say that the systems of 
 direct supply have at least the advantage consequent on less 
 power needed for unit useful work. In the station, the stor- 
 age-battery road has somewhat the advantage; because the 
 dynamos are of the same efficiency in either case, while in 
 charging batteries both dynamos and engines can be worked 
 at an approximately regular load and near the point of maxi- 
 mum efficiency. Taking the friction of the engine at about 
 10 per cent, of its rated full load, which is an estimate in 
 accordance with the facts, the efficiency from steam to elec- 
 trical energy at the charging station may be reasonably taken 
 as about 75 per cent. 
 
 This is something like TO per cent, better than is likely to 
 be obtained on any but the very best overhead systems. 
 The efficiency of the motors will not vary widely in the 
 two cases, although the regulation, by changing the grouping 
 of the batteries which is readily accomplished in the storage 
 system gives it a slight advantage. Another small gain 
 may be secured by using the motors as dynamos to restore a 
 certain amount of energy to the batteries during the opera- 
 tion of stopping the car. To offset this we have on the one 
 hand loss in the line, on the other the inefficiency of the 
 battery; the former is somewhere in the vicinity of 10 per 
 cent. ; the efficiency of the battery, as just mentioned, is 
 hardly likely to rise above 70 per cent., and is more probably 
 below this figure. So that, taken altogether, we find that 
 the storage battery in station and motors gains perhaps 1 5
 
 STORAGE-BATTERY TRACTION. 349 
 
 per cent, in efficiency, and loses between 20 and 25 per cent, 
 in the comparison between losses in battery and losses in 
 line. As a net result, the total commercial efficiency from 
 indicated horse-power at the engine to brake horse-power at 
 the wheel is probably nearly, or quite, 10 per cent, less than 
 in the system of direct supply. A similar figure may be 
 reached ,by considering in detail the efficiency of the series 
 of transformations that occur in a storage-battery plant. 
 
 Taking the efficiency of the station at 75 per cent, and the 
 commercial efficiency of the motors at the same high figure, 
 and combining these with the efficiency of the batteries 
 estimated at 70 per cent. the total commercial efficiency of 
 the system appears to be about 40 per cent. ; while the various 
 tests made on the very best direct systems indicate a total 
 commercial efficiency of a trifle under 50 per cent. 
 
 Aside from this, we still have the greater power re- 
 quired on account of carrying around the batteries, which is 
 of the magnitude previously stated ; this latter disadvantage 
 could be for the most part obviated if it were possible to 
 obtain a light storage battery. In the storage car using 
 the alkaline zincate batteries the total weight of car and 
 batteries is just about the same as that of the standard motor 
 car of the ordinary sort, although the power supplied is not 
 equivalent in the two cases. 
 
 To find the relative commercial availability of the storage 
 system, various other factors in the expense must be con- 
 sidered. 
 
 Accumulator traction undoubtedly has the advantage of 
 requiring a smaller expenditure, and hence smaller interest 
 charges, at the central station. From the regular way in which 
 the dynamos and engines can be operated the capacity of the 
 charging station may be reduced to one-half, or at all events 
 to two-thirds, that required for operating the same number 
 of cars by direct supply. On the other hand, we have to 
 consider the relative cost of the equipment and maintenance 
 of the battery and of the overhead system necessary for dis- 
 tribution. In the first cost the ratio between them will 
 depend on the number of cars per mile operated by the over- 
 head wires, for the battery charges in the accumulator service
 
 2 r THE ELECTRIC RAILWAY. 
 
 vary almost directly with the number of cars. With direct 
 supply the cost of the distributing system increases a little 
 more rapidly than the mileage when the number of cars is 
 uniform. In a long run with few cars the advantage would 
 unquestionably be on the side of the storage battery; 
 on a line with a large number of cars, quite in the other 
 direction. 
 
 As regards cost of maintenance the subject is open to no 
 debate whatever. If an overhead system is properly put up 
 it depreciates but slightly ; 5 per cent, per annum would be 
 a large estimate, while 3 per cent, would cover it in many 
 cases. The depreciation of the batteries is an indeterminate 
 quantity, but is undoubtedly large. Figures on this point 
 are hard to find, for there is a very wide discrepancy between 
 the estimates of the makers on the depreciation of a given 
 battery and the results found when it is put in service for 
 traction purposes. There have been so few storage -battery 
 roads operated, and of these so few have given to the public 
 any definite report, that the question is a puzzling one. 
 
 When the storage system in Brussels was abandoned as more 
 costly than traction by horses, after two years' experience, 
 it was stated that the life of the positive plates was about two 
 hundred days, that of the negative plates considerably longer. 
 An extensive experience in this country with batteries used 
 in lighting railway trains a far less severe service than 
 traction showed that the positive plates were, as a rule, 
 destroyed within a year. In Brussels the cost of maintenance 
 of the batteries proved to be 2^ cents per car mile. 
 
 A very recent report from Birmingham, England, where 
 the storage battery has been given a pretty careful trial, 
 showed the cost of maintenance for a total car equipment to 
 be 4 cents per car mile. The result of our present experi- 
 ence with systems of direct supply seems to show that the 
 maintenance of motor and car apparatus amounts to between 
 i and 2 cents per car mile under favorable conditions, near 
 the former figure. Assuming this amount as a fair estimate 
 for the Birmingham line, we reach about 3 cents per car mile 
 as the cost of the maintenance of batteries in that case. One 
 of the American storage-battery companies recently estimated
 
 STORAGE-BATTERY TRACTION. 2$ I 
 
 the maintenance of the storage-battery equipment at $700 
 per year per car. This is probably low, as in Brussels the 
 contracting company's estimate for a similar type of battery 
 was i l /8 cents per car mile instead of the 2^ cents actually 
 found. Reports from the late Dubuque, la., road gave the 
 life of the batteries at much less than a year. All this would 
 indicate a maintenance account of dangerously near 100 per 
 cent, per annum certainly over 50 per cent. an amount 
 far in excess of the depreciation in overhead lines and motors 
 on the trolley system. 
 
 We are therefore driven to the conclusion that, considering 
 everything, the storage-battery method, at present, is de- 
 cidedly more expensive than the usual system of electric 
 traction, and that, unless the batteries are substantially im- 
 proved in duration of life, it is likely to remain so. Never- 
 theless, this conclusion does not prove that the storage battery 
 cannot now and then be applied under favorable conditions. 
 Two years ago the experience in Brussels showed that its 
 cost was greater than that of traction by horses, while the 
 recent report from Birmingham shows a small balance on 
 the other side of the account. 
 
 There is a very strong prejudice against the use of over- 
 head wires in large cities, and if electric traction is to be 
 introduced under such circumstances the choice lies between 
 conduits and storage batteries. The former will be taken up 
 in a later chapter ; the latter certainly has considerable advan- 
 tages in its favor. Each car is an independent unit, and 
 therefore no accident is likely to cripple the entire system, as 
 might happen with overhead lines. The service, with proper 
 care, can be made reasonably reliable and efficient ; for heavy 
 grade work it is at present out of the question, but over good 
 and level track and under favorable conditions as regards 
 traffic, storage-battery traction even to-day has a field for 
 usefulness. So far as we may be permitted to learn from 
 experience, its cost does not differ very widely from that of 
 horse-car service ; with everything favorable the advantage 
 may sometimes lie on the side of the storage battery ; under 
 other conditions it has often been, and may still continue 
 to be, in favor of horses. In every case, with our present
 
 252 THE ELECTRIC RAILWAY. 
 
 batteries, its cost is greater than electric traction by the over- 
 head system, doing the same amount of work. 
 
 In closing, it should be pointed out that while in any elec- 
 tric system a carefully laid and substantial track is necessary, 
 where storage batteries are employed it is doubly so ; for not 
 only do the irregularities of rough track and the consequent 
 jolting injure the accumulators, but the heavy cars pound 
 the track in a way to cause serious deterioration. The stor- 
 age batteries of to-day are unfitted for heavy grade work ; 
 their place is on a comparatively level track. The experi- 
 ments that are now going on, noticeably at Washington, will 
 undoubtedly teach a very useful lesson as to the present state 
 of the subject.* 
 
 Until definite reports from them are received one is not 
 justified in counting too much on the hopeful results of the 
 preliminary experiments. An approximate statement from 
 the Dubuque road gave 1.5 horse-power hour per car mile as 
 the output required at the station. This figure affords a 
 basis for a rough comparison with the trolley system. The 
 average station output in the latter case is quite nearly i 
 horse-power hour per car mile. Remembering that on ac- 
 count of the greater weight of an accumulator car about a 
 third more power is needed to propel it, the relative effic- 
 iencies of the two systems are shown by the above estimates 
 to be not far from the figure deduced elsewhere. 
 
 Finally, it is only fair to say that if the accumulator can be 
 made durable or of considerably higher weight efficiency than 
 is now usual, it will have a wide field of operations wherever 
 overhead wires are objectionable. But, as remarked by one 
 of the authors more than two years ago in discussing this 
 question, the time is not yet, and it is by no means sure that 
 any of the present types of accumulator will share in the ul- 
 timate success of the system. 
 
 * A report from the late Dubuque road, based on eight months' operation, gave the fol- 
 lowing results : Cost of power at station, including labor, coal, oil, waste, repairs, etc. 
 1.80 cents per car mile ; batteries, renewal expenses of every kind, 3.51 cents per car mile ; 
 ibor connected directly with the batteries, shifting, cleaning, etc., 1.78 cents per car 
 ; repairs to electrical machinery, 0.13 cent per car mile. Total motive power ex- 
 ises, 7.22 cents per car mile. These figures agree quite closely with those already men- 
 l. A thorough investigation of the two experimental roads in Washington, D. C., 
 by a competent engineer, has shown the motive power expenses to be not less than ten 
 cents per car mile.
 
 CHAPTER IX. 
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 
 
 TAKING into consideration -as is purposed in the present 
 chapter only electrical traction as it exists to-day, we must 
 recognize the fact that about 95 per cent, of it is accom- 
 plished on the overhead system of supply generally known 
 as the "trolley" system, and that the parallel method of dis- 
 tribution is well-nigh universally employed. Of the remaining 
 examples of electric roads a portion are operated by the third 
 rail, or ordinary track, system of distribution ; a portion by 
 storage batteries; a smaller portion by conduits of various 
 sorts; and the remainder is composed of miscellaneous tel- 
 pherage systems in a somewhat undeveloped state. The so- 
 called series method of distribution with constant currents 
 has been tried at various times and places, and has been, 
 with the exception of one or two small roads, abandoned. 
 Perhaps the most elaborate trial granted this plan was that 
 by the Short Electric Railway Company in Denver, Colorado, 
 and to some extent elsewhere. The result of this experi- 
 ence has been a complete change to the parallel distribution 
 system. 
 
 There are various good reasons for this. The system of 
 switches and cross-connections necessary in operating several 
 motors in series on an extended line has proved of so for- 
 bidding a character as to deter further pursuit of the method. 
 Another reason, and one which ought to have been recognized 
 from the very start, is that economy in conductors and effi- 
 ciency in the motor system compels the systematic use of the 
 highest voltage that can be employed with entire safety to 
 life. 
 
 This is true of whatever system of distribution is employed ; 
 and if a given amount of energy must be supplied for a fixed 
 line the series system fails in point of economy, if it be lim- 
 ited to a reasonable voltage. Our present electric-railway 
 
 253
 
 254 
 
 THE ELECTRIC RAILWAY. 
 
 lines are operated at a safe pressure, but one that probably 
 cannot be considerably increased without reaching the danger 
 limit. Hence, wherever bare wires are to be employed, the 
 series system, being limited to the same total voltage, must 
 transmit the same current if an equal amount of power is to 
 be employed, and is likely to lose both in efficiency of the 
 motors and regulating apparatus, and in the complications 
 before mentioned^ without gaining anything in the line. 
 
 As regards the three-rail method of distribution referred 
 to, it was employed with a fair degree of success in some of 
 the earlier roads, but proved inconvenient, owing to the 
 position of the bare conductor with reference to the earth. 
 If the third rail be supported above the level of the ground 
 on insulators it gets in the way of general traffic ; if placed 
 on a level with the tracks it still menaces, under ordinary 
 circumstances, vehicles and foot passengers, and becomes 
 the seat of insufferable leakage at any of the voltages now 
 generally used; hence it is generally inapplicable. The 
 same objections hold in case of using the ordinary rails as 
 the conducting system. 
 
 An exception must be made, however, in favor of lines 
 occupying a road-bed peculiarly their own, as is the case with 
 underground and elevated structures, together with such 
 surface roads as may in rare cases be entirely isolated from 
 ordinary traffic. 
 
 Under these circumstances the third rail becomes a very 
 available method of supply, and where the current to be dis- 
 tributed is very considerable, working conductors of the 
 requisite size are perhaps most easily maintained in this way. 
 A noteworthy example is to be found in the City and South 
 London Railway that pioneer of high-power and high-speed 
 electric traction. One or two elevated and underground 
 roads now under contemplation in this country will very 
 probably be operated in a similar manner. 
 
 The miscellaneous systems of supply we may conveniently 
 divide as follows: (a) open-conduit electric roads; (b] closed- 
 conduit electric roads ; (c) telpherage lines automatically con 
 trolled from one or more points, and either underground or 
 on trestles or aerial lines.
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 255 
 
 () The slotted conduit, somewhat similar to that used on 
 cable roads, has been over and over proposed and tried for 
 sheltering the conductors for the most part with very in- 
 different success. The principal exponents of the system 
 have been Bentley & Knight in this country and Siemens & 
 Halske in Germany. Two or three small conduit roads have 
 in addition been operated in England, notably at Blackpool 
 and Gravesend, the latter employing, strange to say, the 
 series system of distribution. 
 
 The Bentley-Knight roads in this country have now been 
 totally abandoned, after long and costly experimenting; and 
 there is to-day no conduit road in operation in America, un- 
 less we except an experimental one in Chicago. 
 
 Abroad, Siemens & Halske have met with considerable 
 success, particularly at Buda-Pesth; but how far this result is 
 attributable to the system employed and how far to the less 
 exacting conditions of the European climate, it is impossible 
 at present to say; although the comparatively slight differ- 
 ences between foreign roads and those abandoned here would 
 at least suggest that climate and accidents of location have 
 much to do with the matter. 
 
 The fundamental difficulty with all slotted-conduit electric 
 roads is the enormous difficulty of proper insulation. This 
 arises from the very nature of the case, for the conductors are 
 placed in a tube of limited diameter in free communication 
 with the open air through the slot. Water, dirt and mud 
 inevitably find their way in, and sooner or later the result 
 has been either a positive short circuit at a single point or 
 general leakage along the line in sufficient quantity to para- 
 lyze its operation. 
 
 A large number of more or less meritorious and ingenious 
 forms of conduit have been proposed, varying in the material 
 and arrangement of the various parts that compose the sub- 
 structure, in the position of the conductors therein, and the 
 methods of insulation employed. The features possessed by 
 all in common are the slot and a pair of bare working con- 
 ductors, placed in various positions \vith reference to the slot, 
 but comparatively close to each other. Conduits with a single 
 working conductor and rail return have been proposed, but
 
 256 THE ELECTRIC RAILWAY. 
 
 inasmuch as any connection between the working conductor 
 and the conduit produces a severe short circuit, they have 
 not passed beyond the experimental stage. Perhaps the 
 most elaborate trial given the conduit in this country was in 
 
 the city of Boston, under the 
 auspices of the Bentley- Knight 
 Company. The construction 
 adopted is shown in Fig. 107. 
 The road was nearly five miles 
 in length, and was in operation 
 less than one year, during which 
 time the continual and irrepress- 
 ible series of electrical misfort- 
 unes showed that climatic con- 
 ditions forbade commercial suc- 
 cess. 
 
 The Siemens & Halske road 
 in Buda-Pesth,Austro-Hungary, 
 is worthy of a little more ex- 
 tended description, as it is al- 
 most unique in its success. 
 
 The total length of the road 
 is now 12.4 miles, and 50 cars 
 are in operation. The first sec- 
 tion a trifle over a mile and a 
 half long was thrown open for 
 use in July, 1889, and the ex- 
 tension has therefore been very rapid. The line was built 
 at the expense of the firm of Siemens & Halske, and may be 
 regarded as the practical demonstration of their system on a 
 large scale. 
 
 The conduit is shown in Fig. 108 ; it is placed directly under 
 one of the rails, and in the main is formed of concrete ; cast- 
 iron yokes with flanges 7 inches wide are placed about every 
 4 feet, supporting the insulators and solidifying the entire 
 construction. The oval conduit is 1 1 inches wide at the 
 widest point and 13 inches deep; while the total depth of 
 the conduit foundations below the top of the rail is 27^ 
 inches. The slot consists of a pair of beam rails, without 
 
 FIG. 107 -BENTLEY-KXIGHT CONDUIT
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 257 
 
 any inside lower flange, secured to the conduit yokes by 
 wrought-iron angle pieces. The slot itself in the finished 
 construction is an inch and five-sixteenths wide, nearly twice 
 as wide as would be tolerated in this country. In each yoke 
 there is a socket on each side of the conduit at its widest 
 point, and in these sockets are carried the insulators to which 
 are fastened the working conductors. These latter are made 
 of angle irons, and are supported by the Y-shaped projec- 
 tions from the insulators. 
 
 As will be seen by a glance at Fig. 108, these are quite high 
 above the floor of the conduit nearly a foot and, further- 
 more, are sheltered by the upper portion of the oval so that 
 they cannot be readily touched from the outside, and are 
 protected from the access of rain. At convenient intervals 
 there are settling boxes, connected with the city sewers; and 
 whatever water enters the slot is collected at the lowest 
 points, and is thus carried away. The distribution of power 
 is at 300 volts, and the mains, instead of being carried on 
 
 FIG. 108. SIEMENS & HALSKE CONDUIT BUDA-PESTH. 
 
 poles, are of lead-covered cables, protected with iron bands 
 and laid in the earth along the line. At convenient points 
 there are junction boxes, from which feeders follow the 
 course of the road and are connected to the working con- 
 ductors in the conduits. 
 
 The road has grades up to about i . 5 and i .6 per cent., and a 
 17
 
 25 8 THE ELECTRIC RAILWAY. 
 
 considerable number of curves none of them, however,, 
 shorter than 84 feet radius. One curve of 148 feet radius is 
 on the steepest grade. 
 
 The speeds permitted are from 9, or even 12, miles per 
 hour in the less densely populated portions of the city to 
 barely 4 miles an hour over street crossings. The daily car 
 mileage per car during the 16 hours of service is from 75 to 
 90. The road has certainly proved a commercial success, 
 and one of the engineers of the company is authority for the 
 statement that the cost of running, exclusive of taxes, is only 
 37 per cent, of the income. This Buda-Pesth tramway is 
 almost the only conduit road in existence to-day of which it 
 can be said that its regular operation has been a success. 
 
 As will be seen from the description just given, the conduit 
 itself possesses no very extraordinary features ; it is simply 
 well made, substantial, and carefully drained. This latter 
 fact is probably the most important factor in the very good re- 
 sults obtained. The concrete conduit is of itself a fair insula- 
 tor, and serves to afford additional security against grounds. 
 
 A half-mile of conduit road is now (Dec. 1892) in opera- 
 tion in Chicago. The conduit is of about the dimensions of 
 that at Buda-Pesth, but is of sheet-iron and the conductors 
 are suspended by rigid insulator blocks from above. The 
 general character of the structure does not insure immunity 
 from the difficulties that have caused the abandonment of 
 similar attempts in the past. 
 
 Granting that the difficulties attending insulation could be 
 successfully surmounted, the slotted-conduit system in any/ 
 of the usual forms is of doubtful commercial value. 
 
 In the first place it is not generally applicable, being con- 
 fined in its operations, by conditions not depending on its 
 special excellence, to large cities ; and, second, even in these, 
 to locations where drainage is at least reasonably good. The 
 former objection is true also of the cable roads, which have 
 proved brilliantly successful commercially. But difficulty of ^ 
 drainage, while disastrous to electric roads, is only a slight 
 practical inconvenience to the cable. 
 
 The most serious objection to the underground electric 
 conduit of the open type is its cost, the necessities of drainage
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 259 
 
 compelling the employment of a tube of considerable dimen- 
 sions. The cost of the substructure is high, $20,000 a mile 
 being a moderate estimate. Doubtless many inventors of 
 such systems may claim an expense vastly less than this, but 
 if they guarantee that limit, they or their backers must be 
 supplied with very well-filled pocket-books. 
 
 This high initial cost puts the conduit road at a very serious 
 disadvantage, compared with overhead roads, and perhaps 
 even with storage-battery systems; for the latter, as has 
 been already shown, can be installed at an initial cost little 
 -exceeding, in many places, that of overhead systems; and 
 were the batteries to be improved in even a moderate degree, 
 the interest charges and maintenance of the conduit and its 
 conductors would be relatively so large as to give the storage 
 battery the advantage. 
 
 In this connection, however, should be mentioned inci- 
 dentally a type of slotted-conduit road that has recently been 
 proposed, which obviates many of the difficulties of insula- 
 tion, but unfortunately substitutes for them others, which 
 will be discussed a little later on in connection with closed 
 conduits. The type of road alluded to is one in which the 
 working conductors are insulated, except for a series of pro- 
 jecting studs furnishing current to an arrow carried under 
 the car, by the passage of which they are thrown into con- 
 nection with the working conductor through a series of 
 switches. The type is an interesting one, although at pres- 
 ent it is purely experimental, and on a very small scale at 
 that. 
 
 (b] Of closed-conduit systems there are many, elaborated 
 by Pollak, Lineff, Gordon and others, abroad, and Wheless 
 and others in America. The violent opposition to the erec- 
 tion of overhead wires in foreign countries has compelled 
 greater activity in this direction there than in America. 
 These systems have very much in common, the fundamental 
 idea of all of them being to put conductors permanently 
 underground, and to connect them automatically by the 
 passage of the car to short sections of working conductors 
 on, or immediately under, the surface and underneath the 
 car itself. By this means the amount of bare conductor
 
 260 
 
 THE ELECTRIC RAILWAY. 
 
 steadily subject to leakage or short circuits is very much 
 reduced. The switching-in of the several sections is accom- 
 plished generally by electro-magnetic means, through current 
 supplied from the car. 
 
 These systems, in first cost, are intermediate between 
 overhead lines and the regular slotted-conduit type. Like 
 the latter, they may be very seriously interfered with by cli- 
 matic conditions, and are therefore somewhat limited in 
 their application ; there is also an additional difficulty, which 
 will be appreciated by every working electrician, due to the 
 multiplicity of switches intended to be automatic in their 
 action. 
 
 The real seriousness of this difficulty cannot be definitely 
 told until some such road has been in operation for a consid- 
 erable time. It will be readily understood, however, that it 
 is by no means negligible, and would probably be the source 
 of a considerable amount of trouble. 
 
 In addition to these closed conduits or modified track 
 methods of supply there exists a class of conduits either 
 
 FIG. 109. VAN DEPOELE CONDUIT WITH FLEXIBLE LIPS. 
 
 closed, but with flexible walls, or partially closed, having 
 only flexible lips to shut the slot, except when the arrow 
 carried by the car opens it. 
 
 A good example of the latter construction, due to Mr. 
 C. J. Van Depoele, is illustrated here (Fig. 109). The 
 former arrangement is one wherein the leads are inclosed 
 in a casing having flexible walls. The upper portion of this 
 casing is taken up by the bare working conductor, and as 
 the car passes along, carrying a trolley wheel on either side, 
 the pressure exerted on the working conductor is sufficient 
 to force it into contact with the main. 
 
 These variations on the conduit theme would, if mechan- 
 ical difficulties could be overcome, be very valuable adjuncts
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 26l 
 
 to the methods of distributing current now available. They 
 are, however, of somewhat dubious practicability; for the 
 conditions to be fulfilled by a flexible insulating wall or 
 insulating lips are so severe as to be almost forbidding. Of 
 course, any of these arrangements can employ either a con- 
 tinuous conductor or one divided into sections automatically 
 switched into circuit, as in the Lineff and other systems 
 previously mentioned. 
 
 The whole group of devices that have just been described 
 are interesting both mechanically and electrically, but have 
 none of them passed the experimental stage; and while it is 
 perhaps not too much to expect that something practical and 
 valuable will be evolved from the amount of ingenuity that 
 has been spent upon them, the time is not yet. This line 
 of investigation, however, is of more than usual promise. 
 
 The number of such modified conduits is too great for 
 anything like elaborate description. Figs. 109-113, how- 
 ever, give a fair idea of the Van Depoele, Wheless, Harding, 
 Lineff, and Pollak-Binswanger plans, which are fairly typical 
 of the directions in which a successful solution of the great dis- 
 tribution difficulty has been sought. We can only suspend 
 judgment on their merits until they have been proved by 
 experience ; for while the difficulties are evident, the prac- 
 ticability of the means taken to avoid them can only be told 
 by protracted trials. 
 
 The use of alternating currents has now and then been 
 proposed for electrical traction, and it is worth while saying 
 something, at least, about their employment for such pur- 
 poses. 
 
 The alternating current has in general two very marked 
 advantages over the continuous current for all sorts of elec- 
 trical distribution and service. The first and most important 
 is the great ease of transformation from higher to lower volt- 
 age and vice versa. This is an immense convenience, and can 
 be made to effect a very great saving in the cost of distribu- 
 tion; since it is possible to distribute the electrical energy 
 in the form of current of enormous voltage and then to em- 
 ploy it in the various translating devices at as low a voltage 
 as may be desirable for the particular purpose in hand.
 
 262 
 
 THE ELECTRIC RAILWAY. 
 
 Inasmuch as it is difficult to insulate dynamos and motors 
 for very high voltages, while transformers can "be easily 
 insulated, it becomes possible by their use to accomplish 'the 
 
 FIG. no. WHELESS CLOSED- CONDUIT SYSTEM. 
 
 FIG. in. HARDING CLOSED-CONDUIT SYSTEM. 
 
 o 
 
 o 
 
 FIG. 113. POLLAK-BlNSWANGER CLOSED-CONDUIT SYSTEM. 
 
 distribution of the current at any desired voltage, and to 
 utilize it at a voltage practicable for the ordinary translating 
 devices.
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 263 
 
 The second great advantage of the alternating current 
 depends on its regulation in connection with inductive resist- 
 ances wasting only a trifling amount of energy. 
 
 If there were at present a thoroughly efficient and durable 
 alternating-current motor which could be run on the ordinary 
 two-wire system of distribution and would start readily under 
 heavy load, it would be possible to employ alternating cur- 
 rents for electric-railway work with the greatest advantage, 
 at least upon long lines ; for a very small amount of copper 
 would be required for the distributing conductors, which 
 would feed the working conductor at intervals by means of 
 transformers. More than this, the motors on the cars might 
 be conveniently and economically regulated by a coil of vari- 
 able self-induction, thereby doing away with one of the diffi- 
 culties which is at present considerable. 
 
 It must not be forgotten, however, that the cost of trans- 
 formers is not an inconsiderable item ; so that we should not 
 be able to reap the full advantage of the alternating current, 
 except in systems of distribution where the cost of the mains 
 aside from the working conductor is very considerable ; the 
 longer the line, the greater the advantage thus to be obtained. 
 
 Just at present, however, from the lack of a suitable motor, 
 little headway is being made toward alternating-current rail- 
 way systems. The " drehstrom" motor has been suggested 
 for this purpose ; but the complications introduced by the 
 multiple leads required are considerable, probably so great, 
 w r hen applied to general traction purposes, as to overbalance 
 the other good features of the machine. 
 
 It is not impossible, however, that some form of alternating- 
 current motor may soon appear that will lend itself readily 
 to traction work ; and in this case it will be likely to have 
 a considerable field in long-distance electric railroading. 
 
 Various other suggestions for the use of alternating cur- 
 rents have been made, some of them even going so far as to 
 propose a primary inducing circuit in a conduit and a secon- 
 dary coil carried on the car. The great difficulty of insula- 
 tion and the lack of an efficient inductive relation between 
 the two coils under such circumstances is enough to dis- 
 miss the scheme without further comment.
 
 264 THE ELECTRIC RAILWAY. 
 
 It will thus be seen that there exist at present, aside from 
 the regularly employed systems of electrical traction, a con- 
 siderable number of arrangements for the distribution of cur- 
 rent, which have not passed the experimental stage and which 
 are possessed of more or less merits and demerits, but no one 
 of them is yet sufficiently important to entitle it to anything 
 more than casual mention here. Some day one or more of 
 these may assume a position of commercial importance, but 
 until that time comes we can do nothing more than bring in 
 the old Scottish verdict of " not proven" against their claims 
 of great practical usefulness. 
 
 Passing, then, from these we reach (c] a group of devices 
 for electrical traction that involve comparatively few new- 
 electrical principles, but are noticeable on account of certain 
 useful properties of their own. We refer to the various sys- 
 tems of automatically controlled electric roads intended in 
 the main for freight and express service, and designed to 
 simplify the conditions of traffic by permitting automatic 
 control of an electrically driven train. 
 
 There are two classes of such devices that may well be 
 separated from each other. First, we may consider the very 
 ingenious systems of telpherage intended for the carrying of 
 freight at a low cost and in a very simple manner. Later 
 we shall have to give some brief mention to the plans for 
 high-speed service of a similar kind, intended to supplant 
 the pneumatic tube and similar apparatus rather than merely 
 to serve as a substitute for carting. 
 
 Telpherage owes its origin to the late Prof. Fleeming Jen- 
 kin, who both devised the system and coined its name. It 
 consists, in brief, of a wire-rope tramway carrying suspended 
 cars or trains of cars driven by electric motors, to which 
 current is supplied from the supporting rods or cables. Sim- 
 ilar apparatus, driven by wire rope, is now frequently used 
 principally for the purpose of transporting earth in making 
 excavations but Professor Jenkin conceived the idea of ex- 
 tending such a service, and enabling goods and material to 
 be transferred across country cheaply and rapidly. 
 
 The appearance of a telpher line resembles that of the 
 wire-rope devices just mentioned, consisting of a double line
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 265 
 
 of cable or rods suspended from stout posts and carrying- 
 trains of small cars. The general appearance of such a line is 
 shown in Fig. 1 14, which represents the system at Glynde, 
 England, which has been in operation for several years, car- 
 rying crude cement from the pits out of which it is taken to a 
 railway siding, whence it can be carried to the cement works. 
 The convenience of such an arrangement is obvious, al- 
 though its application may be somewhat limited. Its greatest 
 merit is the cheapness of the construction and the ease with 
 
 FIG. 114. -GLYNUE TELPHER LINE 
 
 which it can be operated where the laying down of track 
 would be inconvenient. 
 
 The special feature in telpherage systems of this class is 
 the method of supplying current to the car. This is, of course, 
 done through the supporting rods, but to get contact with 
 both positive and negative conductors requires no little inge- 
 nuity. 
 
 Of course, with a pair of cables supporting each car, or 
 with a cable and a trolley wire, the problem would be very 
 straightforward ; but in most cases for cheapness and me- 
 chanical simplicity it has been thought advisable to use 
 only a single supporting cable and no subsidiary conductor ; 
 hence a little electrical complication is necessary to enable
 
 2 66 THE ELECTRIC RAILWAY. 
 
 two cables to act both as positive and negative conductors to 
 trains running in both directions, one on each cable. Two 
 systems are employed for this purpose the one, series dis- 
 tribution at nearly constant current, such as has been pro- 
 posed for ordinary electric tram-car lines ; the other, a very 
 ingenious cross-over parallel system, such as is used in the 
 Glynde telpher line. 
 
 Its electrical peculiarities are fully shown in Fig. 115. The 
 two cables are divided into sections, electrically insulated 
 from each other as regards the consecutive sections of either 
 cable, but cross-connected, so that if any given section of 
 either cable be positive the next will be negative. By using 
 two trolleys for the supply of current to a train somewhat 
 longer than the section, there will be a steady flow of current 
 from one section through the motors to the next. By this 
 
 X X D 
 
 T,i 
 
 FIG. 115. ELECTRICAL CONNECTIONS OF GLYNDE TELPHER LINE. 
 
 device a single pair of cables can operate trains running in 
 both directions. A suspended electric locomotive drags sev- 
 eral cars, and is very readily controlled as it approaches the 
 ends of the lines. In the Glynde line ten small trough- 
 shaped cars are used in each train, and the speed reached is 
 four or five miles an hour. 
 
 The telpher system has never come into any use in this 
 country, but there are at present three lines operated in Eng- 
 land, the longest being a mile and a half. There are, of 
 course, a considerable number of ingenious minor devices for 
 the securing of constant speed and the proper control of trains, 
 and besides, an automatic block system is provided, so that 
 there will be no danger of rear-end collisions. 
 
 Satisfactory work can be done in this line of automatic 
 transportation of goods, and it would not be surprising to 
 see it come into occasional very efficient use in localities 
 where ordinary tramways are, from considerations of cost or
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 267 
 
 difficult country, impracticable. The supporting structure 
 can be made comparatively light ; the cables or rods need be 
 less than an inch in diameter, and a large amount of goods 
 can be carried at a comparatively small cost. Such lines 
 are perhaps at their best in serving as feeders to railway 
 lines, connecting them with manufactories, mines, quarries, 
 and the like. 
 
 In America, however, the system has not yet been intro- 
 duced practically, although small experimental roads have 
 been operated in years past by Mr. Daft and Mr. Van Depoele. 
 Nothing commercial has come from these experiments,, and 
 just at present there is no sign of further development. 
 
 The motors and electrical arrangements generally of tel- 
 pher lines present no specially striking features, the only 
 apparatus in any way peculiar being safety and controlling 
 devices, which may be arranged in a large number of ways, 
 according to the fancy of the individual inventor who has 
 the problem in hand. 
 
 Owing to the beautiful manner in which electrically oper- 
 ated vehicles can be controlled from any given point, the 
 idea of an automatic electric railway, with great capabilities 
 for speed and of a simple and cheap construction, has proved 
 peculiarly attractive to the minds of inventors. 
 
 The telpher system of which we have just spoken, is the 
 mere rudiment of automatic electric traction, and develop- 
 ment in the direction of higher speed and greater permanence 
 of service was very natural. It would seem at first sight to 
 be quite within the bounds of possibility to lay a narrow- 
 gauge track upon a comparatively cheap elevated structure, 
 and to run upon it trains suitable for the transportation of 
 mail and express matter; substitute rails and permanent 
 structures for the cables or rods of the telpher line, increase 
 the speed, and the problem would appear to be solved. 
 
 The possibility of running an electric train in some such 
 fashion at the rate of 100 or 150 miles an hour has been 
 recognized by many of those connected with the development 
 of electric traction. There are, however, a few electrical 
 difficulties, more mechanical ones, and commercial consider- 
 ations of a most forbidding character.
 
 2 68 THE ELECTRIC RAILWAY 
 
 The telpher line carrying its burden of ore or luggage of 
 various sorts over a comparatively short distance is relatively 
 practical; but where a train is to be run at a speed of 100 
 miles an hour over a distance of 100 miles, more or less, 
 peculiar conditions are encountered. In particular, such 
 high-speed service is only desirable for the transportation of 
 things comparatively small in bulk and comparatively great 
 in value, therefore the class of goods which the owners 
 would least care to trust to an automatic service, where acci- 
 dents would be especially serious in their results. 
 
 It is very doubtful, indeed, whether the United States Gov- 
 ernment would ever consent to intrust its mails between 
 New York and Boston, for e'xample, to any automatic system 
 whatever, on account of the great ease with which it could 
 be interfered with and the mails robbed, and the serious 
 delays that even a trifling accident could produce. 
 
 As soon as such a service ceases to be automatic it loses 
 much of its distinctive character, and might as well be ex- 
 tended into a general high-speed electric road intended for 
 transporting passengers as well as express matter. Such a 
 development is highly desirable and quite practicable, but 
 an electrical high-speed package express running unguarded 
 and unwatched over long distances is well subject to the 
 ancient criticism of being " neither fish, flesh, nor good red 
 herring." 
 
 Nevertheless, just such schemes have been brought forward 
 over and over again ; and one of them the Weems project 
 is described in Chapter X. in considerable detail, and has 
 proved of great value to the cause of electric traction by 
 demonstrating on a small scale, but none the less effectively, 
 the possibility of enormous speeds. 
 
 The logical line of development from the experiments is 
 the electric lightning express, as sketched in the same chapter. 
 Very high speeds require remarkably well-constructed track 
 and road-bed, and this is really one of the most serious diffi- 
 culties that has to be encountered. A road-bed and line 
 equipment of suitable character is expensive in a prohibitive 
 degree, unless the traffic is to be heavy and will pay unusu- 
 ally well.
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 269 
 
 Furthermore, the line must be of considerable length in 
 order to render high speed worth attaining. At the rate of 
 100 miles per hour, getting up speed and stopping requires so 
 much space and power that the advantages of so great a speed 
 are almost thrown away unless the road extends over a con- 
 siderable distance. 
 
 All these considerations point to the true electric railway 
 rather than to any automatic system. Nevertheless, several 
 of the latter have been proposed. For the most part they 
 embody no striking features, save a number of ingenious 
 devices for the proper governing of speed, automatic slowing 
 down and stopping, and the regulation of the train from a 
 distant point. These only possess interest in so far as they 
 have been carried out, and they have not been carried out 
 except in the Weems road before mentioned and in one other, 
 of which we will now speak. 
 
 The only automatic high-speed road which is in even ex- 
 perimental operation to-day is that strange freak of an in- 
 ventor's ingenuity the Portelectric system. Abandoning 
 all idea of driving wheels with their motor and motor arma- 
 ture, the inventor of this scheme has proposed to utilize 
 the direct attraction of solenoids for a hollow steel car 
 running on a suitable track. There are required a con- 
 tinuous series of solenoids over the entire length of the 
 line, each of them powerful enough to exert a strong attract- 
 ive pull on the car and furnished with current at the proper 
 time by automatic switches opened and closed by the passage 
 of the flying armature, a plan suggested long ago abroad.* 
 
 Working models were constructed, and, of course, worked 
 well, as model systems always do; and, encouraged by this, 
 still further experiments were made, and there has been 
 until recently near Boston an oval experimental track about 
 half a mile long. It consisted of an upper and lower rail, 
 the car being provided with flanged wheels above and below. 
 The car itself was a wrought-iron elongated shell with ogival 
 ends, 12 feet long, and weighing 500 pounds. The solenoids 
 were placed six feet apart, and each was composed of 20 
 
 * See English patent No. 58, of 1862.
 
 270 
 
 THE ELECTRIC RAILWAY. 
 
 pounds of No. 14 copper wire, having a resistance of about 
 5 ohms ; the successive solenoids were thrown into circuit in 
 front of the car by the action of a contact wheel mounted 
 upon the latter and running upon the upper rail, which was. 
 divided into sections and utilized as a conductor. 
 
 It was expected that enormous speeds would be reached ; 
 this belief, however, has not been justified, as nothing even 
 roughly approximating the speed obtained by one of the 
 authors in an experimental run on the Weems system has 
 ever been reached. The efficiency of such apparatus is ob- 
 viously, from the character of the electro-magnetic action 
 involved, low ; and the sparking from the inductive currents 
 set up in the helices as the contact is broken has proved to- 
 be most severe, although this might be partially obviated by 
 short-circuiting the helices instead of cutting them out. 
 
 An idea of the actual efficiency of the apparatus may be 
 obtained by considering the fact that between 9 and 10 elec- 
 trical horse-power were required to drive the car at a speed 
 of 33 YZ miles per hour 50. feet per second. Portions of the 
 track were level, and in other portions there were grades of 
 small length, but rising as high as 4 per cent. The total 
 weight of the car being 500 pounds, assuming any reasonably 
 plausible traction co-efficient, it is evident that the efficiency 
 obtained was quite low. This much space is devoted to the 
 discussion of this very unusual system simply by reason of 
 its striking peculiarities, and not from any -probability that 
 such a device can ever supplant the electric motor' proper ; 
 unless, possibly, on a small scale, as a substitute for the 
 pneumatic tube. 
 
 As it is, the present chapter has been necessarily some- 
 what disjointed in character, having dealt with a wide range 
 of subjects treating schemes practical and impractical, pos- 
 sible and abandoned as hopeless. 
 
 To the minds of the authors the most hopeful of the 
 methods mentioned are some of the modified conduits. 
 There are no signs as yet that any of these deserves more 
 extended mention than we have accorded to it here. Never- 
 theless, their development should be very carefully watched ; 
 for the direction is one from which something important
 
 MISCELLANEOUS METHODS OF ELECTRIC TRACTION. 27 1 
 
 may be looked for, and from which something practical may 
 be reasonably hoped. 
 
 The slotted conduit, although it has been shown from 
 experience that it may be successful, is both commercially 
 and electrically somewhat discouraging in our American cli- 
 mate; possibly its very best chance for giving good results 
 is in the conversion of some of the existing cable roads into 
 electric roads. With conduits so large as those employed 
 for cables the difficulties of drainage would be reduced to a 
 minimum, insulation would probably be practicable, and, at 
 least on roads of considerable length, as good efficiency could 
 be obtained as with cables. With respect to any and all of 
 the other devices, we can only advise our readers to suspend 
 judgment awaiting further evidence.
 
 CHAPTER X. 
 
 HIGH-SPEED SERVICE. 
 
 UNDER this heading it is intended to consider such service 
 as involves greater speeds, distances, and, usually, greater 
 loads than occur in street-railway work, properly so called. 
 
 To-day there is little to record of actual accomplishment 
 in this direction. The City and South London Railway 
 stands as a unique example of rapid transit for large cities, 
 in which several cars are run in a train, on a special roadway, 
 permitting higher than any allowable street speeds. 
 
 In 1886 Mr. F. J. Sprague conducted experiments on the 
 New York Elevated Railway, having in view the replacement 
 of the steam locomotives in a service requiring five loaded cars, 
 each about 35 feet long, to be hauled at a maximum speed of 
 30 miles per hour. Mr. Leo Daft, during the years 1888 and 
 1889, made similar experiments. The failure to attain im- 
 mediate success may be traced in both cases to several causes 
 immaturity of motor design at that time, limitation of 
 money available for making needed changes, difficulty of 
 experimenting without disturbance of regular traffic, and occu- 
 pation of the experimenters in other absorbing work. 
 
 In the light of the experience of to-day no one of the 
 competent engineers who has had to do with the electric-rail- 
 way work of the last few years would hesitate, if supplied 
 with the proper amount of money, to undertake the success- 
 ful installation of an electric rapid transit system meeting 
 all the requirements of the New York elevated railway or 
 London underground service. The conditions are in many 
 respects much simpler than those met with on the streets. 
 Especially is this true in regard to the placing of the con- 
 ductor system in the common street, where the convenience 
 of the electrical engineer must be subordinated to every 
 immemorial right of man, beast, or tree.
 
 HIGH-SPEED SERVICE. 
 
 2/3 
 
 And the motors may be placed where dust and mud do not 
 corrupt and stones do not break in the fields. 
 
 A brief description of the City and South London plant 
 will illustrate the comparative simplicity of this class of work. 
 The propelling power for each locomotive comes from two 
 motors, the armatures of which are concentric with the axles 
 and keyed directly thereto. Each armature is rated at about 
 50 horse-power, at a normal speed of 25 miles per hour. The 
 motor frames are inclined upward as shown in Fig. 1 16. For 
 the outgoing side of the circuit the working conductor, 
 divided into convenient sections, is of channel steel. Cur- 
 rent is taken from this through a sliding contact shoe attached 
 
 CONDUCTOR RAIL LEVEL 
 
 FIG. n6. -ELECTRIC LOCOMOTIVE OF CITY AND SOUTH LONDON RAILWAY. 
 
 to the locomotive. For the other side, the rails are used as 
 in ordinary practice. The channel steel is insulated from 
 the road-bed by glass insulators, the general line pressure 
 being 500 volts. The motors are series wound and regula- 
 tion is effected through a rheostat placed in the main circuit. 
 
 We have been able to obtain but little information con- 
 cerning the actual operating expenses of this line. Reports 
 indicate that they have at least been satisfactory to the bond- 
 holders. 
 
 The subject of the relative economy of steam and electric 
 train service was discussed by one of the authors in a 
 paper read before the American Institute of Electrical En- 
 18
 
 274 THE ELECTRIC RAILWAY. 
 
 gineers in May, 1890. As the matter was then taken up 
 quite in detail, and as no later developments have been found 
 to alter the conclusions then reached, we draw largely upon 
 that paper for this chapter. 
 
 As will be seen, the question in regard to electricity is not 
 "Can it be done?" but "Can it be done more economically 
 than by steam?" 
 
 In the light of present achievements we may state without 
 argument the following propositions : 
 
 First It is possible to construct motors capable of doing- 
 the maximum work required to-day in transportation. 
 
 Second It is possible continuously to generate electrical 
 energy equal to the capacity of any number of such motors. 
 
 Third At any desired loss and over any desired distance 
 it is possible to supply by the running-contact method the 
 necessary current, at considerable pressure, for the working 
 of such motors. 
 
 Should there still be question in the mind of any as to 
 the value of the running-contact method at speeds much 
 higher than those commonly used, it may be stated that 75 
 amperes at 500 volts have been thus continuously supplied 
 to a car moving at more than 1 10 miles an hour.* 
 
 These premises being established, further discussion di- 
 vides itself into three parts: 
 
 First, As to the mere possibility, without reference to the 
 economy, of steam and electric propulsion under given con- 
 ditions. 
 
 Second, As to the relative cost of exerting in a locomotive 
 any unit of power by electric, as compared with direct steam, 
 motors. 
 
 TJiird, As to the relative amount of power required by the 
 two agents to transport a given paying load under given 
 conditions. 
 
 In using the word steam as above we have in mind only 
 the direct application of steam power on the tnx-ks. The 
 case of cable propulsion is not here compared, as that has 
 
 * See "Report of High-Speed Electric Railway Work," read Feb. 24, 1891, before the 
 American Institute of Electrical Engineers by O. T. Crosby.
 
 HIGH-SPEED SERVICE. 3/5 
 
 within its restricted field of application already often been 
 compared with horse, steam, and electric power. 
 
 The limiting possibilities of locomotion may be understood 
 by considering a prolongation of the lines of present practice 
 in the direction, first, of loads handled; second, grades 
 climbed; third, speeds attained; and, fourth, length of con- 
 tinuous runs. 
 
 Since the effect of grade as compared with load is simply 
 to increase the tractive effort required for a given load and 
 speed, it need not be separately treated, except that there 
 has been some question of increase of adhesion in the ground- 
 return method, which in extreme cases might appear as an 
 advantage for electric propulsion. The matter is not of 
 great importance; and we will refer to it only so far as to 
 say that general experience and some special tests show 
 that the adhesion co-efficient is not increased in any practi- 
 cal degree by the mere passage of the current from wheel 
 to rail. 
 
 The capacity of an electric engine, like that of a steam 
 engine, to haul any given load is measured by the tractive 
 effort possible to be produced and the relation between 
 weight and adhesion for given track conditions. The ready 
 multiplication of cylinders in the one case and armatures in 
 the other, while maintaining mechanical unity and the ready 
 coupling of distinct locomotive units, renders the whole ques- 
 tion of capacity to exert a given horizontal effort, without 
 regard to the time element, unimportant. It goes without 
 saying that, if desired, a single armature may be constructed 
 capable of exerting as great a drawbar strain as any locomo- 
 tive now in use. 
 
 As to limiting speeds, it is not easy to-day to make even 
 " an educated guess," either for steam or electric propulsion. 
 The high figures for steam that have been recently presented 
 both from England and America are higher than the limiting 
 figures as they would have been given by many competent 
 authorities only a few years ago. Eighty-six miles per hour 
 in England, on the Northeastern Railway, eighty-seven miles 
 per hour, and later a rate of 90.45 miles per hour, in this 
 country, on the Reading iRailway, have been reported since
 
 2 -5 THE ELECTRIC RAILWAY. 
 
 Jan. i, 1890. These runs a-re noteworthy not only for 
 the fact of unusual speed, but because, as shown by indicator 
 cards in the English case, and as may be deduced from the 
 consideration of the maximum cylinder power in the Ameri- 
 can case, the train resistances are far below the values that 
 would have been predicted by even the most liberal of the 
 received formulae on the subject. The total resistance per 
 ton, as per indicator card, in the 86-mile run was only 13.4 
 pounds. According to Searles' formula, adopted by Well- 
 ington, it should be 69 pounds, engine and tender being 
 taken at 50 tons. The load of 347 tons was carried at 86 
 miles per hour by an expenditure of 1,068 horse-power this 
 on a level. The engine was compound. 
 
 Would it be possible to attain a speed twice as great, or 
 say 150 miles per hour? 
 
 A driver 24 feet in circumference must revolve 550 times 
 per minute in order to travel 13,200 feet per minute, or 150 
 miles per hour this without slip. Since in the case consid- 
 ered the revolutions per minute reached 309, and since in 
 the Reading case a much higher rate must have been reached 
 the drivers being smaller and since on stationary engines 
 a speed much above 550 revolutions per minute has been 
 attained, it seems beyond question that from this point of 
 view the supposed case is quite possible. 
 
 Considering the matter of steam supply, we are again 
 brought to consider the whole question of train resistances 
 at all speeds. 
 
 Total resistance to motion should 'be sharply divided into 
 two classes: the resistance due to motion through air, and 
 that due to friction and blows between vehicle and track and 
 to friction and blows between parts of the vehicle. 
 
 For the most part, those who constructed the formulas 
 now found in the text-books worked on road-beds far infe- 
 rior to the best work of to-day, at speeds much less than 
 those now attained, and with wrong values for at least one 
 of the species of resistance the atmospheric. On this point 
 there has recently been presented, as the result of experi- 
 ments at high velocities, a formula showing the pressure to 
 be approximately a function of the first, instead of the second,
 
 HIGH-SPEED SERVICE. 2// 
 
 power of the velocity as ordinarily assumed.* A convenient 
 datum point may be given, stating that at 100 miles per hour 
 the pressure on one square foot, normal to the direction of 
 motion, is 13 pounds; while proper shaping of the front may 
 reduce this to 6.5 pounds. The absolute values given, while 
 corresponding quite closely with those of received formulae 
 in the neighborhood of the velocities heretofore experimen- 
 tally attained, depart widely from those assumed for velocities 
 higher than 30 miles per hour, and calculated by the quad- 
 ratic relation between velocity and pressure. Using the 
 more trustworthy values, we are able to separate more nearly 
 than heretofore has been possible the atmospheric from all 
 other resistances met at high velocities. Some inaccuracy 
 still remains, by reason of the difficulty of obtaining exact 
 measure of the resisting areas in a train ; but by study of 
 careful tests made by others on the New York Central and 
 on English roads we find that over the range from about 40 
 up to 80 miles per hour the tonnage co-efficient seems prac- 
 tically constant at 8 pounds. This, of course, applies only to 
 first-class road-bed and rolling stock. Whether this co-effi- 
 cient remains constant at higher speeds we do not know. 
 There is no reason to assume, as has often been done, that it 
 increases with the square of the velocity; and, on the other 
 hand, it will not be safe to assume constancy. From experi- 
 ments made with a single 2. 5 -tons car at about 100 miles per 
 hour, the tonnage resistance at that speed seems to be about 
 20 pounds per ton.f Though this value seems quite high as 
 compared with the eight pounds at 80 miles per hour, the 
 difference is in large part to be explained by the poor condi- 
 tion of the track used for the experiment and a constant 
 curvature which would call for about four pounds per ton. 
 Until better evidence can be had it will *be safe, at least, to 
 assume a value of 20 pounds per ton, on a first-class track, 
 with good rolling stock, at 125-150 miles per hour. 
 
 Having made this necessary digression, we may return to 
 the matter of steam supply, and state that by reducing weight 
 
 * "An Experimental Study of Atmospheric Resistance," by O. T. Crosby, London 
 Engineering, May 30, June 6-13, 1890; New York Engineering Neivs, May 31, June 7-14, 1890, 
 and The Electrical World. May 24, 1890. 
 
 t "Report of High-Speed Railway Work" above mentioned.
 
 278 
 
 THE ELECTRIC RAILWAY, 
 
 ma 
 tric. 
 
 Weight 
 and weig 
 chinery. 
 20 per cen 
 
 t: Load, 
 glit of ina- 
 Klectric. 
 nt. 
 
 JU 
 
 t.2 a 
 
 j; 
 
 2 
 
 1^' 
 ife 
 
 IP 
 
 r 
 
 k 
 
 
 I 1 I
 
 HIGH-SPEED SERVICE. 
 
 279 
 
 JTR 8^88,8 <s 
 
 r^ in^o N t^ O 4 O 
 
 >> ** " 
 
 ^ o "^^ -^ r 
 
 "^ r-: ""' ^- - 1" 
 
 ' S. ov 5" 
 
 m O O O 
 
 <? I; ^ v2 
 
 S. s. 
 I ^ 
 
 s:e 
 
 
 e:e 
 
 go_ 
 
 l-aj 
 S^S 
 
 S^K.'S.Se 
 
 O 6 6 6 O O r 
 
 00 "^ N t^OO 
 
 c % 
 
 sss. 
 
 !^o 
 
 ' ^ -j: 
 
 ; W' 
 
 02 
 
 ? S." 2 
 d 2 - -A o- o- 
 
 , q 
 
 jxg 
 
 II 
 
 I! 
 
 I 
 13 I 
 
 II
 
 2 8o THE ELECTRIC RAILWAY. 
 
 and area both to something less than one-half the existing 
 values in the 86-mile run, the same effort would produce the 
 speed 150 instead of 86. The area cannot be thus reduced, 
 but by assuming a greater reduction in weight say to 100 
 tons, or to little more than engine and tender maintenance 
 of the higher speed might become possible, with nearly the 
 same steam expenditure as in the recorded case. 
 . To attain that speed, from rest, might require such orig- 
 inal weight of fuel and such length of favorable track as to 
 make the feat practically impossible with steam. This leads 
 us to inquire into the dead weight necessary for hauling, say, 
 one ton at different speeds. Table I. gives the horse-power 
 required for exerting the tractive effort for one ton at various 
 speeds, at various efficiencies, with various values of cross- 
 section per ton, and with the two agents steam and elec- 
 tricity. 
 
 Column i shows speed in miles per hour from 20 to 140; 
 column 2, corresponding tonnage co-efficient, or resistance, 
 in pounds per ton, exclusive of atmospheric resistance. Col- 
 umns 3 to 8, inclusive, show horizontal effort needed for 
 overcoming atmospheric resistance under various assumptions 
 as to area exposed per ton, from i square foot to o. i square 
 foot per ton. The former figure corresponds nearly to the 
 case of a heavy locomotive propelling itself alone. As load 
 is put on behind it other ratios are formed. Oblique surfaces 
 are supposed to be reduced to equivalent normal surfaces. 
 Columns 9 to 14, inclusive, show rate of work in horse-power 
 per ton for the various cases of area exposed to atmospheric 
 resistance, efficiency of locomotive being taken at 90 per 
 cent. Columns 15 to 17, inclusive, show horse-power per 
 ton for the extreme and middle cases of exposed area, and 
 for locomotive efficiency of 80 per cent. Columns 18 to 20, 
 inclusive, show same for efficiency of 60 per cent. Columns 
 21 to 26, inclusive, show weight of coal and water per ton 
 carried for one hour, assuming 5 pounds of coal and 25 pounds 
 of water per horse-power hour on a steam locomotive. The 
 coal figure is very close to actual practice. The water figure 
 is less, but makes allowance for scooping water at convenient 
 intervals. Continuous scooping is not considered practical
 
 HIGH-SPEED SERVICE. 28 1 
 
 or economical. Columns 27 to 29, inclusive, show weight of 
 steam locomotive and tender required to generate the required 
 horse-power per ton, under the assumption of 100 pounds 
 per horse-power and 90 per cent, efficiency. Only three 
 cases of exposed area are taken ; that is, one foot, one-half a 
 foot, and one-tenth of a foot per ton. The weight of steam 
 locomotives is not calculated for any other efficiency figure 
 than 90 per cent., as this seems to be quite constantly attained 
 or surpassed. The assumption of 100 pounds per horse-power 
 is closely true for many good types of locomotive when 
 working at speeds from 60 to 80 miles. At lower speeds this 
 figure is too low, but it is assumed that for any ruling speed 
 engines may be built of minimum weight for that speed. In 
 passing through lower than ruling speeds both electric and 
 steam motors work at low output per pound of weight; hence 
 the assumption of constant weight per horse-power will not 
 introduce error materially affecting comparative results. At 
 higher speeds than 80 miles existing engines would show 
 less than 100 pounds per horse-power; but as their boiler 
 capacity is reached at that speed the necessary increase for 
 any regular w r ork would carry the weight figure to very nearly 
 the figure given for the 60 to 80 mile running. Columns 30 
 to 32 show weight of steam locomotive and tender, plus weight 
 of fuel and water, per ton hauled ; also weight of load freight 
 and freight car that may be hauled by such weight of motive 
 power; the load figures being obtained by subtracting the 
 motive power weights from 2,000 pounds. Columns 33 to 
 41 show corresponding figures for electric locomotives, under 
 the assumption of 60 pounds per horse-power, and at the 
 three efficiencies 90, 80, and 60 per cent. In 30 to 41, 
 inclusive, weight of load is written first, w r eight of locomotive 
 (plus tender, fuel, and water for steam) being written below. 
 The 60 pounds per horse-power covers weight of the con- 
 taining car for motors. We cannot here go into detailed 
 figures on this point, but believe that any investigation will 
 find the figure safe, supposing always that the unit be, say, 25 
 horse-power or more. Columns 42 to 53, inclusive, show 
 the horse-power required to be exerted for hauling one ton 
 of load, i.e., freight and freight car, the relation between
 
 282 
 
 THE ELECTRIC RAILWAY. 
 
 these two being taken as the same for either steam or electric 
 propulsion, hence not necessary to enter here. These col- 
 umns apply to steam at 90 per cent, and to electricity at 90, 
 So, and 60 per cent., and for the three cases of exposed area. 
 They are readily obtained from the previous columns by 
 making allowance for the horse-power necessary to haul that 
 part of every ton, total weight, which must go into motive 
 power, machinery, and fuel. Columns 54 to 62, inclusive, 
 show the ratios between horse-power required by the two 
 agents for hauling a ton of load (freight and freight car) at 
 the different speeds, efficiencies and area relations. 
 
 It is plain that if we can now obtain the ratio of cost per 
 horse-power, as given by the two agents, in the corresponding 
 cases, we can easily determine the speeds at which the one 
 or the other agent becomes the more economical. 
 
 Let us first obtain the cost of electric propulsion. 
 
 TABLE II. 
 
 COST OF ONE H. P. HOl' 
 
 ELECTRIC, IN CENTS. 
 
 
 
 
 800 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 5 ' 
 
 
 Engineer 0.4 
 Fireman 0.3 
 Dvnamo man 0.4 
 Helper . 02- 
 
 0.13 
 o. 10 
 0.13 
 
 0.08 
 0.06 
 
 0.037 
 
 0.04 
 0.03 
 0.04 
 
 0.04 
 
 0.03 
 0.04 
 
 0.04 
 0.03 
 
 0.04 
 
 0.03 
 0.04 
 
 0.04 
 
 0.03 
 0.04 
 
 0.04 
 0.03 
 
 0.04 
 
 0.04 
 0.03 
 
 0.04 
 
 Superintendence [0.30 
 
 O.IO 
 
 0.06 
 
 0.037 
 
 o.o 3 J 
 
 0.02 
 
 0.01 
 
 0.01 
 
 0.01 
 
 0.01 
 
 0.01 
 
 Oil, waste, and water j 0.15 
 Interest and depreciation 
 
 0.15 
 
 ~-' 5 
 
 0.15 
 
 o 028 
 
 0.15 
 
 0.15 
 
 0.15 
 
 0.15 
 
 0.15 
 
 0.15 
 
 Ditto electric plant | 0.057 
 Ditto building i 0.028 
 
 0:026 
 
 0.044 
 
 0.033 
 
 -.028 
 
 o'oii 
 
 o'oi 
 
 O.OII 
 
 o'oii 
 
 o'oii 
 
 O.OII 
 
 For this, form Table II., showing the elements in the cost 
 of one horse-power in stations of various capacity from 100 
 to 6,000 horse-power. Engineers and dynamo men are as- 
 sumed to receive 40 cents per hour, and to superintend a 
 maximum of 1,000 horse-power. This would produce in 
 some cases fractional engineers, as fora i,5oo-horse-power 
 plant ; but such complication has been avoided by assuming 
 a constant value per unit of power in the pay-roll element in 
 plants exceeding 1,000 horse-power. Firemen and helpers 
 are taken at 30 and 25 cents per hour, respectively. Super- 
 intendence at 30 cents per hour is apparently low, but is 
 equivalent to 60 cents for daylight hours, since the plant is
 
 HIGH-SPEED SERVICE. 
 
 283 
 
 able to run 24 hours with the same general superintendence 
 as for 12 hours. It is further supposed that the total of this 
 item will not require increase until the capacity reaches 3,000 
 horse-power, beyond which it remains constant per unit of 
 power. As no other element of cost is supposed to vary 
 beyond this point, the table shows here a minimum total cost 
 per unit and a constant cost beyond it. 
 
 Coal is assumed to cost $3 per ton, and to be consumed 
 at the rate of 3.2 pounds per electric horse-power hour in the 
 dynamo. A slight error is made in taking the rate of con- 
 sumption as constant while changing the capacity of the 
 engines. The cost of the steam plant is taken to vary from 
 $50 per horse-power, in a small plant, to $20 in a plant of 
 1,500 horse-power. The dynamo plant is taken to vary from 
 $50 to $20 per horse-power, in going from 100 to 1,500 
 horse-power. 
 
 TABLE III. 
 
 TOTAI. COST OF ONE H. P. HOUR. 
 
 Output in per 
 cent, of c a- Hours of work 
 pacity while per day. 
 working. 
 
 Capacity of Station. 
 
 ,00 
 
 300 
 
 500 
 
 800 
 
 1,000 
 
 1,500 
 
 2,000 
 
 3,ooo 
 
 J 
 
 
 4 
 
 42 
 
 is 
 
 
 .29 
 30 
 52 
 
 
 .06 
 
 III 
 
 0.9148 
 28 
 
 o.Sfo 
 0.888 
 0.955 
 
 0.835 
 0.855 
 0-95 
 
 0.830 
 0.849 
 0.895 
 
 0.825 
 0.829 
 0.867 
 
 1 
 
 
 
 4 
 
 62 
 
 
 36 
 
 
 12 
 
 0.947 
 
 0.88 
 
 0.86 
 
 0.85 
 
 0.83 
 
 90 1 
 
 
 8 
 
 77 
 
 
 
 
 17 
 
 
 .98 
 
 O.OI 
 
 0.88 
 
 0.87 
 
 0.853 
 
 
 
 3 
 
 
 
 .61 
 
 
 29 
 
 
 .06 
 
 0.99 
 
 0.94 
 
 0.925 
 
 0.89 
 
 
 
 4 
 
 8 
 
 
 45 
 
 
 18 
 
 .987 
 
 0.92 
 
 0.91 
 
 
 .88 
 
 0.86 
 
 Bo < 
 
 
 8 
 
 3.06 
 
 
 
 
 
 
 o 98 
 
 0.936 
 
 
 
 
 
 
 
 3-42 
 
 
 73 
 
 
 37 
 
 
 13 
 
 1.06 
 
 
 
 .98 
 
 -95 
 
 J 
 
 
 4 
 
 8.19 
 
 
 57 
 
 
 26 
 
 
 .04 
 
 0.96 
 
 0.925 
 
 
 .91 
 
 0.89 
 
 7 { 
 
 
 8 
 
 3-42 
 
 
 .68 
 
 
 3 2 
 
 
 .09 
 
 
 .01 
 
 0.952 
 
 
 94 
 
 0.91 
 
 I 
 
 
 3 
 
 3-82 
 
 
 .90 
 
 
 48 
 
 
 .19 
 
 
 .09 
 
 i. 02 
 
 
 .01 
 
 0.97 
 
 
 
 M 
 
 8 
 
 3-62 
 3-85 
 
 
 
 
 36 
 44 
 
 
 !i6 
 
 
 .01 
 
 .06 
 
 0-975 
 015 
 
 
 .96 
 
 .00 
 
 0.94 
 0.965 
 
 
 
 3 
 
 
 
 .11 
 
 
 63 
 
 
 30 
 
 
 17 
 
 .085 
 
 
 .07 
 
 1.015 
 
 50 J 
 
 
 4 
 8 
 
 4.22 
 4-50 
 
 
 95 
 
 
 it 
 
 
 .20 
 
 .27 
 
 
 
 045 
 .09 
 
 
 .076 
 
 39 
 
 
 
 3 
 
 5-09 
 
 
 .40 
 
 
 83 
 
 
 43 
 
 
 .28 
 
 .19 
 
 
 '7 
 
 
 
 
 4 
 
 8 
 
 5-i! 
 
 
 .28 
 
 .0 
 
 
 8 7 i 
 
 
 35 
 
 
 .21 
 
 15 
 
 
 '3 
 
 i 
 
 40 
 1 
 
 
 
 5-47 
 5.20 
 
 
 12 
 
 
 
 
 $ 
 
 
 33 
 .50 
 
 3 
 
 
 .23 
 35 
 
 ^28 
 
 
 
 4 
 
 6.63 
 
 2.84 
 
 
 10 
 
 
 59 
 
 
 .40 
 
 325 
 
 
 3 
 
 .26 
 
 3 1 
 
 
 8 
 
 7.08 
 
 3.02 
 
 
 27 
 
 
 
 
 .40 
 
 
 38 
 
 3 2 
 
 1 
 
 
 2 
 
 8.06 
 
 3-52 
 
 
 65 j 
 
 [98 
 
 
 73 
 
 57 
 
 
 54 
 
 44 
 
 The cost of buildings is taken to vary from $25 to $10 per 
 horse-power. This is the most indefinite item. Interest, 
 maintenance and taxes are roundly assumed at 10 per cent, 
 per annum on the whole plant.
 
 284 THE ELECTRIC RAILWAY. 
 
 With Table II. as a basis, Table III. has been calculated, 
 giving the total cost per horse-power hour in stations of 
 various capacities, working at various percentages of their 
 full capacity and for 24, 18 and ^^hours per day, respect- 
 ively. 
 
 A glance at the table shows that in a loo-horse-power plant 
 the cost varies from 2.42 cents for a 24 -hour run at full 
 capacity, to 8.06 cents for a 3<D-per-cent. output continued 
 only 12 hours per day. 
 
 This extreme case would doubtless be ameliorated by 
 dropping the superintendence and combining engineer and 
 fireman though to this the unions might object. This 
 capacity is smaller, however, than need be considered. 
 
 The minimum cost given by the table is 0.816 cent; this 
 is for a 3,ooo-horse-power plant, working at full capacity 24 
 hours per day. 
 
 The next element of cost that of the conductors for the 
 current is obtained from Table IV., showing the investment 
 in dollars for the copper required to transmit one horse-power 
 a distance of one mile, at varying initial and final pressures. 
 The constants for this table were thus obtained : Taking a 
 well-known line wire, it is found that from No. 4 to No. 
 ooo B. W. G. the average weight ratio of insulated to bare 
 wire, per unit of length, equals 0.844. When bare copper 
 sells at 15 cents per pound this insulated wire sells at 20 
 cents to 22.5 cents on the copper alone, when insulated. If, 
 therefore, we take copper at 25 cents per pound we provide 
 for a very good insulation. The cost of one mil-mile is thus 
 found to be $0.004. Combining this with the familiar for- 
 mula 
 Q -^ _ 16,600 X h. p. transmitted X distance in feet. 
 
 ~ E. M. F. at motor X volts lost X motor efficiency' 
 the tabulated values have been determined from the resulting 
 formula : 
 
 Cost- 76 ' 32 m 
 
 - (E^Tvyv 
 
 in which E = E. M. F. at station, V = volts lost on line ; motor 
 efficiency is taken at 90 per cent, and distance at 5,280 feet. 
 The tabular figures give the investment. To reduce to actual
 
 HIGH-SPEED SERVICE. 285 
 
 cost per horse-power hour, the rates of interest and depreci- 
 ation and the ratio between power transmitted and power 
 possible to be transmitted must be had. For one year the 
 possible horse-power hours = 365 X 24. Take annual inter- 
 est and depreciation at 1-16 the investment and ratio of actual 
 to possible power transmitted per annum at i.o, 0.8, 0.6, 0.4, 
 0.2, o. i , 0.05, then the divisor of the tabular number becomes 
 140,160, 112,128,84,096, 52,064, 28,032, 14,026 and 7,008, 
 respectively. 
 
 It remains to obtain values for the distance of transmission. 
 
 Let n = number of miles of line supplied from one station. 
 
 h = horse-power required for unit locomotive. 
 
 K = maximum number locomotives per mile at any time. 
 
 n h K = total power to be transmitted at time of K. 
 
 A = percentage of dynamo power lost on line. 
 
 n hK 
 
 = maximum power to be generated. 
 
 = lost on line. 
 
 b = cost in cents of generating one horse-power hour in 
 station. 
 
 r = ratio of average power required to maximum power 
 required. 
 
 L interest and depreciation on supporting structure. 
 
 This last may be omitted from the calculation determining 
 division of line into sections, since it remains the same what- 
 ever that division may be. Omitting this, the expression for 
 cost of transmission becomes : 
 
 760,320 n 3 h K r bnhKA 
 
 ~~ (E V)Vx 140, 160 ' r (100 A) 100 
 _5.42n s hKr bnhKA 
 
 : ~(E _ V) V * r (loo A) 100 
 
 Supposing that the time schedule of trains, horse-power 
 required per train, and efficiency of motors be known, and 
 that the initial E. M. F. be in all cases taken as high as the 
 state of the art permits, the only variables remaining in this 
 expression are n, b, and A. This latter, the value for the 
 drop on the line, will generally be determined by conditions 
 other than those of strictest economy as shown by getting a
 
 286 
 
 THE ELECTRIC RAILWAY. 
 
 minimum value for cost of transmission. We must have a 
 reasonably uniform E. M. F. all along the line, in order that 
 the motors may work satisfactorily. It will not be wide of 
 the mark to assume 10 per cent, as a limiting variation of 
 line potential. This is the figure generally assumed in cal- 
 culating wire for street railways. 
 
 The cost of one horse-power hour, b, must vary with the 
 capacity and conditions of working of the station ; hence it 
 is a function of ni.e., length of unit section, of A, and of 
 that inexpressible variable the conditions of the service. 
 This stands in the way of obtaining any perfectly general 
 definite expression for n, which, but for this, would result 
 from placing the first differential co-efficient of C, with respect 
 to n, equal to zero, and solving to find the value of n giving 
 a minimum value for C. If trains be run at very short inter- 
 vals, increase of n would be followed by proportional increase 
 in capacity of station ; but, as above shown, this need not go 
 
 g 3.5 
 
 "C 3. 
 
 CAPACITY IN HORSE POWER 
 FIG. 117. VARIATION OF COST OF CURRENT WITH CAPACITY OF STATION. 
 
 beyond 3,000 horse-power unless the service be so heavy as 
 to require practically contiguous stations of that capacity. 
 If we suppose a case of this short interval service, so short 
 that a change in n will not be followed by any change at the 
 station in the relation between maximum capacity and aver- 
 age output, or in the number of working hours, but only in
 
 HIGH-SPEED SERVICE. 
 
 28; 
 
 the normal capacity, the relation between b and n for two 
 cases of average output and working hours is shown by Fig. 
 117. The equation of the 50 per cent, curve from 200 to i ,000 
 horse-power seems to be very nearly 
 
 _ n 2 QSn -f- i ,000 
 
 12 (n 100) 
 
 If the trains be run at very long intervals we may require 
 no greater station capacity for 20 than for 10 mile sections; 
 
 .5 .6 .7 .8 .9 1-Ct. 1.1 1.2 1.3 1.1 1.5 1.8 1.7 
 COST PER H.P.PER HOUR RUN 
 
 FIG. n8. VARIATION OF COST OF CURRENT WITH CAPACITY OF STATION. 
 
 but the relation of output to normal capacity will vary, and 
 possibly also the number of hours during which the working 
 force would require to be kept on pay. 
 
 Taking the case of a change in average output only for a 
 i,ooo-horse-power station, we have the curves in Fig. 1 18 for 
 24, 1 8, and 12 hours' work. The equations are nearly of the 
 same form, and are approximately the equations of arcs of 
 circles referred to an. origin outside the circle. But each 
 would contain different constants, and would vary more or 
 less from the exact formula for the circle. 
 
 If in the application to a particular case the algebraic 
 expression for b above given, Or any other resulting from 
 any of the possible progressions through Table III., be sub- 
 stituted in equation (2), we may, by differentiation, solve that 
 particular case for the most economical value of n. If the 
 relation between b and n cannot be algebraically expressed, 
 then the proper value for n may be determined by a few trial 
 values, the corresponding values of b being taken from the 
 table.
 
 2 88 THE ELECTRIC RAILWAY. 
 
 For the present purposes of comparison we will assume a 
 case not more favorable than might often be met on busy 
 steam lines i.e., a station of 2,000 horse-power normal 
 capacity, working 18 hours per day at 40 per cent, of its nor- 
 mal output, the cost per horse-power being 1.25 cent. 
 
 To obtain the cost of the line we will assume that the 
 average distance of transmission is five miles. This would 
 correspond to one station for every twenty miles of road. 
 We will also assume 5,000 volts initial E. M. F. and 10 per 
 cent. " drop. " From Table IV. the copper investment is found 
 to be 34 cents for one mile. Then for five miles the invest- 
 ment equals $8.50. Making assumptions as to service cor- 
 responding to those for the station, we have cost for one 
 horse-power equaling $8.50 -=- 20,000 := 0.042 cent. 
 
 The structure for carrying the conductors may be built for 
 $2,000 per mile. Interest and depreciation would then be- 
 come 200 per annum. This total is almost wholly inde- 
 pendent of the power transmitted; hence the cost per unit of 
 power will vary inversely with the number of units transmit- 
 ted. Assuming a constant distribution of one joo-horse-power 
 train for every 20 miles of line, we have the cost of this item 
 for one horse-power hour : 
 
 20,000 ~ 365 X 24 X 25 = 0.09 cent. 
 
 Reaching the locomotive, we must add, supposing an aver- 
 age output of 500 horse-power, 0.08 and 0.06 cent, respect- 
 ively, for driver and his assistant. The latter is necessary 
 only as a substitute for his principal in case of emergency, 
 but as such he would doubtless always be placed on trains of 
 considerable value. 
 
 Repair on electric locomotives is not as yet well defined. 
 That the repair bill must be far less than in the case of steam 
 locomotives follows almost necessarily from the great reduc- 
 tion in the number of parts, especially of moving parts. 
 
 From Mr. Arthur Wellington's very valuable work on 
 railways, we take the figures showing percentage distribution 
 of locomotive repairs by parts. 
 
 Boiler, 20 per cent. ; running gear, 20 per cent. ; machin- 
 ery, 30 percent.; lagging and painting, 12 percent.; smoke- 
 box, etc., 5 per cent.; tender (running gear, 10 per cent.;
 
 HIGH-SPEED SERVICE. 
 
 28q 
 
 gi 
 
 O. 
 
 O 
 
 I 
 
 SP 
 
 1 
 
 OcoO CO *) O Ococo r^o min-i-'^-cococ^c^rommco'TO 
 
 ooooooooooooooooo 
 
 OOOOOOOOO 
 
 o 
 
 c3"3- c SS'~222 c 8' v ^?1r3- 
 
 d d d d d d d d d d d d d d d d d 
 
 co o co -t 
 
 o o o o o o o"o 
 d d d d d d d d 
 
 co" 
 
 ^* ?1 ^ 8 - :r "- 2 .o^o3' v 82 
 
 d d d d d d d d d d d d d d d d d d d d d d d d 
 
 r^ 
 
 i-'ui'c^^ ?! f4^ M M n^'o t o'o V o > o 
 
 d d d d d d d d d d d d d d d d 
 
 d d d d d d 
 
 o" 
 
 co-)--)-fNcocNOcoOinc<OO Oco co co 
 
 CiSOvrCOdOIClMMMMMMOOOO 
 
 "< d d d d d d d d d d d d d d d d 
 
 'S'g'S 2 
 
 d d d d d 
 
 1 
 
 OO^tOOMOcOO*t<NC-.OOOOOONOO 
 
 d d d d d d d d d d d d d d d d 
 
 O O 
 
 L, ""* 
 
 
 4 IS. Wl 4 m Ci CI N M M M M M M M M M M M <9 
 
 oooooooooooooooooooo 
 
 W ^ 
 
 Qco 5 -to C-ICOO n OcocoOO vn -tO 
 
 CO MO 
 
 M d d d d d d d d d d d d d d d d 
 
 d d d 
 
 3 
 
 J3 rf 
 
 OOcoc4O-tNOco-tNOOcocOOO! 
 
 O -1- 
 
 wwOOOOOOOOOOOOOOC 
 
 
 
 8 
 
 
 CO 
 
 d 
 
 ci -i d d d d d d d d d d d d d d d 
 
 | 
 
 
 
 ci d d d d d d d d d d d d d d >-' 
 
 8 
 
 S^S^g^SS^^fi 
 
 
 CO OOOOOOOOOOOO 
 
 1 
 
 8 2 2>S 8 ci"Rc 8^ C J 
 
 
 rtMHMOOOOOOO-C< 
 
 1 
 
 *SR&^R**aS 
 
 
 
 1 
 
 ^:ii^!!5 
 
 , 
 
 8S8 
 
 0*0 
 
 
 mM 
 
 
 88888888888888888 
 
 M pj Jn^tnO r-co ^5 M to co too
 
 2 g THE ELECTRIC RAILWAY. 
 
 body and tank, 3 per cent.), 13 per cent.; total, 100 per 
 cent. 
 
 Of these items we may at once and with certainty strike 
 out boiler, smoke-box, etc., and tender thus dropping 38 
 per cent, of the total. Having no boiler to carry, the running 
 gear will be less in quantity. 
 
 The wear will be less, due to the use of rotary instead of 
 reciprocating effort. It will then be fair to reduce this item 
 by half, making another saving of 10 per cent. So in the 
 machinery item there can be no question that with the rapid 
 advance toward slow-speed motors, reducing gear, and sound 
 insulation methods, the great advantage of having only one 
 moving part in the motor itself must operate to reduce very 
 largely the repair figure, probably to half its value in a steam 
 locomotive, leaving it at 15 per cent. In lagging and 
 painting, the omission of boiler and other parts must 
 again effect a reduction, say to 6 per cent. The total 
 reduction thus plainly indicated must be then very nearly 
 70 per cent. 
 
 The actual cost of repairs to-day on steam locomotives is r 
 on the Pennsylvania Railroad, nearly 0.75 cent per horse- 
 power hour. Reducing as above, this figure becomes 0.22 
 cent for electric traction. This refers to engines of consid- 
 erable power, say from 400 to 1,000 horse-power capacity. 
 The figure for both steam and electric motors w^ould go 
 up for smaller powers. 
 
 The interest charge results from considering the cost of 
 an electric locomotive as $50 per horse-power, and the duty 
 as six hours per day, full capacity. The average duty of 
 steam locomotives is only about three hours. The higher 
 figure results from a smaller number of repairs necessary, 
 due to greater simplicity of parts. Then 
 
 ^ ,- 5O.OO X O.o; 
 
 Interest for h. p. hour = - -~- = o. 1 1 cent. 
 
 Before summarizing we must know something of the 
 efficiency of the system. If the locomotive be of 90 per 
 cent., or 80 per cent., or 60 per cent, efficiency, we must 
 generate, respectively, 1.25, 1.4, and 1.85 horse-power hour 
 in the station for every horse-power hour actually delivered
 
 HIGH-SPEED SERVICE. 
 
 2 9 I 
 
 to drivers (line loss being supposed constant at 10 per cent.). 
 We should then have in the station 1.56, 1.75, and 2.30 
 cents, respectively. Then, for total: 
 
 Station 
 Conductor system 
 Structure for same 
 
 1.56 
 0.042 
 0.09 
 
 i-75 
 0.04 
 o 09 
 
 2.30 
 0.04 
 
 O OQ 
 
 Wages 
 Repairs 
 
 o. 14 
 
 o '>'> 
 
 0.14 
 
 0.14 
 
 Interest 
 
 O. I I 
 
 0. I I 
 
 0. I I 
 
 Totals 
 
 2.16 
 
 2. ^C 
 
 2 QO 
 
 
 
 
 
 Of the whole amount it is to be noted that the station item 
 is three-fourths. Taking the most favorable case shown by 
 the table 3,000 horse-power capacity, working for 24 hours 
 at 100 per cent, of capacity those figures become 1.5, 1.62, 
 and 2.00 cents, nearly, the reductions in the other items 
 being still a little indefinite, without making another series 
 of independent assumptions. 
 
 The cost of one horse-power hour exerted by steam loco- 
 motives is to be obtained only by some circumlocution, the 
 reports of cost being based on train-miles. As quoted by 
 Wellington, the coal consumed per passenger train-mile on 
 the Pennsylvania Railroad is very closely 50 pounds, and an 
 average performance will show 5 pounds coal per horse-power 
 liour while under way. This shows that one train-mile equals 
 10 horse-power hours. The coal consumption of a freight 
 train-mile is much higher, but the divisor would also be 
 higher, on account of the greater number of stops, in backing 
 and switching, and greater delays while on a trip, thus 
 increasing waste of coal. 
 
 Again, the Pennsylvania Railroad reports show a cost of 
 fuel per train-mile of 5 cents. The cost of coal to that com- 
 pany, as nearly as we can learn, is about $1.50 per ton. 
 Hence it would appear that 66 pounds per train-mile are 
 consumed. As the terminal losses of fuel and delay (getting 
 up steam and drawing fires, etc.) are known to be in the 
 neighborhood of 25 per cent, of the total consumption, we 
 have 6.6 (pounds) as the divisor, and the same ratio again 
 appears. The cost of one train-mile as to motor power
 
 29 2 THE ELECTRIC RAILWAY. 
 
 alone is given by the Pennsylvania Railroad as 22 cents. 
 
 And this is practically equal to the average for the United 
 States. The itemized statement is very carefully made up, 
 and seems to cover everything except interest on the engine 
 investment. 
 
 Knowing the annual mileage per locomotive to be about 
 20,000 and the cost to be $10,000, interest charge per train- 
 mile (not quite, but nearly, equivalent to engine mile) be- 
 comes 2.7 cents. Total cost becomes 24.7 cents, or 2.47 
 cents per horse-power hour. 
 
 We may safely use this figure in the comparison to be 
 made, since any positive error in the calculation of coal per 
 horse-power hour or negative error in horse-power hours per 
 train-mile will be offset by the difference in cost of coal per 
 ton to the Pennsylvania Railroad as compared with the value 
 assumed in Table III. 
 
 At $3, instead of $1.50, the fuel item would be 10 cents- 
 and the total motive power 29.5 cents. Leaving the statis- 
 tics of actual cost, we may reach, by the method of Table 
 III., nearly the same figure, considering the locomotive as a 
 i,ooo-horse-power steam plant of low first cost, burning 6 
 pounds of coal per horse-power hour, and working at one- 
 third capacity for about three hours per day. The result 
 thus reached is about 10 per cent, higher, but the 2.47 cents 
 seem more reliable. 
 
 These values multiplied into the corresponding values in 
 the last nine columns of Table I. give the following values of: 
 
 Power units, steam cost per unit, steam. 
 : <y/ l 
 
 Power units, electric cost per unit, electric. 
 
 This value, Table V., when greater than unity indicates 
 greater economy by electricity than by steam, and vice versa. 
 
 A glance at the table shows the dominating necessity of 
 increasing the efficiency of the mechanism delivering energy 
 from the electric line to the vehicle. We cannot count upon 
 a higher efficiency than 90 per cent, for the motor. Hence, 
 save in the case of putting the armature directly on the axle, 
 we cannot hope to reduce the total loss to less than 20 per 
 cent. a case permitting one set (i.e., two gears) of spur 
 gearing between armature and axle. As the ordinary rela-
 
 HIGH-SPEED SERVICE. 
 
 293 
 
 tion beween tonnage and resisting area will lie between the 
 second and third columns of each efficiency table, it appears 
 that with a 20 per cent, loss electricity becomes cheaper than 
 steam at about 70 miles per hour, the frequency of service 
 being such as is assumed above. 
 
 Ratios of cost of motive power = cosi ^ steam 
 
 cost by electricity 
 
 Efficiency of electric 
 engine. 
 
 10 per cent. 
 
 20 per cent. 
 
 40 per cent. 
 
 Tonnage and area 
 
 relation. 
 
 2& 3 
 
 2& 5 2 &8 
 
 2& 3 
 
 2& 5 
 
 2&8 
 
 2& 3 
 
 2& S 
 
 2 & 8 
 
 Speed. 
 
 5 
 9 
 5 
 
 57 
 
 .16 
 .19 
 24 
 
 :l 
 
 0.92 
 o-95 
 
 1. 00 
 
 0.85 
 
 0-95 
 
 1. 00 
 
 0.85 
 0.95 
 0.95 
 
 0.56 
 
 3 
 
 o^57 
 0.58 
 
 ?1 
 
 1-36 
 
 p p p p p p p 
 
 20 
 40 
 60 
 80 
 100 
 120 
 140 ^ 
 
 -94 
 .82 
 
 3 
 
 i 
 
 11 
 
 1.20 
 
 I.I3 
 
 0.66 
 0.80 
 
 
 
 
 In the case of 40 per cent, loss our new agent betters the 
 old only at 140 miles per hour. And yet this loss 40 per 
 cent. is about the best we do with our present systems of 
 electric street car propulsion. 
 
 Considering the very short life of the art, these results are 
 excellent. Indeed, excellence is shown in the mere fact of 
 success in competing with horses under conditions very trying 
 for any mechanism not made of india-rubber or whit-leather. 
 How great that success has been we need not here proclaim. 
 But the steam locomotive is a foeman more worthy of our 
 steel, or, rather, of our annealed soft iron and 99 per cent, 
 conductivity copper. 
 
 Consideration of all that precedes leads to the following 
 general conclusions: 
 
 1. A slow-speed armature placed on the car axle would 
 place the electric motor in the lead at all service speeds. 
 
 2. For speeds above 70 miles per hour an electric motor of 
 90 per cent, efficiency, working through gearing of 90 per 
 cent, efficiency, would prove more economical than the steam 
 locomotive save in cases of very infrequent service on very 
 long lines. 
 
 3. On lines for heavy traffic steam would be more eco-
 
 294 THE ELECTRIC RAILWAY. 
 
 nomical than electricity if motor and gearing have a combined 
 efficiency as low as 60 per cent., up to 100 miles per hour. 
 
 4. At speeds of 100 miles per hour and upward neither 
 steam at 90 per cent, nor electric apparatus at 60 per cent, 
 efficiency is commercially practicable. 
 
 5. Inasmuch as the saving of coal in stationary, as com- 
 pared with locomotive, engines is one of the chief causes of 
 the greater economy of electric propulsion at any speed this 
 advantage will increase with that difference and also with the 
 price of coal. 
 
 6. Any cause other than inefficiency of motor which in- 
 creases the power required to haul a ton of freight increases 
 the advantages of electricity, since it enlarges the value of 
 the coal difference and the dead-weight difference. 
 
 Thus, bad roadways and large areas exposed to atmospheric 
 resistance, as in street-railway work, lower the spee<J at which 
 electric motors of any efficiency become cheaper than steam. 
 
 7. In descending to small locomotive units the electric 
 motor loses less, relatively, of its advantage another reason 
 for success on street lines. 
 
 8. Multiplying the number of motors should be as far as 
 possible avoided. 
 
 9. In special cases cleanliness and compactness of electric 
 machinery may be of great value ; in case of very frequent 
 stops the possibility of returning to the line the energy now 
 wasted in brakes may be of considerable value. This, how- 
 ever, can be obtained only by sacrifice in the matter of dead 
 weight, as normal working is implied to be at comparatively 
 low magnetization. Loss due to low efficiency in starting can 
 scarcely be avoided, either in steam or electric engines. 
 
 10. Other minor pros and cons may be enumerated, but, 
 considering the general economic results of the two systems, 
 we reach no more definite conclusions. 
 
 While only one condition of station working has been 
 taken for final comparison, that case is an average one, and 
 comparative results would be but slightly affected by ordinary 
 variations. Extreme cases may be readily determined from 
 the tables and formulae presented. 
 
 1 1 . Some difference of opinion as to the proper values for
 
 HIGH-SPEED SERVICE. 295 
 
 the various constants is to be expected, but these differences 
 taken all along the line would probably nearly balance between 
 positives and negatives, leaving the general results and the 
 method unchanged. 
 
 Whether these conclusions be accepted or not, we believe 
 the tables and formulas here presented will be found valuable. 
 
 Consulting Table V., we find that rapid transit service for 
 large cities can be accomplished electrically with a saving in 
 motive power of about 20 per cent, on the cost by steam. 
 This supposes the electric motor to have an efficiency of 
 about 90 per cent., at 30 miles per hour, and contemplates 
 the construction of the armature concentric with the axle. In 
 respect to this point there would, at this writing, be little, if 
 any, difference of opinion. It should be added that in con- 
 sidering city service a more favorable case of station expense 
 might be assumed than that which has entered into Table V. 
 There we have taken a total capacity of 2,000 horse-power 
 working at 40 per cent, of that capacity during 18 hours, the 
 cost per horse-power hour being 1.25 cent. 
 
 But in the case of city service the load will be tolerably 
 uniform and the station capacity may be so determined that 
 the average output may be, say, 60 per cent, of the capacity. 
 We might, indeed, fairly take i cent per horse-power hour 
 for the station cost. This assumption would be followed by 
 an increase of about one-eighth in the figures given for u 10 
 per cent, loss" in Table V. We might, then, fairly say that 
 an economy of 30 per cent, may be had over the cost of 
 motive power by steam. Now, in the average rapid transit 
 system the motive power expense may be taken at abotit one- 
 third of the total operating expense. On this total, then, the 
 saving may be said to be about 10 per cent. 
 
 The conclusions here drawn from the tables above given 
 have been checked by careful detailed estimates made for 
 special projects now on foot in Chicago, 111., and by engineers 
 not familiar with the tables. The results were very nearly 
 identical, tending to show that, at least in all general discus- 
 sion of such work, very satisfactory guidance is given by the 
 figures here presented. 
 
 Concerning speeds considerably beyond those now com-
 
 296 THE ELECTRIC RAILWAY. 
 
 monly attained by steam, no experiments have been made 
 and reported save those made by one of the authors at Laurel, 
 Md.* The work was done for the Electro- Automatic Transit 
 Company of Baltimore, Md. The organizer of this company 
 was Mr. David G. Weems. Lack of means has thus far stood 
 in the way of accomplishing on a proper scale the demon- 
 stration required in order that the world may become familiar 
 with, and practically interested in, the practicable increase 
 of speed of transportation up to 125 or 150 miles per hour. 
 
 In the experiments above referred to a car about 30 inches 
 high by 24 inches wide by 20 feet long, carrying two motors 
 with their armatures keyed directly to the axles and weighing 
 about 2.5 tons, was driven around a circular track (2 8 -inch 
 gauge, 2 miles in circumference) at a final speed of about 
 1 1 5 miles per hour. Control was effected from the power 
 station, placed inside the circle. The voltage of the line was 
 500. An upper rail carried on a framework built over the 
 track served as one side of the circuit, the lower track rails 
 and earth as the return circuit. 
 
 Current was received from the upper rail through an up- 
 ward bearing brush. The track was of exceedingly light 
 
 FIG. 119. -PLAN AND ELEVATION OF EXPERIMENTAL CAR.-WEEMS SYSTEM. 
 
 construction quite unfit for the work attempted to be done. 
 Figs. 1 19 and 120 show the car and station. Valuable infor- 
 
 * See "Report of High Speed Electric Railway Work," by O. T. Crosby. Read before 
 the American Institute of Electrical Engineers, February 24, 1891.
 
 HIGH-SPEED SERVICE. 297 
 
 mation, however, was obtained concerning atmospheric and 
 tonnage resistance. 
 
 In the American Institute paper above referred to these 
 results were reported quite fully, and plans were shown for 
 
 
 FIG. 120. STATION AND TRACK AT LAUREL, MD.-WEEMS SYSTEM. 
 
 a passenger service or passenger and mail service at 125 
 miles per hour. Since nothing more definite has yet been 
 done in this direction we can only quote some of the con- 
 clusions reached, and further refer the reader to the paper 
 itself for details. 
 
 A. For purposes of demonstration it was proposed to the 
 company that they should build a track about 4 miles (not 
 less) in circumference, and should run thereon a train of two 
 or three cars drawn by one locomotive. Calculations were 
 based on the following data: 
 
 (1) A speed of 150 miles per hour on a level was to be 
 aimed at. 
 
 (2) Cross-section of car should be a minimum consistent 
 with the seating of passengers. This was taken at 6 feet by 
 5 feet 30 square feet (crowning higher along middle of car). 
 
 (3) Gauge of track should be standard 4' 8.5". 
 
 (4) Character of track should be simply the best that art 
 could contrive, rails to weigh from 65 to 90 pounds per yard. 
 
 (5) Electromotive force should be as high as the art of 
 insulation would permit. 
 
 (6) Whatever might be the number of cars in a train, all 
 were to be so connected as to present a continuous exterior, 
 thus presenting only one cross-section to the atmosphere.
 
 298 
 
 THE ELECTRIC RAILWAY. 
 
 (7) Atmospheric resistance at 150 miles per hour would be 
 taken at 15 pounds per square foot of cross-section, a wedge 
 or parabolic locomotive head being used. This value is 50 
 per cent, in excess of that indicated by experiments. 
 
 (8) Traction co-efficient, exclusive of air resistance, would 
 be taken at 150 miles per hour to be 25 pounds per ton this 
 for reasons set forth previously. 
 
 From the two resistance co-efficients just given, it followed 
 that for every square foot of cross-section, regardless of 
 weight, 6 horse-power would be required, and for every ton 
 of weight, regardless of cross-section, 10 horse-power must 
 be supplied. 
 
 FIG. i2i. HALF SECTION OF MOTOR FOR 
 HIGH-SPEED SERVICE. 
 
 FIG. 122. HALF PLAN OF MOTOR FOR 
 HIGH-SPEED SERVICE. 
 
 For a locomotive of about 600 horse-power, having 30 square 
 feet cross-section, the dead weight was calculated to be about 
 1 8 tons. Steel cars of equal cross-section were designed, 
 weighing, empty, about 5 tons, with a carrying capacity of 
 about 5 tons. 
 
 The important question as to power required could then 
 be tabulated thus : 
 
 Locomotive alone 
 
 Locomotive and one car loaded . '. . . . . . 
 
 Locomotive and two cars loaded 
 
 Locomotive and three cars loaded 
 
 150 miles 
 per hour. 
 
 360 H. P. 
 460 
 
 560 " 
 760 " 
 
 120 miles 
 per hour. 
 
 288 H. P. 
 369 " 
 446 " 
 528 "
 
 HIGH-SPEED SERVICE. 
 
 299 
 
 To perform the work here indicated two motors were to 
 be provided, one armature directly on each of the two axles 
 of the locomotive car. These motors, outlined in Figs. 121- 
 124, were of the Manchester type. They were supposed 
 to be at first connected in arc across a i,5OO-volt circuit, each 
 taking about 130 to 150 amperes and delivering about 250 to 
 300 horse-power. Should it be desired to experiment with 
 
 FIG. 123. HALF LONGITUDINAL SECTION OF ELECTRIC LOCOMOTIVE. 
 
 higher line potential, the motors in series would permit 3,000 
 volts to be easily tried. The armatures of the Gramme type 
 were 30 inches outside, 23 inches inside, diameter. The ar- 
 mature conductors were to be of about 40,000 circular mils 
 cross-section, or from 500 to 600 c. m. per ampere. Many 
 successful machines designed to work indoors go as low as
 
 3 oo 
 
 THE ELECTRIC RAILWAY. 
 
 this figure, and since the dissipation of heat must in such 
 case proceed more slowly than in the case here considered, 
 the figure seemed very safe. Moreover, it was intended to 
 introduce a blast of air from the car front, which should con- 
 tinuously flow through the open center of the armature. 
 
 FIG. 124. HALF TRANSVERSE SECTION OF ELECTRIC LOCOMOTIVE. 
 
 How far the rate of dissipation would in practice be carried 
 by such means cannot now be known, but all recent evidence 
 goes to show that the heat capacity would be much increased. 
 The cross-section of core in the armature was 3.5* X 40" = 
 140 square inches gross, or, deducting 1 5 per cent, for paper
 
 HIGH-SPEED SERVICE. 301 
 
 insulation between disks, and 8 per cent, for conductor slots 
 cut in periphery, no square inches net. Through this ar- 
 mature it was desired to force 22,000,000 lines (C. G. S. units) 
 or 100,000 per square inch = 15,200 per square centimeter. 
 The magnet coils of the two armatures were calculated to 
 be in series with each other, but in shunt with respect to the 
 two armatures. This variation from present street-car prac- 
 tice was thought to be justified by the following considerations : 
 
 First, no system of commutation could obviate the need 
 of a considerable external resistance, which could be easily 
 made sufficient within itself for speed control; second, the 
 maximum torque for a given armature current is always 
 desirable and can best be obtained by permanent saturation ; 
 third, this constant and maximum magnetization would tend 
 to diminish sparking an evil to be specially avoided in long- 
 distance work, involving continuous runs of several hours. 
 
 While it was thus intended to use shunt coils at first, a 
 little experience would soon demonstrate whether the advan- 
 tage lay with this or the series winding. 
 
 The commutator, Fig. 124, was designed to be 23 inches in 
 diameter, which would of course give rise to a very high 
 speed relative to the brushes. It was thought better to face 
 the mechanical trouble that might be connected with great 
 circumferential speed rather than to diminish the diameter 
 "by decrease of sections, since this in turn might produce 
 sparking. 
 
 Without going into the mechanical details of the commu- 
 tator, we may add that a single bar could be taken out or 
 replaced without removal from the shaft. 
 
 The rheostat for the armature circuit was designed to be 
 of iron wire, with the expectation, however, of experimenting 
 also with liquid or carbon resistance. The resistance and 
 heat capacity of this main line rheostat were determined with 
 reference, first, to the current necessary for starting ; second, 
 to currents needed for maintaining low speeds. 
 
 A small rheostat and a condenser were designed to be 
 placed in the circuit of the magnet coils, and a rheostat also 
 in the lamp circuit, these resistances giving delicate control. 
 
 The problem of retardation for a mass of, say, 40 tons run-
 
 202 THE ELECTRIC RAILWAY. 
 
 ning at 1 50 miles per hour is a serious one. The Gallon and 
 Westinghouse tests indicate a very low value for the co-effi- 
 cient of friction between brake-shoe and wheel at high 
 speeds ; they report it in some cases at 60 miles per hour as low 
 as 0.04. As, however, it rapidly increases at lower speeds, 
 an average of o. i may be taken. The total retarding effort, 
 i.e.', brake resistance and track and atmospheric resistance, 
 must be kept below that which will just produce sliding of 
 the wheels, and the co-efficient of sliding may be taken on an 
 average at 0.08 of the weight on the wheels. If, then, we 
 represent by P the maximum allowable brake pressure on 
 the wheels, by R resistances other than brake friction, and 
 by W the weight on braked wheels, we may wr.ite the follow- 
 ing inequality : o.i P R < 0.08 W. 
 
 The value of P should be determined as that which, with 
 a small margin, will satisfy this condition of inequality. 
 
 Following this calculation, it was found that a brake press- 
 ure of about 5,000 pounds should be applied to each wheel. 
 This was designed to be produced by magnetic brakes similar 
 to those used by Mr. Daft on the New York Elevated Rail- 
 way. The form and approximate dimension of those brakes 
 are shown in Fig. 125. 
 
 A mass of 80,000 pounds moving at 1 50 miles per hour rep- 
 resents 64,000,000 foot-pounds of energy. Average resistance 
 due to brakes = o. i X 60,000 6,000 pounds. Average re- 
 sistance from track and atmosphere for average speed of 75 
 miles = 400 pounds. 
 
 In addition, the motors may be required to generate a cur- 
 rent which shall be wasted over the rheostat. Since the 
 field circuit in the case of shunt motors is independent, mag- 
 netization may remain constant, and the average E. M. F. of 
 i,500-volt motors (i.e., motors generating 1,500 volts E. M. F. 
 at 1,100 revolutions per minute) would be, while the train 
 was coming to rest, 750 volts. Make the external resistance 
 such that, the two armatures being in series, an average cur- 
 rent of 200 amperes shall flow. Then the average dynamo 
 resistance, in pounds =2,000. Total retarding effort = 
 6,000 -f- 400 -J- 2,000 = 8,400 pounds. Dividing 64,000,000 
 by this last quantity, we have 7,620 feet as the length of
 
 HIGH-SPEED SERVICE. 
 
 303 
 
 run in coming to rest, and the time of stopping about 100 
 seconds. 
 
 For the mechanical construction of the locomotive two 
 plans were contemplated. One had a 12 -foot rigid wheel 
 base and no pilot wheels; the other had a 7-foot wheel base 
 for the drivers and a pony axle in front, free to move later- 
 
 HALFPLAN 
 
 LONGITUDINAL SECTION 
 FIG. 125. ARRANGEMENT OF ELECTRIC BRAKE FOR HIGH-SPEED LOCOMOTIVE. 
 
 ally over a certain distance, dragging the drivers in the same 
 direction. This is the general principle of pilot wheels, so 
 largely used on high-speed engines. The second arrange- 
 ment is preferred, for in it nearly all the weight goes on the 
 drivers.
 
 304 THE ELECTRIC RAILWAY. 
 
 In the first design of locomotive car the operator was to sit 
 between the two motors, where also would be placed the 
 controlling devices. In the second design the operator would 
 be placed over the pony axle, the devices being chiefly in 
 the cylindrical or parabolic head. Detailed drawings of 
 freight and passenger cars were completed, but need not 
 here be referred to. 
 
 As to the supply of current, it was proposed to use the 
 double metallic system. This departure from the ground 
 return, now so commonly employed in street-railway work, 
 seems justified by the following considerations : 
 
 i st. For a given voltage it diminishes the cost of insula- 
 tion, per se. 
 
 2d. It diminishes the danger of accident within the motors. 
 
 3d. It diminishes danger of accident to workmen. 
 
 4th. By being thus more readily handled it is probable 
 that a higher voltage will be found practicable than with the 
 ground return; hence, in fact, even first cost might not be 
 greater. 
 
 As designed, the conductors were to be placed in an in- 
 verted wooden trough, attached to posts placed at 1 2-foot 
 intervals on each side of the track. The trough was to be 
 about 5.5 feet above the ground. The conductors for each 
 side of the circuit were to be two one insulated and contin- 
 uous, the other a bare, flat strip broken into sections, these 
 being normally out of circuit ; the locomotive trolley arm 
 operating a switch to throw out the rear section and throw 
 into circuit the section ahead. Such a switch was designed 
 in detail. 
 
 As to what the line voltage should be, progress in the art 
 of insulation can alone determine. It would, of course, not 
 be necessary on a roadway for long-distance travel to consider 
 the death-producing voltage as a limitation. 
 
 In the important matter of limiting curvatures calculations 
 were made in the ordinary way, using particular values for 
 weight and height of center of gravity. A safe speed was 
 then tabulated, being just one-half the speed that causes the 
 resultant of weight and centrifugal force to pass through the 
 outer rail of a curve. The beneficial effect of super-elevation,
 
 HIGH-SPEED SERVICE. 305 
 
 neglected in the calculation, would, of course, practically 
 much increase the factor of safety. 
 
 The following table resulted from the method described: 
 
 Safe speed, by rule above, 
 Radius in feet. in miles per hour. 
 
 1,000 71 
 
 2,000 98 
 
 3,OOO I 20 
 
 4,000 140 
 
 5,000 158 
 
 6,000 172 
 
 7,ooo 187 
 
 8, ooo 198 
 
 Of course, naught save practical trial could determine 
 whether the assumed factor of safety, which is quite large 
 enough at present speeds, would in fact be right for the 
 higher speeds aimed at. 
 
 COMMERCIAL ASPECTS. 
 
 Copper calculations were made, based on a station potential 
 of 3,000 and also of 6,000 volts, stations being 50 miles apart 
 (thus supplying current 25 miles on each side), and on a ser- 
 vice requiring one train on each 2 5 -mile section, the drop on 
 the line being 33 per cent, of station potential. Then, for a 
 line of i ,000 miles, as from New York to Chicago, with copper 
 at 15 cents per pound, and for 3,000 volts initial pressure: 
 
 Per mile. 
 
 Cost of conducting system, including frame-work = $7,000 
 
 station per mile, including steam power = 3,000 
 
 double track, including depots, etc = 55,000 
 
 train equipment, 3 cars each, 50 trains and engines.. . . = 1,000 
 
 Total per mile $66,000 
 
 In this, the cost of track construction is based on road-bed 
 figures reported for the Erie Railway Company. To arrive 
 at operating expenses, suppose 20 trains each way each day 
 and a schedule speed of 125 miles per hour, i.e., 8 hours for 
 the trip. Suppose a station development of 800 horse-power 
 per train on the line, or 6,400 horse-power hours per trip. 
 In stations of large capacity working under the supposed 
 conditions the cost of one horse-power hour = 0.90 cent, 
 
 20
 
 3 o6 THE ELECTRIC RAILWAY. 
 
 or, per trip, for power, a cost of $57.60. For trainmen, let 
 there be two men on each train and one per train in reserve ; 
 and suppose $3 per man per trip of 8 hours, then, cost 
 per train trip = 3 X 3 = %> Interest charge per train trip, 
 $66,000,000 X 0.05 -r- 365 X 40 = $225.42. For maintenance 
 of way take $i ,000 per mile of double track per annum, or per 
 day for the supposed line $2,700, or per trip, 2,700 -h 40 = 
 $67.50. For general expense of operating company, sup- 
 pose $1,000 per day, or per train trip $25. For wear and 
 tear, oil, etc., on train itself, take $10 per trip. Then total 
 = 56.60 -f 9 + 225.42 + 67.50 -f 25 + 10 = $394-42. 
 For receipts we should have about the following : 
 Suppose the average train be of two cars (such as designed) t 
 having each a capacity of about 10,000 pounds freight (con- 
 sidered to be such as express and mail matter), or, say, of 15 
 passengers. Suppose average train load of freight 15,000 
 pounds, of passengers 20. Take freight at 33 cents per cwt., 
 passengers at $25. The income per trip in either case 
 $500, showing profit over fixed charges of, say, $100 per 
 trip, or $4,000 per day. The total daily load of freight, 
 each way, required to justify the service above treated 
 is 300,000 pounds, or if 200 through passengers be car- 
 ried (i.e., 10 trains for passengers each way each day), then 
 only 150,000 pounds of freight. The present movement 
 between New York and Chicago, in mail and express, is not 
 far from this figure. A soo-mile line connecting Boston, 
 New York, Philadelphia, Baltimore and Washington would 
 be even more profitable. It is to be remembered also that 
 a service of small trains is contemplated, making fixed charges 
 relatively very large. Much larger trains could be, and, we 
 think, will be, run at the speeds here considered; but we 
 believe that the smaller effort would precede the larger. It 
 is further to be noted that the use of 6,000 volts potential 
 and the extension of some existing roadway instead of the 
 building of a new one would diminish first cost. As grade 
 crossings would be out of the question, and as cities would 
 require to be entered above or below the surface (country 
 roads could be carried over on bridges) , it is perhaps best to 
 consider only the larger figure already used.
 
 HIGH-SPEED SERVICE. 307 
 
 To demonstrate all, or nearly all, that has been outlined 
 liere would cost about $300,000, covering the construction and 
 operation for a reasonable time of the 4-mile circular line 
 proposed above. Last year the Baltimore company met un- 
 expected difficulty in efforts to raise the needed money, and 
 their operations were suspended just at the time when plans 
 were completed even to working drawings. 
 
 Unfamiliar as is the project, the reader will perhaps be the 
 more ready to share the confidence of its promoters on read- 
 ing the words from Prof. Henry A. Rowland and Dr. Louis 
 Duncan, of Johns Hopkins University, to whom were sub- 
 mitted full details of the plans partially described in this 
 paper. After viewing the figures for power required, they 
 say: 
 
 " We believe from the data obtained that the values given 
 are not too low, and that the horse-power which Mr. Crosby 
 calculates is not less than the amount required. While we have 
 investigated carefully and considered all the data obtainable, 
 yet the existing experiments are not sufficient to accurately 
 fix a limit to the train resistance. We believe, however, that 
 the value assumed by Mr. Crosby is safe. 
 
 " The motors, the calculations and drawings for which are 
 in the appended statement, will develop horse-power for 
 which they are designed, namely, 500 horse-power. We 
 point out some modifications which will be beneficial. We 
 Relieve that they will drive the train [locomotive and two cars 
 O. T. C.] at the required speed [speed then considered being 
 120 miles per hour on a level]. 
 
 " The possibility of a train being derailed by an obstruction 
 on the track increases with the speed. At speeds up to 90 
 miles, however, there seems no increase in the number of 
 derailments. In the case in question the center of gravity 
 of the cars is very low, and it would be difficult to derail them 
 on straight parts of the track. The radius of the curves should 
 of course be great, but not so great as would be required for 
 an ordinary train going at these high speeds. The question 
 of safety is, however, almost wlwlly a question of track construc- 
 tion. Considering the form of the proposed train, its compar- 
 atively light weight making a less demand on the track, it is
 
 308 THE ELECTRIC RAILWAY. 
 
 certain that with a carefully constructed road it could attain 
 with safety speeds which would be impossible with trains as 
 at present constructed. As these latter have several times 
 made 86 miles, and often made 80 miles, it would seem that a 
 speed of 120 miles or even more, with the electric cars, would not 
 be outside the limits of safety. 
 
 " The plan for supplying current to the motors is feasible. 
 
 " The design of the motors is discussed in the detailed 
 report. It is generally good. 
 
 " Should it be demonstrated by an actual test that passenger 
 trains can be run safely and economically at a speed over IOG 
 miles per hour from here to Chicago, the financial aspects 
 of the case would certainly be improved. We are of the opin- 
 ion that the chances are in favor of this being accomplished by the 
 present scheme" (Italics are by the present authors.) 
 
 These words, though guarded and accompanied by criticism 
 of the detail, are yet a substantial approval of what was sub- 
 mitted. 
 
 A recent paper by Mr. Carl Zipernowski, setting forth 
 plans (not yet well matured) for a single-car, 125-mile-per- 
 hour passenger service between Vienna and Buda-Pesth, 
 shows that the interest in and efforts toward the great step 
 soon to be made are not confined to the American people 
 usually more adventurous (though perhaps less thorough- 
 going) than their European cousins.* 
 
 * Since the first edition of this work a very notable contract has been entered into be- 
 tween the Baltimore and Ohio Railway Company and the General Electric Company. 
 This contract calls for the construction of three 8o-ton electric locomotives capable of 
 doing service as follows : To pull a i2oo-ton train at 15 miles per hour over an 6.8 per cent 
 grade, and to pull a soo-ton train over the same grade at a speed of 30 miles per hour. 
 The locomotives are to operate in a tunnel about three miles long, which the Baltimore 
 & Ohio Railway Company has built through and under the city of Baltimore, Md. 
 
 The locomotive will probably be constructed to permit of much higher speed with 
 lighter loads, those mentioned being maximum for the freight and passenger service, 
 respectively, which are contemplated as possible within the next ten years. It is ex- 
 pected that these locomotives, which will be driven by direct-coupled slow-speed genera- 
 tors, will be in operation in the summer of 1893. 
 
 The confidence thus shown by two large companies in the future of electric traction 
 on a large scale is most reassuring to those who follow the progress of the art.
 
 CHAPTER XI. 
 
 COMMERCIAL CONSIDERATIONS. 
 
 IN treating of the commercial value of any enterprise we 
 are led to discuss: 
 
 1. Original investment. 
 
 2. Cost of maintenance and operation. 
 
 3. Gross income. 
 
 Of the original investment required for a street railway the 
 two most important items are those to which it is most diffi- 
 cult to assign average values that may serve as useful guides. 
 These two items are cost of franchise and cost of track con- 
 struction. It may be said that the franchise may cost 
 from $0.00 to $500,000 per mile of roadway. Track con- 
 struction may run from $5,000 to $50,000 per mile of single 
 track. 
 
 It should be added, however, that variation in this item is 
 often due to the fact that it is made to bear a part of the 
 charge for franchise, imposed in the shape of restrictions as 
 to the how or where of performing the necessary engineering 
 work. In treating the matter here we shall omit considera- 
 tion of the franchise cost, and give moderate values for track 
 construction, not applicable where special engineering diffi- 
 culties are to be met. 
 
 The minimum figure above given $5,000 per mile of 
 single track cannot often be reached. It supposes a 45- 
 pound T-rail, cross-tie work, with very light grading and no 
 paving. 
 
 In the general summary of cost intended here to be pre- 
 sented we shall take $10,000 per mile of single track as a 
 safe average figure. This estimate, however, is not designed 
 to cover the expense required to make of the track a good 
 conductor. Let us suppose for this purpose a No. o bare 
 copper wire to be laid and bonded to each rail, as described in 
 
 309
 
 3IO THE ELECTRIC RAILWAY. 
 
 " Instructions to Linemen" (see Appendix B). For this item 
 it is safe to estimate $200, covering labor, and $400 for ma- 
 terial. 
 
 Going next to the poles, a fair figure for round, trimmed 
 cedar poles, laid down, not set, may be taken at $2.50. 
 Squared poles of Georgia pine maybe had for $3.50; and 
 in either case setting, if without concrete or blasting, need 
 not cost more than 2 to $2.50. For a good sawed pole, 
 with one cross-arm, it is safe to say that $600 per mile 
 will cover the cost, the cross-suspension method being sup- 
 posed. (As the poles are set about 125 feet apart, longitu- 
 dinally, and one on each side of the street, we count on 
 about 90 poles per mile.) 
 
 Iron poles may be had for from $18 to $27 each. The 
 pole of lower cost will stand ordinary strains. At curves 
 or other special points something heavier may be required. 
 The standard round pole, in three sections, of 6, 5, and 4 inch 
 diameter, respectively, may be had for about $22. Setting 
 of these poles in concrete, without blasting, should not cost 
 more than $5 each. Blasting alone may cost from $10 to 
 $20 per pole. For a mile of iron poles (90), set, we take 
 $2,500 as a fair figure. 
 
 They will last longer than wood, look somewhat better 
 than the best sawed pole, but are less desirable in one re- 
 spect, namely, that the liability to "grounds" is somewhat 
 greater than with wooden poles. 
 
 The bare trolley wire will cost from $175 to $285 per mile, 
 according to size the latter figure covering No. o B. & 
 S.. at present market rates. Cost of span wire, insulators, 
 etc., maybe taken at about $100 per mile; labor of erect- 
 ing wires, superintendence of contingencies, at about $325 
 per mile. Taking $275 for the cost of the bare trolley 
 wire, we have $700 per mile, covering the trolley wire in 
 place. 
 
 The amount of feed wire, of course, is indefinite, as has 
 been explained in Chapter IV. 
 
 The following costs per mile of a good insulated wire show 
 how the amount may be made up, having determined the 
 length and cross-section of feeders:
 
 COMMERCIAL CONSIDERATIONS. 311 
 
 No. o B. & S. per mile $500 
 
 No. oo B. & vS. per mile 650 
 
 No. ooo B. & S. per mile 850 
 
 No. oooo B. & S. per mile 975 
 
 The cost of erecting this wire runs from $75 to $100 per 
 mile. As a safe general figure we may take $1,000, cover- 
 ing feeder wire in place. Resuming, we have: 
 
 Track construction $10,000 
 
 Earth circuit 600 
 
 Wooden poles 600 
 
 Trolley wire, etc 700 
 
 Feed wires. . 1,000 
 
 Total, track and line, per mile $12,900 
 
 If iron poles be used, add $2,000, or, in round figures, 
 say $13,000 with wooden poles, $15,000 with iron poles. 
 
 Car bodies, ready for motors, may be had for from $750 to 
 $1,500, the higher figure referring to those of unusual con- 
 struction. A very good body may be had for $1,000, for 
 a 1 6-foot car. The longer bodies cost from $1,250 to 
 $2,000 each. Trucks for these cost about $600. Electri- 
 cal equipment, consisting of two 15 -horse-power motors 
 and accessories, may be taken at from $1,800 to $2,500, 
 according to the make. If one motor be used, from $1,100 
 to $1,800. Assuming $2,250 for the electrical equipment, 
 the car ready to operate costs $3,500. (All the figures 
 here and elsewhere used suppose goods are purchased for 
 cash. No treatment can here be given to the increase of 
 figures due to various forms of credit.) 
 
 Passing now to the station, it is to be noted that the cost 
 per horse-power of the steam and electric plant varies con- 
 siderably with the size of the installation. For convenience 
 here, let us suppose the steam and the electric plant to be 
 each of 500 horse-power capacity. We may then assume, 
 per horse-power installed, for a complete high-speed steam 
 plant, including boilers and fittings, $50* for a complete 
 
 * For a low-speed steam plant *hese figures should be somewhat increased say 20 per 
 cent.
 
 2J2 THE ELECTRIC RAILWAY. 
 
 electric plant, $40, or $90 per horse-power for the combi- 
 nation. It has been shown that there should be provided 
 from 10 to 20 horse-power at the station for each car on the 
 line. Taking 15 horse-power per car, it appears that an 
 investment of $1,350 must be made in station machinery per 
 car to be operated. 
 
 The cost of ground space for station, car-barn and office 
 will, of course, vary widely. The cost of buildings will also 
 vary considerably, there being in this item a considerable 
 "personal equation" of the particular management. Since 
 we desire to present some reasonable figure, we find that 
 $50 per horse-power of station capacity may serve in a 
 majority of cases to cover all real estate. Reduced to the 
 investment per car to be operated, this becomes $750. If, 
 now, there were any fixed number of cars per mile of track 
 or mile of street, the whole investment might be expressed 
 at a certain amount per car operated. This ratio, however, 
 varies widely being principally determined by the size of 
 the city in question. 
 
 It may be roughly stated that cities of population from 
 15,000 to 100,000 show from one to two cars per mile of track, 
 while from 100,000 to 300,000 inhabitants there should be 
 from two to four cars per mile. 
 
 So, in still larger cities, a slightly higher ratio. * 
 
 Let us take two cases first, that in which we find one car 
 per mile of track. The investment per 1 6-foot car then 
 becomes: 
 
 One mile track and line, wooden poles $13,000 
 
 One car ready to run 3, 500 
 
 Steam and electric plant per car i,35o 
 
 Real estate. . . . 700 
 
 Total, per car $18,550 
 
 Then, second, for three cars per mile, iron poles on line, 
 the items are as above, except track and line, which now 
 
 * The West End Company of Boston serves about 650,000 people. It has 260 miles of 
 track and 2,745 cars. New York City has 276 miles of track and 2,519 cars.
 
 COMMERCIAL CONSIDERATIONS. 313 
 
 enter at $5,000 (one-third of $15,000) per car, or the total 
 per car $10,550. 
 
 We may now find the interest charge per car-mile, which 
 is a convenient perhaps the most convenient unit of oper- 
 ating- expense and earning capacity. 
 
 Each car may be taken to make 35,000 miles per year a 
 little less than a steady average of 100 miles per day, for all 
 cars. 
 
 At 5 per cent, on $18,550 we have per annum. . $927.50 
 or, per car-mile, 2.65 cents. . 
 
 At 5 per cent, on $10,550 we have per annum. . 527.50 
 or, per car mile, 1.5 cents. 
 
 Thus far the reports of operating expenses of electric rail- 
 ways 'have been widely divergent. Some order is now 
 beginning to appear. It should be recognized, however, 
 that considerable differences as between one company and 
 another must continue to exist until local prices of labor and 
 material shall be everywhere uniform. We have seen many 
 useless contentions as to the proper average figure for oper- 
 ating expenses, one contestant having in mind labor of 
 motor-men and conductors at 22 cents per hour, as in Boston, 
 Mass., the other considering the same item at 11 cents per 
 hour, as in Knoxville, Tenn., and very many other cities of 
 comparatively small population. 
 
 So, the one has in mind coal at about $5 per ton, as in 
 Boston, the other coal at about $1.75, as in Knoxville. 
 
 Again, one may consider a road of 5 or 6 cars which runs 
 its own independent station at a necessarily high figure per 
 car, while another refers to a similar small road that may 
 have its dynamos in an electric-light station, an equitable 
 arrangement being made for the additional coal and wear 
 and tear the additional attendance and general expense 
 being almost nil. 
 
 Let us take up the various parts of operating expense, 
 seriatim. 
 
 As to the cost of getting upon the line the electric power 
 needed for one car -mile of ivork, it could be, in a station of 
 about 500 horse-power capacity, not far from i.o to 1.2 cents, 
 as shown by Table III., Chapter X. (i.e., table of cost of one
 
 3I4 THE ELECTRIC RAILWAY. 
 
 horse-power hour in various stations). The cost per horse- 
 power hour is a little more than this figure, since one horse- 
 power hour will produce a little more than one car-mile of 
 service, say 1.2 miles. For long cars the figures should be 
 increased by at least 50 per cent. ; we would then have a 
 station cost of 1.5 to 1.8 cents. 
 
 These figures, too, may hold for -a standard 1 6-foot car 
 drawing a trailer. In the figures here given, taken from the 
 table, coal enters at 0.475 cent P er horse-power hour. It is 
 assumed to cost $3 per ton and to be burned at the rate of 
 3.2 pounds per horse-power hour. 
 
 As a matter of fact, the consumption per horse-power hour 
 generally goes far beyond this. Bad design of station, as in 
 wrong relation between capacity and service of engines, poor 
 piping, etc., together with poor firing, may double, even 
 treble, this figure. We must, therefore, be understood as 
 speaking of what can be done by good practice, not of what is 
 done by bad practice. The effect upon the cost per car-mile, 
 should coal consumption be trebled, would be to add about 
 0.8 cent, making nearer to 2.0 than i.o as the round figure. 
 We have seen reports of less actual cost than that resulting 
 from the table and mentioned above. But in a majority of 
 cases actual practice is worse than the tabular figure. We have 
 therefore averaged several cases of practice generally called 
 good, and find 1.35 as a fair figure. Many statistics show 
 higher. * 
 
 The difference between maximum and minimum values is 
 striking. But it should be noticed that, in the average, roads 
 operating 3 cars are grouped with those operating 140. In 
 many respects, but especially in the cost of power, must there 
 be very wide differences between very small and very large 
 plants. It scarcely seems useful to average results obtained 
 
 * Mr. J. S. Badger (Tiie Electrical World, October 30, 1891) reports as the average of 
 twenty-two trolley roads a cost of 2.32 cents for this item. It is made up as follows. 
 Maintenance of power plant, repairs on engines, dynamos, etc. 
 Highest. Lowest. Average. 
 
 -86 0.05 0.36 
 
 Fuel, wages, oil, waste, etc., at station. 
 Highest. Lowest. Average. 
 
 4-95 0.48 i. Q 6' 
 
 These two items should be summed to get cost of power in the line, making 0.36 -f 1.96 = 

 
 COMMERCIAL CONSIDERATIONS. 315 
 
 under such widely different conditions. The figures given 
 in the text are not expected to apply to very small plants. 
 They will be useful as applied to companies operating, say, 10 
 cars or more. See curves (page 286) showing change in cost 
 of one horse-power hour according to change in magnitude of 
 station. The change up to about 500 horse-power is very 
 rapid. 
 
 As an example of good station service the East Harrisburg 
 Street Railway Company may be cited. They kindly report 
 to us as follows: 
 
 POWER PLANT EXPENSE FOR MONTH SEPTEMBER, 189!. 
 
 Coal at $2 . i o per ton $24 1.12 
 
 Dynamo oil . 5.55 
 
 Cylinder oil 5.80 
 
 Waste and coal oil 3.81 
 
 Engineer and fireman 206.63 
 
 Sundries 21.24 
 
 Total $484. 1 5 
 
 Number of cars, 18; mileage, 63,028. Cost per car-mile, 
 0.75 cent. 
 
 Many contracts heretofore made between lighting com- 
 panies and railway companies have been on a car-day basis, 
 with some expressed or implied assumption concerning car- 
 miles per day. Unless the parties to such contracts are sat- 
 isfied to let matters rest on a guesswork basis, it is certainly 
 best to at least come as near to the facts as possible by the 
 car-mile unit. The most accurate guide for proper charge is 
 to be found in watt readings taken at the station, all the 
 peculiarities of service (pro and cou the railway company) 
 being thus taken into account. 
 
 When it is remembered that occasional line leaks have been 
 found in one case reaching a flow of 60 amperes the wis- 
 dom of this course will be better appreciated. True, there 
 are as yet no satisfactory large wattmeters for railway sta- 
 tions: but several small ones may be used in multiple, or 
 frequent reading of the ammeters (checked by frequent
 
 316 THE ELECTRIC RAILWAY. 
 
 reading of the voltmeter) will generally give a closer deter- 
 mination than can be made by estimating the consumption 
 per car-day, or even per car-mile. 
 
 In case a contract of this sort is about to be made we would 
 advise the following course : Agree upon a rate per horse- 
 power hour, a minimum of consumption being fixed, or on a 
 sliding scale with respect to amount of power used ; have 
 five-minute readings on the ammeter, and fifteen-minute 
 readings on the voltmeter taken during a period of, say, four 
 weeks ; keep a close record of the car-miles made during the 
 same period. 
 
 In this way a very reliable figure for power per car-mile 
 can be had, and the frequent readings may be discontinued 
 to be taken up again when the season changes, as from sum- 
 mer to winter, and to be occasionally renewed as a check on 
 any changes in service. As soon as reliable recording watt- 
 meters can be had for all stations there need be no more 
 uncertainty in a contract of this sort than is the case in a 
 contract for the supply of gas to a building. 
 
 Some existing contracts call for a certain sum per day for 
 each generator of a given size that may be in operation. 
 This method is of course less accurate than the others. A 
 just relation between service and cost may be hit upon and it 
 may not. 
 
 For a small number of cars, say six, it is not uneconomical 
 to pay 4 cents per car-mile rather than erect a suitable sta- 
 tion. There will not be a great profit in this to the lighting 
 company unless they have already all-day attendance as a 
 part of the lighting service. Four cents per horse-power 
 hour or per car-mile leaves a handsome percentage of profit 
 to the power station, and yet the railway company can scarcely 
 compete with that figure by their independent production on 
 a very small scale. 
 
 In the matter of motor repairs the advances of the current 
 year will make a great and welcome change. There has 
 been a wide variation in the experiences of different com- 
 panies; averages have been less useful guides than- will now 
 be the case. Perhaps the widest fluctuation of "luck" has 
 been found in connection with armature repairs ; next would
 
 COMMERCIAL CONSIDERATIONS. 317 
 
 follow field windings. It is safe to say that ten roads might 
 be chosen from which, selecting two sets of five each, the 
 average armature and field repair account for the one set 
 would be five times that of the other. Individual instances 
 may be found in which one road has had five times as much 
 intelligence in its management as a. neighboring road, and 
 may have had armatures made with five times or with one- 
 fifth of the care given to those that chanced to be received 
 by that neighbor. 
 
 The winding of electrical apparatus is largely a matter of 
 conscience as well as skill. Constant supervision cannot be 
 given to every workman, yet a moment of negligence or ill- 
 will on his part may fatally weaken an armature. Shop- 
 testing is important should never be omitted ; yet it cannot 
 bring out all weaknesses as they are brought out by bad 
 handling on the road, or even by the exigencies of legitimate 
 service. In all this matter the most marked improvement 
 recently made is the use of Gramme or hollow-cylinder arma- 
 tures instead of Siemens or solid-core armatures. 
 
 Field windings have been much improved by good detail 
 work and by decreasing the heat developed in them, or 
 increasing the rate of dissipation of that heat. In gears, 
 too, there has been a great change for the better. The 
 number has been diminished, the speed reduced, and those 
 that remain have been carefully inclosed in dust-proof cas- 
 ings. These, besides excluding dust, pebbles, etc., serve also 
 as reservoirs for oil or grease, into which the larger gear 
 (axle gear) dips half an inch or more, thus lubricating itself 
 and the armature pinion. Experience with these improved 
 forms has not been extended enough to show accurately what 
 their repair rate is ; but that it will be greatly less than that 
 of the earlier high-speed, exposed motors, we have priina 
 facie evidence of the strongest character, and also the grati- 
 fying results of some months of practical use for some hun- 
 dreds of such motors as are here described. 
 
 The following table, based on the returns of several 
 millions of car-miles of hard work under good management, 
 shows the life of various wearing parts in double-reduction 
 motors. These machines were covered on the side, rather
 
 2!g THE ELECTRIC RAILWAY. 
 
 loosely, by canvas, and under the bottom a metal pan was 
 hung. Dust was not excluded, but the protection from peb- 
 bles, etc., was good. 
 
 Life in Firct rr>=f Cost per car- 
 
 Article, car-miles. First cost. mi i e in cents. 
 
 Axle gear 29,600 $7.50 0.025 
 
 Intermediate gear 28,600 6.25 0.022 
 
 Intermediate pinion 10,050 6.75 0.067 
 
 Armature pinion 8,260 6.05 0.073 
 
 Trolley wheel 5-75 i-4 0-025 
 
 Armature bearing 24,200 3.70 0.015 
 
 Intermediate bearing 35, 100 4.40 0.013 
 
 Total 0.240 
 
 It may be said that many roads have shown a repair account 
 for motors, gears and trolleys running from 2 to 3 cents per 
 car-mile some even higher. We can see no reason why, 
 with slow-speed machinery, this item need go beyond i cent. 
 It can be shown to be considerably less than this for the 
 period during which the single-reduction motors have been 
 in use. 
 
 Maintenance of the line is well covered by 5 per cent, on 
 its cost. Averaging this in iron and wooden pole construc- 
 tion, at $3,000 per car for line investment and at 35,000 
 car-miles per year per car, this item amounts to 0.4 cent 
 per car-mile. At $1,000 line investment per car it becomes 
 0.14 cent. It is an interesting fact that on curves, where 
 traffic is very dense, the sides of the trolley wire are worn by 
 the trolley flanges so as to require renewal of the wire in a 
 very few years. 
 
 Maintenance of dynamos and engines has already been 
 accounted for in taking the cost of power from the table. 
 
 Maintenance of track for electric railways is still a some- 
 what uncertain factor. Typical, well-constructed electric 
 railway road-beds are not numerous; use of them has not 
 been long; reports have not been full; paving expenses 
 (varying widely according to municipal requirements) have 
 been combined with those of the track proper. 
 
 It is found that OQ steam railways the injury to track is 
 approximately in proportion to the number of locomotive 
 miles, rather than total ton-miles, including train weights. 
 In other words, each passage of a set of driving wheels does
 
 COMMERCIAL CONSIDERATIONS. 319 
 
 a certain amount of injury, and this amount is great in com- 
 parison to that done by the car wheels following. In street 
 railway practice car-miles and driving-wheel miles are almost 
 always as i to 2 (or i to 4), since, except for trailer service 
 (rarely exceeding two trailers) , locomotive and carrying car 
 are combined. The West End Street Railway Company of 
 Boston, Mass., operating about 325 electric cars over tracks 
 not yet completely relaid for electric services, reports the 
 track maintenance at i.oS cents per car-mile. The same fig- 
 ure is reported by the Pleasant Valley Railway Company, of 
 Pittsburg, Pa.* Until more extended figures can be obtained 
 we may adopt these as a fair guide. No distinction is here 
 made between effect of long cars and sho'rt cars, though the 
 former are in considerable number on the West End lines. 
 
 Maintenance of car bodies and trucks is thoroughly covered 
 by allowing 20 per cent, per annum on their cost, i.e., $250, 
 or 0.72 cent per car-mile. 
 
 We have already stated that in the matter of wages of con- 
 ductors and motor-men wide differences are to be found. 
 The pay per hour is known to vary from 10 cents to 22 cents. 
 Let us take 18 cents as an average. We then have 36 cents 
 for the two men. They will run very nearly 8 miles per 
 hour while under pay , hence the cost per car-mile is 4. 5 cents. 
 
 On a few electric roads, in rather small cities, no conductors 
 are employed. Even with horse cars, however, it has been 
 found of advantage to have conductors in nearly all towns of 
 considerable size. Of cities in the United States having 
 more than 100,000 inhabitants we now recall only one, New 
 Orleans (population 242,000), in which conductors are not 
 employed. The higher speed made, the greater number of 
 passengers carried, and the occasional attention required for 
 the trolley are all considerations which give greater value to 
 the conductor on an electric than on a horse car. 
 
 * Rather curiously, just half of this amount is reported by Mr. J. S Badger ( The Electrical 
 World, Oct. 30, 1891) as the average given by twenty-two trolley roads for maintenance of 
 way. The highest figure given in the averages is 1.86 cents, the lowest o. 10 cent. Prob- 
 ably reconstruction of old track will in part explain the very high figure; use of perfectly 
 new track for a short time probably explains the very low figure. Mr. Badger's report 
 appeared after the text above was written, and is a valuable contribution to our knowl- 
 edge of the actual operating expense. In this particular item, however, we think it safer 
 to adhere to the larger figure, 1.08, reported by companies operating a considerable 
 mileage for a consi.'lc-rablo *-ime.
 
 220 THE ELECTRIC RAILWAY. 
 
 The charge for accidents is another widely varying item 
 On the same road it varies much from month to month, or 
 from season to season. From an examination of a number 
 of cases we find that 0.25 cent per car-mile is not far from a 
 fair average. Insurance companies have recently placed 
 accident risks for a number of electric railway companies, 
 generally for a fixed percentage of gross receipts. This per- 
 centage has varied from 0.75 of i per cent, to 2.25 per cent., 
 the higher figure in cities having a bad accident record, due 
 to unusually crowded streets. To show the relation existing 
 between the figure just assumed and those held in view by 
 the insurance companies a certain railway company has 
 gross receipts of about 20 cents per car-mile. Its rate is i 
 per cent, (very nearly). This means 0.20 cent per car-mile, 
 a little less than the figure assumed. Three months' average 
 April, May and June, 1891 for this item on the West 
 End Railway Company of Boston shows .57 cent per car-mile. 
 Mr. Badger, in the report referred to in the last foot-note, 
 takes 0.06 1 cent. This seems entirely too low. 
 
 General expense, covering officers' salaries, office expenses, 
 superintendent, taxes, insurance, etc., runs from i.o to 2.25 
 cents per car-mile. In making general estimates it is best 
 to take at least 2.0 cents. 
 
 Resuming, the various items for an average case appears 
 thus: 
 
 COST PER CAR-MILE (STANDARD 1 6-FOOT CAR). 
 
 Power delivered on line 1.35 cents. 
 
 Repairs on electric machinery of car i.oo " 
 
 Repairs on line 0.43 " 
 
 Conductors and motor-men 4. 50 " 
 
 Repairs on cars and trucks o. 72 " 
 
 Maintenance of roadway i .08 " 
 
 General expense 2.00 " 
 
 Accidents 0.25 " 
 
 Total 
 
 11.33 cents. 
 
 Throughout the above calculation good management is sup- 
 posed.
 
 COMMERCIAL CONSIDERATIONS. 321 
 
 An inspection of these items shows the great advantage to 
 be attained from running trail cars whenever their use is 
 justified by the traffic. The consumption of power for the 
 two cars will not exceed 50 per cent, of that required for the 
 motor car alone, since all losses of transmission are met in 
 the former. One active conductor will take care of both 
 cars. Maintenance of line, of track, and of car machinery 
 are scarcely increased at all. Nor does it appear that gen- 
 eral expense would be considerably increased. Assuming, 
 however, that this item must always be in proportion to the 
 volume of business done, we have still, on the above basis 
 for a motor car, not more than, say, 5 or 6 cents for the cost 
 Cx running a trail car-mile. 
 
 Comparing the cost of operating a two-car train (motor car 
 and one trailer) with that for one long double-truck car, it 
 would appear that they are very nearly the same. Service 
 based on the long car is, however, much less elastic in meet- 
 ing traffic demand than is that based on the use of trailers. 
 The additional expense over that for the standard car must 
 be met at all times, whether traffic be heavy or light. More 
 tim > is required to take on and let off passengers from the 
 long car than from the two shorter ones, or even from a 
 single standard car carrying the same load as the long car. 
 
 Let us see approximately what it costs to stop a car, without 
 now considering the loss of schedule speed (or the greater 
 consumption of current, if speed be maintained) due to in- 
 crease of time for stops. For convenience, take the even 
 figures 12 cents and 16 cents as the cost per car-mile for 
 short and long cars respectively. At 9 miles per hour, 
 schedule speed, we have the cost per hour of running equal to 
 1 08 and 144 cents respectively. This, per minute, equals 1.8 
 cents and 2.4 cents respectively. Suppose the average time 
 lost per stop, from full speed to full speed again, be 10 sec- 
 onds and 15 seconds, respectively, for short and for long cars. 
 These figures correspond very nearly to the observed facts 
 in several cases. Then for cost of a short-car stop we have 
 1.8 -7- 6 = 0.3, for a long-car stop 2.4 -=- 4 = 0.6 cent, or just 
 twice as great. 
 
 The advantage of the long car over the motor car and
 
 322 THE ELECTRIC RAILWAY. 
 
 trailer lies in its smoother motion over the track. How far 
 this will enter as an earning factor it is difficult to say. We 
 know that improvement, the refinement of service, generally 
 carries with it increase of travel. The pedestrian, seeing a 
 chance to get a seat, enters the car which, if already crowded, 
 he would not have entered. High speed, carpeted seats, 
 good lights, nice trimmings, general neatness, politeness of 
 employees all these increase the number of passengers. 
 And all these may be given, together with greater ease in 
 entry and exit, in the motor and trail car. Whether the 
 smooth riding of the double truck will attract additional pas- 
 sengers, only a long experience can determine.* At present 
 there is a reaction toward the trail car. 
 
 As to how far it is best to go in furnishing seating capacity,, 
 there are, indeed, certain conditions of heavy travel as when 
 every standard 1 6-foot car that can be run shows standing 
 crowds for which it is plain that either the long car or the 
 train must be applied. For traffic less than this, yet quite 
 heavy for the single car, and which might be increased by' 
 further comfort in riding, we can only say that each partic- 
 ular case must be determined by trial and good judgment. 
 
 Passing now generally to the question of earnings, we can. 
 do little more than refer to the facts as disclosed by reports. 
 It is evident from the figures above given how much must 
 be earned in order that the enterprise may be self-supporting. 
 We may say roughly that an electric car must earn 1 5 cents 
 gross per mile run in order to pay 5 per cent, on the invest- 
 ment no account being taken of cost of franchise. In large 
 cities, where labor is considerably higher, this figure must 
 be increased to 18 or 20 cents. 
 
 Since, in the United States, the fare is almost everywhere 
 the same 5 cents it follows that for every mile run the 
 company should receive from three to four passengers in 
 order to live. Expressed in passengers per car per day, as- 
 suming 1 10 miles per day as made by each car, the required 
 numbers are from 330 to 440. The relation between the 
 
 * The question of side entrances for these long cars is now considered as a serious one 
 by all the companies who have had any experience in their use. A recent double-decked 
 long car, manufactured by the Pullman Company, of Pullman, 111., shows both side and 
 end entrances. 

 
 COMMERCIAL CONSIDERATIONS. 
 
 323 
 
 population of a city and the total number of people who ride 
 in street cars is an interesting inquiry. It is also full of 
 irregularities. The "lay" of the city, the occupation and 
 habits of its inhabitants, the quality of the service given, are 
 all of great value in determining the number of fares to be 
 taken from a given number of people. This ratio is fre- 
 quently expressed in this way that the whole population is 
 carried so many times per year or per day. On pages 324 
 .and 325 we give a table setting forth important commercial 
 
 
 c 
 
 URVE 
 
 OF CO-EF^ICIE'NT OF 
 
 
 
 , I 
 
 
 
 
 
 FilC 
 IN C 
 
 ES P 
 TIES 
 
 ER C 
 OF T 
 
 VPITA 
 HE U 
 
 8. 
 
 AVE 
 
 RAGES OF SAM! 
 
 
 wrOH, 
 
 >- 
 
 
 
 
 
 
 
 
 
 
 ^t 
 
 ^a,,,o 
 
 M CHIC 
 
 GO PHI 
 
 AND 
 
 
 
 
 
 
 
 ^ 
 
 *^ 
 
 IITIES- 
 
 
 .000 T 
 
 
 . 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 I/ 
 
 
 
 
 
 
 .000 
 
 
 
 
 
 
 
 
 
 
 
 
 . 80 
 
 <s 
 
 LTl 
 
 4 CIT,( 
 
 S-POP. 
 
 88.000 
 
 TO 107 
 
 000 
 
 
 
 
 
 
 
 * 
 
 AVERAGE OF 9 CITIES-POP. 65 
 VERAGE^ OF.5 ClTIES-POP. 52. 
 
 000 TC 
 
 88. 000 
 3.000 
 
 
 
 
 
 
 
 
 DC 
 40 
 
 7 ' 
 
 ,8C,T,ES.PO 
 
 
 
 
 
 
 
 
 
 
 JAVE 
 
 IAGE O 
 
 9 TOW 
 
 IS-POP 
 
 3.500 
 
 18.0 
 
 W 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 x> 260 a 
 
 30 400 500 6 
 
 00 7 
 
 X) 800 900 1000 1100 1200 ' 
 
 
 POPULATION OF CITIES BY THOUSANDS 
 FIG. 126. 
 
 facts of the street-railway service in a number of cities of 
 the United States. 
 
 The " cars possible to be supported" column shows the num- 
 ber of cars that could make a living, assuming from the above 
 discussion that 400 passengers must be carried per car per 
 day. 
 
 The last column shows the actual number of cars reported 
 from these cities, including every car, running or not. 
 
 The increase of what may be called the riding co-efficient 
 with increase of total population appears plainly from the 
 table. This relation is displayed more clearly by the above 
 curve, Fig. 126.
 
 324 
 
 THE ELECTRIC RAILWAY 
 
 SW D U3d 
 
 i i| |*^s saa*ff*as'RR l ssfe6 > < -s^-s-^ss 
 
 
 rfxo 
 
 -oxoa* 
 
 3.ocn8M H 
 
 , rr 
 
 B^Pf &RjW8R & fiSff9W m?i ? ft 
 
 % 5 u "S 
 
 
 3JIPJ 
 
 
 7l?d.ofJd d 
 
 ?3; .pg S .-<,j ^-Rg.si'ggygKg-gg's." 8.^*55588 
 
 Pd 'luiV 
 
 |^o5,2n^; Nm ...^^^ N c. M m 2*,j 
 
 !- 
 
 im^9 M > 8 njin R iv 
 
 Xup jad 
 
 
 
 
 
 
 g'E 
 
 IM^fH^I sHt?fxHII-li? 1^1 Hsl 
 
 1" 
 
 H^a*^"- 
 
 - C 
 
 ^^g-fS.?^^^^??^^^^^^!???, R8?2.R<8<8. 
 
 
 
 *| 
 
 Ilil^sil RS| ?1> ? 8 ? ?li S>l Rl s *** | ^"- ::< s as 
 
 o 
 
 ^ r /: cr 
 
 ji 
 
 **-" 
 
 Jl ' 
 
 
 
 
 < 
 
 ^ WM fn m w M M MMro-wtH tt ro * M M S N H 
 
 
 S.1 i gl ^Hi S ?^^ s H iff ? n S5 1 H ? i SI 1 
 
 
 p||||ilf^^^^Kss it g;;5??;Ci$R^^ss;ss^ 
 
 1 
 
 ^^|^>j^^^^^^^^|^^ 
 
 u 
 
 g g|1 pl||f| 1 11 1= >-i g |.s glU j l||lf 1 1 | 
 
 aaSaMasS^A^K^JsSiHsSj^enSs^JSlofeKtljSjW
 
 COMMERCIAL CONSIDERATIONS. 
 
 32$ 
 
 uad 
 
 .>*0c ^ r^o 
 
 i 
 
 
 
 
 | 
 
 
 
 
 *c 
 
 
 
 * 
 
 
 
 
 E 
 
 
 
 
 5 
 
 
 
 t 
 
 
 
 
 2 
 
 
 
 
 
 
 S-rB") XO'l 
 
 
 
 
 
 
 l( _ | 
 
 
 
 
 S 
 
 
 
 ts. 
 
 
 
 
 3 
 
 
 
 
 c 
 
 
 
 
 
 
 
 i 
 
 
 
 
 ^ 
 
 o 
 
 
 o*>*>^ os N ooooo 
 
 s 
 
 
 
 
 B 
 
 jj 
 
 
 
 n 
 
 S0 
 
 
 ^ 
 
 
 
 
 | 
 
 | 
 
 
 i 
 
 | 
 
 *j 
 
 c gj "8 
 
 
 
 
 
 
 | 
 
 | 
 
 
 a 
 
 J 
 
 - s -~ S u t! 
 
 
 1 
 
 
 
 
 I 
 
 
 
 r 
 
 JB 
 .H 
 
 oj|| | 
 
 
 
 
 
 
 1 
 
 I 
 
 
 | 
 
 X 
 
 S 
 
 'wdo V aad d 
 
 ^^g g. ? ^-s^sss 
 
 ? 
 
 
 
 
 .2 
 
 J 
 
 | 
 
 
 ^ 
 
 c 
 
 1 
 
 P!M 'J^V 
 
 M C, ro M 
 
 f 
 
 
 
 
 .3 
 
 E. 
 >. 
 
 
 I 
 
 S 
 
 8 U S 
 
 * 0,0 + N m^O ~ ~ CO 
 
 M 
 
 3 
 
 
 a 
 1 
 
 H 
 
 ,s 
 
 
 
 
 ^ 
 
 ?3 y '5, 
 
 p-> f-oo m M ej M N w m N ~ 
 
 00 
 
 i- 
 
 
 n 
 
 
 C 
 
 
 b 
 
 J 
 
 -53 
 
 
 " 
 
 2 
 
 
 
 
 8 
 
 | 
 
 i 
 
 Ok, 
 
 
 
 
 2 
 
 
 
 
 
 
 pi 
 
 t 
 
 c 
 
 
 
 
 S 
 
 A B p J3d 
 
 |IJp-2S | 5 H?si 
 
 1 
 
 1 
 
 
 
 4 
 
 a 
 
 
 1 
 
 1 J 
 
 paujBO SSBJ 
 
 
 ': 
 
 3 
 
 
 I 
 
 i 
 
 J3 
 
 
 M 
 a 
 
 d 
 
 1 & 
 
 
 
 
 c 
 
 
 7. 
 
 8 
 
 u 
 
 
 | 
 
 '* 1 
 
 C 
 
 ^ SLI^ !> 5 J; J ? > K?J'p:^ 
 
 -.? 
 
 CO 
 
 
 '- 
 
 . 
 
 2 
 
 
 
 "o >. 
 
 11 
 
 ^^o^^co; g> " y^'^^ 
 
 ? 
 
 T3 
 
 1 
 
 pa 
 
 P 
 
 I 
 
 
 c~ 
 
 ^ J 
 
 
 o.o^ * *MS*m 
 
 1 
 
 S 
 
 
 | 
 
 C 
 
 'c 
 
 
 o 
 
 s '1 
 
 3 
 
 ft, 
 
 
 
 ^ 
 
 
 jj 
 
 I 
 
 
 
 
 L 
 
 
 111 
 
 
 i 
 
 
 
 "5 
 
 
 1 
 
 j 
 
 from year to 
 
 
 al number re 
 
 p of old cars, 
 ral with all st 
 
 o O 
 
 7j '^ 7j V 
 
 00 
 
 3 
 
 
 
 H 
 
 01 
 
 
 
 
 3 S 
 
 
 
 CO Z M CO 
 
 
 
 J 
 
 
 g 
 
 
 
 i 
 
 
 V 
 
 1 & 
 
 c38, 
 
 
 
 . | 
 
 
 o 
 
 x 
 
 
 1 
 
 | 
 
 i 
 
 E g 
 
 S to 
 
 1! 
 
 M - 
 
 t 
 
 | 1 
 
 
 I 
 
 | 
 
 | 
 
 | 
 
 s 
 
 t 
 
 -s 
 
 3 
 
 55 H 
 
 
 
 x 1 
 
 
 -- 
 
 .= 
 
 jj 
 
 1 
 
 !: 
 
 1 i 
 
 d 
 
 M 
 
 10 
 
 " ? S S 
 
 "" 
 
 J S 
 
 II 
 
 | -o 
 
 .' 
 
 n of Pittsbur 
 md Cohoes. 
 
 n of Lynn ar 
 
 cars reported 
 
 i 
 
 4 
 _g 
 
 :t is clearly d 
 
 ro 
 
 | 
 
 .2 
 
 J 
 
 irs that this s 
 old cars is in 
 
 
 
 
 <u **^ 
 
 i & 
 
 5 
 
 *o 
 
 
 s 
 
 r- 
 
 5 J 
 
 HO.xvin.od 
 
 $ % R'S S % S S.'S %<? 5 |S S * " ^ 
 
 I 
 
 a ,* 
 c o 
 
 in - 
 
 3 . 2 
 
 1 "o 
 
 | 
 
 | 
 
 
 
 1 
 
 | 
 
 1 f 
 
 
 
 d 
 
 k. v 
 
 V U 
 
 
 
 c 
 
 - 
 
 ^" 
 
 e 
 
 2 1 
 
 ui 
 
 
 
 <S g. 
 
 1 1 
 
 ?> 
 
 u 
 
 - 
 
 S5 
 
 a 
 
 c % 
 
 5 
 
 J gj^ S S :>;>; S >:> S^J S 5 " J 
 
 
 2 E 
 
 
 | 
 
 H 
 
 
 8 
 
 "o 
 
 1 *5 
 
 H 
 
 CO 
 
 SrtSggsssssssaiassaissai^ 
 
 
 s 
 
 ^ ^ 
 
 | 
 
 M 
 
 
 
 .X 
 
 5 
 
 1 
 
 s * 
 
 
 
 
 S5 
 
 o 5 
 
 1 1 
 
 ^ 
 
 B 
 y. 
 
 .a 
 
 
 
 B 
 
 H 
 
 g 8 g 
 
 S'?S 
 
 H 
 U 
 
 ijjjjijy^iijjjj^ 
 
 
 * This inc! 
 t Based or 
 
 jj 
 
 1 Based on 
 
 < 
 
 ER 
 
 ,' 
 
 o 
 
 I 
 
 1 
 
 1 
 
 It is perso 
 cars for set 
 Thiscond 
 
 
 a - g-o J = f f^-S "SJ ^ -S g i 
 
 
 
 
 
 
 \ 
 
 
 
 j> 
 
 
 5l=ll^ s lzll6'IJ^cJI 
 
 
 
 
 
 
 | 
 
 
 
 1
 
 32 6 THE ELECTRIC RAILWAY. 
 
 In endeavoring to treat of any particular case, the wide 
 variations from which the above averages were made should 
 be borne in mind. 
 
 It has become a common practice to express the cost of 
 operating as a certain percentage of gross earnings. Such a 
 relation is a useful one to be denned. Yet it is to be remem- 
 bered that unless the service in a given city has been well 
 studied and carefully managed, the figures shown for the 
 actual business may be but a poor guide as to what is possi- 
 ble. If approximation be desired and this method be used, 
 operating expense may be taken at 60 to 70 percent, of gross 
 earnings. 
 
 Experience thus far shows that the substitution of electric- 
 ity for horses is followed by an increase in gross receipts 
 ranging from 25 per cent, to 300 per cent, over the original 
 receipts. Even when the service with horses has been good 
 considerable increase (rarely less than 30 per cent.) is 
 marked. No better evidence could be had of the desirability 
 of the electric railway. The expense per car-mile will gen- 
 erally prove to be less than by horses. We say generally, 
 because in a city where horses and forage are exceedingly 
 cheap, while coal and mechanical skill are unusually high, it 
 might be impossible to secure the economy usually counted 
 upon. As an illustration of the difference, exemplified on a 
 large scale, we subjoin a statement made by the West End 
 Railway Company of Boston, showing the greater economy 
 per car-mile of -electric traction. It is to be noted in this 
 case that the figures for horses represent the result of years 
 of effort toward the best attainable practice, while for electric 
 traction the figures cover contract prices for repairs made 
 very early in the history of the business and consequently 
 at a higher price than would now be made. Moreover, 
 the practice generally being new, it remains for time to 
 develop those small economies which show a system at its 
 best. 
 
 For both horses and electricity these figures of cost are 
 high as compared with the rates attainable elsewhere. Coal, 
 forage and labor are, in Boston, well above the average for 
 the entire country. Other statements are also given in a
 
 COMMERCIAL CONSIDERATIONS. 
 
 327 
 
 OPERATIONS OF THE WEST END STREET RAILWAY 
 CO., BOSTON, MASS. 
 
 1891. 
 
 
 
 ELECTRIC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Gross receipts 
 
 $134.321-00 
 
 $144,638.00 
 
 8145,31900 
 
 $i44,553-oo 
 
 $136,246.00 
 
 General expenses 
 
 8,193.00 
 
 7.796-00 
 
 7,465 oo 
 
 6,956.00 
 
 
 Maintenance expenses 
 
 
 8,891.00 
 
 8,078.00 
 
 1 1 ,999.00 
 
 12,498.00 
 
 Transportation expenses 
 
 37,653.00 
 
 36,55200 
 
 31,556.00 
 
 3-,8 9 5.o 
 
 31,123.00 
 
 Motive power 
 Total operating expenses 
 
 30,194.00 
 85.835.00 
 
 30,924.00 
 84,163.00 
 
 26,360.00 
 73,459.00 
 
 26.399.00 
 77,249.00 
 
 25,630.00 
 75,278.00 
 
 Net earnings 
 
 48,487.00 
 
 
 71,860.00 
 
 
 60,068.00 
 
 Miles run 
 
 394,459 
 
 376^321 
 
 360,567 
 
 377',i9i 
 
 363,836' 
 
 Ratio of mileage 
 
 26.68 
 
 25-58 
 
 25.15 
 
 25.19 
 
 24.92 
 
 Per cent, operating expenses 
 
 63-36 
 
 58.18 
 
 50.55 
 
 53-44 
 
 57-'3 
 
 EXPENSES PER MILE RUN. 
 
 CTS. 
 
 CTS. 
 
 CTS. 
 
 CTS. 
 
 CTS. 
 
 Motive power per mile 
 
 07.65 
 
 08.22 
 
 07-31 
 
 07.00 
 
 07.04 
 
 Car repairs per mile 
 
 01.39 
 
 01.33 
 
 01.18 
 
 01.17 
 
 01.57 
 
 Damages per mile 
 
 00.75 
 
 00.89 
 
 
 
 00.23 
 
 Conductors and drivers per mile 
 
 07-33 
 
 07-36 
 
 07.25 
 
 06.92 
 
 06.85 
 
 Other expenses per mile 
 Total expenses per mile 
 Earnings per mile run 
 
 04.63 
 21.75 
 
 34-05 
 
 04.56 
 22.36 
 38.43 
 
 04.47 
 20.37 
 40.30 
 
 05.27 
 20.48 
 38.32 
 
 05.00 
 20.69 
 
 37-45 
 
 Net earned per mile 
 
 12.30 
 
 
 19.93 
 
 17.84 
 
 16.76 
 
 1801 
 
 
 
 HORSE. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Gross receipts 
 
 8344,396-00 
 
 8374.6o6.00 
 
 8404,225.00 
 
 $409,878.00 
 
 $392,474.00 
 
 General expenses 
 
 22,514.00 
 
 22,682.00 
 
 22,218.00 
 
 20,657.00 
 
 18,160.00 
 
 Maintenance expenses 
 Transportation expenses 
 
 22,692.00 
 114,001.00 
 
 17,748.00 
 
 18.560.00 
 106,836.00 
 
 27.228.00 
 
 18,669.00 
 105,953.00 
 
 Motive power 
 Total operating expenses 
 
 117,740.00 
 276,947.00 
 
 118,972.00 
 269,556.00 
 
 116,211.00 
 263,825.00 
 
 72,'888.oo 
 
 "3,35i.oo 
 256,133.00 
 
 Net earnings 
 Miles run. . 
 Ratio of mileage 
 
 67,449.00 
 1,083,887 
 73-32 
 
 105,049.00 
 1,094,683 
 74-42 
 
 140,399.00 
 1,073,218 
 74-85 
 
 36,990.00 
 ', 20,377 
 
 136,341-00 
 1,096,376 
 75.08 
 
 Per cent, operating expenses 
 
 80.62 
 
 7"-95 
 
 65-27 
 
 [66.58 
 
 64.52 
 
 EXPENSES PER MILE RUN. 
 
 CTS. 
 
 CTS. 
 
 CTS. 
 
 CIS. 
 
 CTS. 
 
 Motive power per mile 
 
 10.86 
 
 10.86 
 
 10.83 
 
 w.38 
 
 10.33 
 
 Car repairs per mile 
 
 00.93 
 
 00.60 
 
 00.61 
 
 
 00.03 
 
 ] damages per mile 
 
 00.78 
 
 00.37 
 
 00.15 
 
 oo!o6 
 
 
 Conductors and drivers per mile 
 
 08.28 
 
 08.24 
 
 
 08.23 
 
 otio 
 
 OthT expenses per mile 
 Total expenses per mile 
 Earnings per mile run 
 
 04.70 
 25-55 
 31.77 
 
 04-55 
 24.62 
 34-22 
 
 04.74 
 
 24.58 
 37.66 
 
 05-07 
 
 33 
 
 04.81 
 23-37 
 35.80 
 
 Net earned per mile 
 
 06.22 
 
 09.60 
 
 
 12.23 
 
 12.43 
 
 Allowances should be made for the omission of cents in the above statement.
 
 32 8 THE ELECTRIC RAILWAY. 
 
 foot-note,* showing results obtained elsewhere on a somewhat 
 smaller scale. 
 
 Those companies having severe snows to contend with must 
 expend a considerable, but widely variable, amount of money 
 in order to clear their tracks. In some of our most northern 
 cities a period of inaction is forced by the magnitude of ex- 
 pense that would be incurred in clearing tracks, and also 
 because such effort on the part of the railway company would 
 render the streets unfit for the service of " runners" wheeled 
 vehicles being out of the question for any transportation. 
 
 The electric snow-sweepers now being put on the market 
 will undoubtedly diminish the interruption to traffic. But 
 they will demand a good supply of power. Even with their 
 assistance the severe winter months must remain but slightly 
 profitable. The average road lying north of 40 north latitude 
 
 * In February 1891 President D. F. Henry, of the Federal Street and Pleasant Valley 
 Passenger Railway Company of Pittsburg, Pa., made a report of the operation of the 
 company for the preceding six months, in which was embodied some interesting infor- 
 mation m this line. We quote from it as follows . 
 
 Passengers carried, six months, 3,370,531 ; receipts. $168,526.55; mileage, 31 cars, averag- 
 ing each 108 miles per day , total daily average, 3,348 miles. Cost per mile : 
 
 Conductors and motor-men per mile, 6.80 cents. 
 
 Motor and electric repairs " 1.68 
 
 Mechanical repairs 1.14 " 
 
 Motive power ' 1.54 " 
 
 Overhead system ' 0.45 " 
 
 Maintenance of way " 1.08 " 
 
 General expense ' 0.88 " 
 
 Stables ' 0.46 " 
 
 Officers and salaries " 0.84 " 
 
 Interests " 2.71 ' % 
 
 Tolls 0.25 " 
 
 General labor " 2.43 " 
 
 Total 20.26 cents. 
 
 as total cost per mile of operating expense, fixed charges, salaries and interest, as 
 against receipts per car-mile of 27.55 cents ; showing net earnings per mile of 7.29 cents. 
 
 Separating the above into strictly operating expense and fixed charges, we have oper- 
 ating expense. 12.74 cents per mile ; and in comparing the same with the cost of operating- 
 the horse line, which was, 10 cents per car-mile, we must remember that we then paid but 
 one man on a car. where we now pay four, or 3 cents per mile, against 6.80 cents now. 
 This increased cost per mile for conductors and motor-men is a necessary adjunct of rapid 
 transit, and is not peculiar to the system. Allowing for this, we have a difference of 1.04 
 cents per car-mile in favor of electricity as against animal power. 
 
 1 call attention to the fact that 'we have one of the most difficult roads in existence 
 to operate, with streets having over one hundred curves, many of which have a short 
 radius, and with heavy grades, leaving but little straight and level roadway. We have 
 also six steam and six cable railway crossings. 
 
 President John N, Beckley of the Rochester Railway Company, at a recent meeting 
 of the Street Railway Association of the State of New York, gave the following as the 
 result of the experience of his company 
 
 n the month of May last the Rochester Railway Company operated 44 18 foot vestibule 
 electric cars. The gross receipts from passengers riding on these cars during the month
 
 COMMERCIAL CONSIDERATIONS. 329 
 
 in the United States will not be wide of the mark in calculat- 
 ing that its profits must be made in nine months of the 
 twelve the other three just paying their own expenses. 
 
 As an assistance in the commercial management of an 
 electric railway we give in Appendix D a set of accounting 
 rules. These were very carefully prepared by Mr. H. I. 
 Bettis, of Boston, while serving as auditor of the accounts of 
 the Thomson-Houston Electric Company, in their contract 
 work of maintaining electric equipment and lines of the West 
 End Railway Company the work being under the direction 
 of Mr. W. E. Baker. Mr. Baker has kindly added to these 
 rules a few words on the economical organization and ad- 
 ministration of the working force of an electric railway. 
 
 A comprehensive statement of monthly operations may be 
 made on a form similar to form "C." It will be seen on this 
 
 were $37,053, or 23.15 cents per car-mile for a mileage of 159,567 miles. The total expense of 
 operation of these cars for that month was $18,332, thus leaving a net profit of $18,721. 
 The total cost of operation per car-mile was 11.4 cents, and the profit per car-mile was 
 therefore 12.11 cents It may be observed in passing that the operating expense was a 
 trifle under 50 per cent, of the gross receipts The cost of operating was divided as 
 follows : 
 
 Motive power 2.8 cents. 
 
 Car repairs 0.7 
 
 Conductors and motor-men 4.9 " 
 
 Other expenses 3.0 
 
 During the same period the company operated 62 horse cars, all of them without con- 
 ductors. Most of the horse cars were one-horse or bobtail cars. The total cost of operat- 
 ing the horse cars, without conductors, during this period was about 10 cents per car- 
 mile, but the total receipts per car-mile were but little above 12 cents. 
 
 In the month of June the Rochester Railway Company operated 54 electric cars and 60 
 horse cars. The electric cars earned each per day $23.60, or 22.77 cents per car-mile, and 
 the total expense of operating them per day was $10.50, or 11.07 cents per car-mile. The 
 cost of operation per car-mile was divided as follows : 
 
 Motive power 2.40 cents. 
 
 Car repairs i.oo " 
 
 Conductors and motor-men 5.66 " 
 
 Other expenses 2.01 '' 
 
 Making a total per car-mile of 1 1.07 cents 
 
 The cost of operating the horse cars during the same month per car-mile was 11.06 
 cents, and they earned 14.37 cents per car-mile. My experience in the operation of street 
 railroads has convinced me that the most economical system of operation is the electric 
 system. I have not, in the statements which I have now made, taken into consideration 
 the greater fixed charge in the operation of an electric railroad as compared with a horse 
 rnilroad, due to the much greater cost of the former; but in arriving at the conclusion 
 which I have above expressed, due consideration has been given to this element of 
 increased cost. We know that when a horse railroad is changed over and operated by 
 electricity the receipts are very largely increased. It is safe in any case to say that the 
 increase in gross receipts will be at least 15 per ce.it., and the average increase is 
 probably as high as 30 per cent. Some of the increase is undoubtedly due to the greater 
 mileage which the cars make, and still more is due to the cleaner, more rapid and more 
 comfortable transportation of the people.
 
 330 
 
 THE ELECTRIC RAILWAY. 
 
 form that the final result of a month's operation is expressed 
 in four different manners namely, per cent, of expenses to 
 earnings, net earnings, expense per car-mile, and expense 
 per passenger. All these have a certain value for comparison. 
 Neither, alone, tells the whole story. 
 
 The next matter is how to arrive at and keep those ex- 
 penses sufficiently accurately and without too much clerical 
 labor; and this brings us primarily to the stock or supply 
 account. All material while in the stock room, and until 
 used, evidently bears the same relation to the company as 
 cash in the till or bank, except for availability, and should 
 be so treated. Material should not be charged to the expenses 
 of the road until used, and should be charged when used, 
 regardless of the bill for this material being received or paid. 
 
 A careful record should be kept of all material 'received 
 and all material used on blanks somewhat similar to forms 
 " E" and " D," bound in book form. It is sometimes consid- 
 ered better to have the form " D" on loose sheets instead of 
 bound, and there are apparent variations to be made in the 
 blank to suit individual cases. 
 
 The " Material Received" book, or form " E," will be found 
 to pay for itself in checking bills and preventing duplication. 
 The material used is arrived at by charging out of the stock 
 room on "Foreman's Requisition," in small quantities, as 
 used, or obliging a foreman lo make on a proper form a daily 
 or weekly or monthly report of the material he uses. A 
 simple form is that of form "F." One man should receive 
 all the material for the road ; he may, of course, be a store- 
 keeper for this purpose if the road is sufficiently extensive ; or, 
 if not, this duty may be added to the other duties of the clerk. 
 
 An invoice book with copies of all bills should be kept, 
 and a monthly summary of both books should be made giv- 
 ing from the invoice book all the charges to stock, and from 
 the material used all the credit to stock, and all the charges 
 of material to expense properly classified. If in addition a 
 stock ledger is kept, it will be found that all errors and mis- 
 takes in foreman's and clerk's pricing, etc., will correct 
 themselves ; and it will be easy at any time to check up the 
 ledger with-the material actually on hand in the stock room.
 
 COMMERCIAL CONSIDERATIONS. 33! 
 
 In regard to the labor, there are many forms of time books 
 used; the one here presented as form " P" is convenient, 
 and the foremen soon learn to classify their time with little 
 trouble and few mistakes. On the pay roll should be classi- 
 fied these expenses, and the total of all expenses is thus 
 easily found. 
 
 The mileage sheets from " A" will be found very complete, 
 and will give a record of the mileage of each car per month. 
 The original records can be taken from the conductor's 
 report most easily on a blank book, and transferred to these 
 reports monthly. The conductor's reports should cover a 
 report of the number of passengers carried on each trip, the 
 number of trips made, and the route. There are many 
 excellent forms for these. 
 
 To summarize, if the superintendent, as is frequently the 
 case, has an office distinct from the office of the company, he 
 would be required to have the following books : stock ledger, 
 time book, pay rolls, material used and material received 
 books. Copy books as follows will be found convenient: A 
 letter book, a copy book for pay rolls, a requisition book (un- 
 less he buys his own supplies wholly, when this becomes an 
 order book), a daily report of earnings book or deposit book, 
 a motor report book for copying reports made on blank form 
 "S,"and a statement and bill copying book for bills made 
 against other parties. Monthly reports should be made about 
 as follows: mileage report, material used, balance of stock 
 ledger, receipts report, invoice summary. The daily re- 
 ports would be of daily earnings and of condition of motor 
 cars. 
 
 On a road of any considerable size, say of 20 cars and 
 upward, the superintendent will need the assistance of six 
 good men namely, a chief conductor, engineer at power 
 house, line foreman, foreman of car repairs, foreman of track, 
 and clerk ; and, in fact, this organization will need only to 
 be extended to operate a road of the largest size, when the 
 foreman of car repairs becomes the master mechanic, the 
 clerk the auditor, etc. 
 
 As to the force employed with an equipment of two 15 
 horse-power motors per car, we can allow about as follows:
 
 332 THE ELECTRIC RAILWAY. 
 
 One day mechanic to 12 cars. 
 
 One inspector to 50 cars. 
 
 One cleaner to 1 5 cars. 
 
 The motors ought to be cleaned thoroughly every third 
 day. When less than 25 cars are run from one car house, 
 the repair foreman may take the place of one mechanic, and 
 in case less than 12 cars are run from one car house one 
 mechanic may do the cleaning and repair work. The fore- 
 man will be in addition to the figures given, and these men 
 will take care of the motors only; if additional work is re- 
 quired to any extent in repairs of the car bodies, etc., more 
 men will be needed. 
 
 A proper manner to handle the bills for supplies and ex- 
 pense is as follows : provide a stamp with which every bill can 
 be stamped on receipt, and which provides a place for the 
 storekeeper to certify that the goods have been received, and 
 have the bill checked and a place for a clerk to certify that 
 any extensions and additions on the bill are correct; also a 
 place for the same person ordering the goods to certify to the 
 same ; lastly, a place for the approval of the superintendent. 
 It is important that all the names be written on the bill certi- 
 fying to these facts, as then the whole history of the bill can 
 at any time be ascertained.
 
 CHAPTER XII. 
 
 HISTORICAL NOTES. 
 
 THE history of the electric railway presents some very 
 curious features. Important though this particular applica- 
 tion of electricity is to-day, it is almost impossible to trace 
 out the various steps in its development with anything like 
 certainty of giving credit to whom credit is due. For not only 
 has the electric railway been a gradual evolution of ideas in 
 the hands of many inventors working almost simultaneously, 
 but it is almost unique in that it was worked out so far as, 
 invention is concerned during two distinct periods ; first, 
 at the time when the only source of electricity was the bat- 
 tery, and then over again after a commercial source of elec- 
 tric current had been found in the dynamo. 
 
 So long as the efforts of inventors were confined to the use 
 of batteries, almost nothing could be done in the way of 
 practical electric railway work; for its cost was evidently so 
 prohibitive as to deter inventors from spending time and 
 money on the solution of the various problems involved. Not 
 until the modern dynamo had been invented and the great 
 principle of its reversibility discovered was it possible to 
 bring electric motive power to anything like a practical state. 
 
 This fact of there being two distinct periods of evolution 
 in the history of the electric railway has had, perhaps, a 
 happy effect on its modern development, in that the funda- 
 mental principles and methods had already become the 
 common property of inventors, and hence the growth of the 
 art has not as yet been checked by any basic patents of so 
 formidable a description as to deter improvement on the 
 part of others than some single individual or company. 
 
 The inception of the electric railroad and the first period 
 of its history had its scene of action in the United States ; 
 the working out of the broad principles on which modern 
 
 333
 
 334 THE ELECTRIC RAILWAY. 
 
 electric traction is based, and the first great steps toward 
 the reduction of these principles to practice, took place 
 abroad ; and then again the scene changed back to America 
 so completely that one may say, without risk of doing any 
 injustice, that electric traction as a whole is of American 
 development. 
 
 The electric motor, from which has sprung so many 
 important practical applications, may probably be said to 
 have had its origin in the Barlow wheel in 1826; and, oddly 
 enough, this first effort was the forerunner of a rather ad- 
 vanced type of machine, being the motor counterpart of 
 Faraday's disk machine and the unipolar dynamos of later 
 years. Inventors were to pass through a long period of per- 
 plexity and roundabout expedients before they came back 
 again to the principle of a reversed dynamo. 
 
 In 1830 the Abbe Salvatore Dal Negro, an Italian, pro- 
 fessor of natural philosophy in the University of Padua, de- 
 vised a reciprocating form of motor in which a permanent 
 magnet oscillated between the poles of an electro-magnet 
 commutated at each movement. A year or two later many 
 inventors worked in the same line, but the first step toward 
 the electric railroad was taken b)' Thomas Davenport, a 
 blacksmith living in Brandon, Vt., who, quite independently 
 of previous researches, devised a rudimentary electric motor 
 working on the general principle of revolving an electro- 
 magnet by its attraction for fixed armatures or magnets, the 
 current being commutated at the proper time to let the poles 
 pass the first set of armatures and take up the work at the 
 next. This he applied to an automobile electric car supplied 
 by batteries carried upon it, and as early as 1835 he con- 
 structed a little circular electric road. 
 
 This was exhibited during the autumn of 1835 in Spring- 
 field, Mass., and also for two weeks in December of the same 
 year in Boston. The road consisted of an automobile car car- 
 rying its own batteries. Of course nothing came of this rudi- 
 mentary experiment, and the first extension of the same idea 
 was at the hands of Robert Davidson, of Aberdeen, Scotland, 
 who resides there to-day perhaps the oldest living electri- 
 cian. He began his experiments about 1838, at the period
 
 HISTORICAL NOTES. 335 
 
 when wide attention was called to the possibilities of the 
 electric motor by its use in the hands of Jacobi for propelling 1 
 a small boat on the river Neva. The power in this case was 
 furnished by primary batteries, and the speed reached by the 
 boat 28 feet long and 7 feet beam about 3 miles per hour. 
 
 Davidson was filled with the idea that the electric motor 
 would find a place in the ordinary locomotive work of rail- 
 ways. He built with this in view a powerful motor, and 
 fitted it on a truck of the ordinary railway gauge. The motor 
 carried batteries for supplying the power; it was 16 feet 
 long, 5 feet wide, and its total weight including the bat- 
 teries was about 5 tons; 40 battery cells, were employed, of 
 a type claimed by Davidson as his own, although this was 
 disputed by Mr. Sturgeon and J. Martin Roberts. The ele- 
 ments consisted of plates, 12 by 15 inches in size, of iron and 
 of amalgamated zinc immersed in dilute sulphuric acid. 
 The current obtained from these was delivered to an electric 
 motor consisting of a pair of drums, each bearing two sets of 
 iron bars parallel to the axles. Eight electro-magnets were 
 placed on the bottom of the car in two opposite rows, and 
 put the drums in rotation by attracting the armatures fas- 
 tened upon them, while a commutator revolving with the 
 armatures made the necessary changes of current in the 
 magnets. Davidson's experiments are principally interest- 
 ing on account of the large scale on which they were carried 
 out. The locomotive just mentioned made several successful 
 trips on some of the Scottish railways, and, finally, while the 
 machine was being taken home to Aberdeen, it was found 
 one morning in the engine-house at Perth shattered beyond 
 repair by some mischief -making intruder. The railway en- 
 gineers of the time felt very strongly that this innovation 
 might be destined to supersede their own machines, and 
 evidence pointed to the destruction of this first electric loco- 
 motive as having been at their hands. 
 
 After the Davidson experiments nothing on so large a 
 scale was attempted for more than a decade, when Prof. C. 
 G. Page, of the Smithsonian Institution, well known as 
 a versatile genius in many fields of research, took up 
 the subject of the application of the electric motor to
 
 236 THE ELECTRIC RAILWAY. 
 
 railroading and performed some exceedingly interesting 
 experiments. 
 
 Page, like Davidson, was thoroughly persuaded that the elec- 
 tric motor might be successfully used in regular railway work, 
 and a machine of 10 horse-power was built by him in 1850. 
 The general design followed was similar to a steam engine, 
 two solenoids replacing the cylinder, and drawing backward 
 and forward a soft iron piston rod which actuated a fly-wheel 
 by means of an ordinary connecting rod and crank, while the 
 motion of the piston shifted the connections from one solenoid 
 to the other at the end of the stroke. 
 
 From this design he passed to that of two double solenoids 
 acting on U-shaped magnet cores in a precisely similar way, 
 and he eventually wound his solenoids in sections so that 
 by a sliding commutator the current could be so applied as to 
 produce both a long and a strong pull. 
 
 His first experiment with the electric locomotive made on 
 this model was tried April 29, 1851, along the Washington 
 and Baltimore Railroad to Bladensburg. This later machine 
 was of 1 6 horse-power, and derived its supply of electrical 
 energy from 100 large flat Grove cells stored on the locomo- 
 tive, each having platinum plates 1 1 inches square. 
 
 The speed attained in this experiment was quite high, 
 being at some points in the run at the rate of 19 miles an 
 hour. Trouble was encountered, however, from breakage of 
 the light earthenware porous cups, and after several subse- 
 quent' experiments Prof. Page dropped the subject, having 
 proved that considerable power and speed could be obtained 
 at, however, a prohibitive cost. 
 
 Two other experiments on electric railroading, at about 
 the same time, are worth mentioning, although on a very 
 small scale. One was by Prof. Moses G. Farmer, who built 
 a model locomotive, running on a track of 1 8-inch gauge and 
 carrying a small car, which could accommodate two passen- 
 gers. The power supplied was but small, furnished by 48 
 pint cells of Grove battery. In 1851 he built another little 
 model road, which is important as showing a decided devel- 
 opment in the art. It was constructed by Mr. Thomas Hall, 
 of Boston, Mass., and was exhibited at the Charitable Me-
 
 HISTORICAL NOTES. 337 
 
 chanics' Fair. The track on which it ran was 40 feet long, 
 and the gauge was but 5 inches. The motor was such an one 
 as had been frequently used by Page and others, consisting 
 merely of a permanent magnet, with an electro-magnet, pro- 
 vided with a suitable commutator, rotating between its poles. 
 This armature shaft was provided with a worm, which en- 
 gaged the teeth of an intermediate gear wheel that drove 
 the axle. 
 
 The track was so arranged that when the car reached either 
 end of it a pole changer would automatically be thrown over 
 and reverse the direction of motion. The specially interest- 
 ing feature of this road, however, is the fact that the current 
 was supplied to the motor through the rails, and not from 
 batteries carried on the car ; further, the motor armature was 
 geared to the driving axle with a considerable speed reduc- 
 tion, which embodied the principle of economical working 
 that has since been generally followed. 
 
 This model of Mr. Hall's very plainly shows the combina- 
 tion of a moving motor car supplied from a stationary source 
 of energy, and, with some similar anticipations, has gone far 
 to prevent any fundamental patents from being granted dur- 
 ing the second stage of the development of the electric rail- 
 road. In fact, the same idea had been disclosed four years 
 previously by Mr. Lilley and Dr. Colton, of Pittsburg. 
 Their device was simply a little model electric locomotive 
 running around a circular track. The rails were insulated, 
 and one was connected with each pole of the battery. 
 
 In 1840, however, a provisional patent had been granted 
 in England to Henry Pinkus, in which not only was the 
 general idea of supplying current to a moving train from a 
 fixed source plainly shown, but it was even suggested that 
 an electric motor might be used to drive the train itself. 
 The patent had primary reference to ordinary railroad work, 
 and its provisions are not very clear ; but the idea was very 
 distinctly put, and Pinkus even suggested the possible use 
 of mechanical generators to replace the batteries shown in 
 his plans. 
 
 It may also be mentioned that an English patent to Swear, 
 in 1855, on telegraphy from moving: trains, involved in a very
 
 338 THE ELECTRIC RAILWAY. 
 
 obvious way the idea of taking current from an uninsulated 
 conductor running along the line, thus forestalling the mod- 
 ern trolley system. 
 
 The same principles are shown and described in French 
 and Austrian patents granted to Major Alexander Bessolo in 
 1855. The French patent shows what is now known as the 
 third-rail system, with the return through the ordinary rails. 
 It was proposed originally for train telegraphy, as in the 
 case of Pinkus and others. Bessolo also stated that he pro- 
 posed to use an overhead wire for one side of the circuit and 
 the rail for the other, the current being taken from a sta- 
 tionary generator. In a supplemental specification to the 
 Austrian patent the inventor explained that he proposed to 
 use this system of distribution for driving vehicles on ordi- 
 nary roads and on railways, and stated, as in the French 
 patent, that the current may be conveyed to the locomotive 
 machine by means of a circuit formed by a conductor insu- 
 lated from the ground and suspended in a manner analogous 
 to telegraph wires and by the rails of the track. Bessolo 
 further suggests that such a system of locomotion would have 
 great advantages on account of the ability of the current easily 
 to reach independent locomotives or vehicles composing 
 small trains, and that, besides, the generators were to be so 
 located that they could be controlled from the stations. 
 
 All these early efforts at electric traction, however, resulted 
 in nothing practical, for the simple reason that power had to 
 be derived from batteries and was consequently too expensive 
 for commercial use. It was not until the dynamo had ap- 
 peared and the principle of its reversibility had been estab- 
 lished that anything could be hoped for along this partictilar 
 line of progress. 
 
 In 1864 Pacinotti made his now famous electro-magnetic 
 machine, and, crude though it was, and unappreciated at the 
 immediate time, it was the forerunner both of the modern 
 dynamo and motor ; for Pacinotti understood perfectly well 
 that his machine was reversible, and that if current were 
 supplied to it power could be obtained ; while if power were 
 supplied current could be obtained. 
 
 Three years 'later Wheatstone and Siemens, independently
 
 HISTORICAL NOTES. 339 
 
 and almost simultaneously, devised self-exciting dynamos 
 that of the former being shunt wound, that of the latter 
 series wound ; and at about the same time Siemens found 
 that his machine could be used as a motor, although his 
 discovery produced no immediate results. 
 
 He is said, however, at that period to have contemplated 
 its adaptation to electric traction. Six years later the use of 
 the dynamo as a motor was clearly demonstrated by Messrs. 
 Fontaine and Breguet at the Vienna Exposition. Just pre- 
 vious to this time Siemens and Gramme had brought out 
 commercial machines, and the latter had observed in his 
 workshop that his dynamo could be used as a motor. 
 
 Fontaine and Breguet had intended to exhibit the principle 
 by running one of the dynamos as a motor from a primary or 
 secondary battery, but failing in this they coupled two ma- 
 chines together, and finding that the experiment was thor- 
 oughly successful, used the motor to drive a pump at the 
 formal opening of the Machinery Hall, June 3, 1873. 
 
 The experiment, of course, created a great sensation, and 
 during the next four years it became familiar in lectures, and 
 was on a few occasions used for the transmission of a small 
 amount of power. The scene of action now shifted to 
 America, and the experiments of Mr. George F. Green, of 
 Kalamazoo, Mich., began. Green was a poor mechanic, but 
 possessed of more than ordinary skill and ingenuity, and for 
 years had been interested in the study of electricity. About 
 1875 his ideas took shape, and he constructed a little model 
 electric railway quite similar to that devised by Farmer, but 
 on a somewhat larger scale. 
 
 Green used the track rails to transmit the current from the 
 source of electricity a battery in his experiments to the 
 moving car, which was driven by a small electric motor of 
 the pole-changing type. He also proposed, as shown by his 
 drawings of that date, to use an overhead wire as one of the 
 sides of the circuit, but the experiment was not carried out. 
 
 During the next few years he was working at the problem 
 as steadily as his occupation permitted, and in 1878 or 1879 
 built a larger model, with a car large enough to seat one or 
 two people, propelled in substantially the same way.
 
 3 4 o THE ELECTRIC RAILWAY. 
 
 He fully understood the advantages of employing a dynamo 
 instead of batteries, but no dynamo was available for the 
 experiments. Not until August, 1879, did he apply for pat- 
 ents; and then, through lack of funds, he was obliged to act 
 as his own patent attorney, with the customary result of 
 encountering great difficulties in the Patent Office. He went 
 into interference with other inventors, and his claims, after 
 being rejected on appeal to the Commissioner of Patents, 
 were taken to the Circuit Court of the District of Columbia, 
 which tribunal alone has the right to order a patent issued 
 under such circumstances. 
 
 Here he was successful, but it was not until the latter part 
 of 1891 that the decision was finally reached, and in December 
 a patent was issued covering more broadly than would a pri- 
 ori be supposed possible the principles of the electric railway. 
 His experiments formed the connecting link between the 
 older and the newer period of inventions in the line of electric 
 traction. 
 
 It was not until 1879, however, that the electrical trans- 
 mission of energy was taken up on a serious scale. On 
 Ascension Day, 1879, there was carried out at the little vil- 
 lage of Sermaize, near Paris, by Chretien and Felix, an ex- 
 periment on plowing by electricity which marks so decided an 
 epoch in the subject as to be worthy of more than a hasty de- 
 scription, particularly as it has been too lightly passed over in 
 most historical notes on the transmission of power. 
 
 Plowing by steam power had already been tried in England 
 and the United States, and it occurred to the engineers who 
 instituted this notable experiment that the electric transmis- 
 sion of power offered great opportunities in facilitating the 
 convenient application of mechanical power to such tasks. A 
 rectangular plot of ground 220 meters wide had been laid out 
 for the experiment, and the method employed was to estab- 
 lish at each end of a furrow a massive wagon carrying a 
 motor, dragging a double-ended gang-plow by means of a 
 cable and drum. A similar device had been previously used 
 in steam plowing with tolerable success. 
 
 In this case there was mounted on each wagon a " type A" 
 Gramme machine of the sort ordinarily used for electric
 
 HISTORICAL NOTES. 34! 
 
 lighting; these two were united by a conductor 250 meters 
 long, forming part of a complete metallic circuit of copper 
 wire 3 millimeters in diameter connecting them with two 
 precisely similar Gramme machines for furnishing power. 
 These two generators were distant, at various times during 
 the experiment, from 400 to 620 meters from the motors. 
 
 The wagons themselves were automobile, as an axle of 
 each could be driven by means of throwing the motor into 
 gear with a large gear wheel connected to a wheel upon the 
 axle by a sprocket chain. Probably about 3 horse-power was 
 developed at the motors in this experiment, which was en- 
 tirely successful. The speed reached by the plow was 40 to 
 50 meters per minute. The drums on which was wound 
 the cable were driven by the motors through the interven- 
 tion of friction gearing. 
 
 The power was supplied from the Sermaize beet sugar 
 factory, and during the latter part of the previous season 
 (1878) Chretien and Felix had arranged at the river front, 
 about 100 yards distant from the factory, an electric hoist 
 operated from the machines in the factory and serving to 
 transfer the beet root from the boats in which it was brought 
 down the river to the factory wagons. Up to the time of 
 the famous plowing experiment some 4,000 tons had been 
 thus handled, at a cost of 40 per cent, less than unloading 
 by hand. 
 
 The experiments just mentioned are in all probability the 
 first examples of the transmission of power on a considerable 
 scale either to fixed or movable motors. 
 
 During the summer of 1879 the fi rst electric railway was 
 put in operation, by the firm of Siemens & Halske, at the 
 Industrial Exposition in Berlin. An oval track about 300 
 meters in circumference was laid down, and upon it ran. a 
 little electric locomotive dragging a single platform car. A 
 third rail placed midway between the others served as the 
 working conductor, and the current was taken from it by 
 means of a sliding contact under the locomotive. The motor 
 used was one of the regular Siemens dynamos, placed with 
 its armature spindle parallel to the track ; and the power was 
 transmitted to the axle by a double-gear reduction, including,
 
 342 THE ELECTRIC RAILWAY. 
 
 of course, a pair of bevel gears. The outer rails served as a 
 return for the current through the wheels of the locomotive. 
 Eighteen or 20 passengers constituted the full load of the 
 little train, and the time required for a complete circuit was 
 from one to two minutes, corresponding to a speed of perhaps 
 8 miles per hour. 
 
 This road is notable as being the practical starting-point 
 of modern electric traction. It is beyond all question the first 
 working electric railroad on a practical scale. Up to that 
 time nothing had been constructed which bore even the sem- 
 blance of a commercial electric railway, and such few plans 
 as had existed before, both in Europe and America, were 
 mainly on paper. 
 
 The success of the Berlin experiments stimulated invention 
 in every part of the world, and the natural result was repeti- 
 tion of them under varied conditions and in different places. 
 During the exposition at Vienna in the succeeding year Egger 
 showed a model electric railway in which the third rail was 
 dispensed with and the two working rails alone constituted 
 the metallic circuit. Very early in the same year Siemens 
 had been negotiating with the city authorities of Berlin for 
 a franchise for an electric road on a considerable scale. 
 About the same time some experiments were conducted in 
 Paris with a view toward the utilization of a very small elec- 
 tric road as a substitute for pneumatic tubes, constituting a 
 sort of forerunner of automatic electric traction. 
 
 Not until the middle of 1880 was any real work in electric 
 railroading done in America. Mr. T. A. Edison was, so far 
 as experimental work goes, first in the field, although in 
 projecting such a system priority was awarded in the course 
 of Patent Office litigation to Stephen D. Field, who filed a 
 caveat on the subject May 21, 1879, and an application for a 
 patent the succeeding March. 
 
 During the last months of 1878 and the first of 1879 Mr. 
 Field had been doing some experimentation on electric mo- 
 tors, arranging among other things a small electric hoist; 
 but so far as railroading is concerned his plans remained on 
 paper until the latter part of the year 1880, a full year after 
 the Siemens experimental road had been operating in Berlin.
 
 HISTORICAL NOTES. 343 
 
 Mr. Edison, too, only began work in 1880, and it was 
 June 5 before he applied for a patent, nearly a year after 
 the Siemens road was in successful operation, and some 
 months after the same distinguished pioneer had been nego- 
 tiating with the city authorities of Berlin for an electric road 
 on a considerable scale. 
 
 Edison laid down a short experimental track near his lab- 
 oratory at Menlo Park, N. J., and conducted some crude but 
 interesting experiments; in his first locomotive the power 
 was transmitted to the car wheels by means of belts, and the 
 two rails were utilized as conductors. 
 
 The next year, 1881, saw decided changes and develop- 
 ments. Mr. Field had by that time put a small experi- 
 mental road at Stockbridge, Mass., under way, and Edison 
 was continuing his experiments. Meanwhile Siemens, the 
 practical originator of modern electric traction, had not been 
 idle, and plans of an elaborate character never realized in 
 their original form were drawn up for electric rapid transit 
 in Berlin. In May, 1880, these had taken shape, and were 
 in the main for an elevated road of a meter gauge, the rails 
 of which were to form the conductors, while the power was 
 to be transmitted from the motor armature to the driving 
 wheels by means of belts. At the same time Siemens was 
 drawing up plans for electric rapid transit express service 
 somewhat like that proposed in Paris some months before. 
 
 His scheme was for an automatic electric road of a little over 
 a foot in gauge running completely incased in a sheet-iron tube 
 of square section about 20 inches on a side. One especially 
 interesting feature of this proposition, which was fully illus- 
 trated in the Electrotechnische Zcitsclirift and La Lumiere Elcc- 
 triqne at that time, was that the armature of the motor was to 
 be directly on the axle of the driving wheels, much after the 
 plan that is being followed in the gearless motors of to-day. 
 The speed proposed was about 40 miles an hour, and the 
 service was to be entirely automatic. 
 
 It was not until the next year, however, that the first com- 
 mercial electric road for regular service was put into opera- 
 tion. This was formally opened to the public May 12, 1881, 
 at which time the Lichterfelde line which is in operation
 
 344 THE ELECTRIC RAILWAY. 
 
 to-day was finally completed. In this, the third-rail method 
 
 of supply is employed. The construction of the road was 
 begun in the latter part of 1880. This Lichterfelde line was 
 the first commercial electric road, as the little line of 1879 at 
 Berlin was the first extensive experimental road. 
 
 It is worth noticing that July 20, La Lumiere Electrique 
 announced that the Lichterfelde line had up to that time 
 been running without any serious accident, that a company 
 for continuing it had already been formed, and in addition, 
 that Siemens & Halske were just then changing the horse 
 railroad that ran from Charlottenberg to Spandau into an 
 electric road. In this installation to avoid some of the 
 inconveniences encountered in supplying the current through 
 the rails overhead wires were to be used, to which com- 
 munication was made by a trolley towed by the car. 
 
 The experimenters were then hesitating between the use 
 of the double trolley line and the single trolley using the 
 rails as return, as practiced to-day. A month or two later, 
 at the Paris Exposition, the system of overhead supply was 
 used, in a somewhat different form, however, from that just 
 mentioned. The Siemens road exhibited there was \vorked 
 on the double trolley system, but the conductors, instead of 
 being wires, were a pair of slotted tubes in which slid contact- 
 makers connected to and dragged by the car. 
 
 Meanwhile the accumulator had not been forgotten. In 
 1880 a locomotive driven by accumulators w r as put into oper- 
 ation at the linen bleaching establishment of M. Duchesne- 
 Fourmet, at Breuil en Auge, the total length of the track 
 being over a mile. In June, 1881, an electric car operated 
 by accumulators was tried upon the Vincennes tramway line, 
 and from about that time the development of accumulator 
 traction went gradually on, both in England and on the Con- 
 tinent. 
 
 The next noteworthy electric road to be put in operation 
 was that at Portrush, in the north of Ireland. Soon after the 
 franchises for a tramway line over the route had been ob- 
 tained it was suggested to Dr. Werner Siemens that it was 
 a good place for installing an electric road, and work was at 
 once begun. The line was about 6 miles long. It was com-
 
 HISTORICAL NOTES. 345 
 
 pleted in the early summer of 1883, and the regular running 
 of the trains began November 5 of that year. At the start 
 the power was generated by steam, but a pair of turbines was 
 soon installed at a convenient point on the river Rush, 1,600 
 yards distant from the nearest point on the line, and for the 
 last seven years the road has been in steady and successful 
 operation. The method of supply is by a third rail. 
 
 Up to 1883 the electric road in this country was practically 
 undeveloped all the advances had been made elsewhere; but 
 soon the scene of activity was to shift. The conflicting inter- 
 ests of Field and Edison were consolidated after a couple of 
 years of litigation, and the Electric Railway Company of the 
 United States was organized early in May, 1883. The next 
 month the Chicago Railway Exposition was to open, and by 
 the greatest energy and push an electric locomotive was con- 
 structed and a road laid around the gallery of the main 
 building, the total length being a little over 1,500 feet. It 
 was of 3-foot gauge, and the third-rail method of supply was 
 used, the two ordinary rails serving for the return. Auxil- 
 iary conductors were supplied to improve the conductivity 
 of the rails. 
 
 The road began running early in June, and after a suc- 
 cessful career of something less than a month was exhibited 
 at the Louisville Exposition during the later part of the same 
 year. By this time American inventors had settled down to 
 hard work on the electric railway problem, and soon pushed 
 on far in advance of their European competitors. 
 
 In the spring of 1883, too, Mr. C. J. Van Depoele began 
 work on electric traction, and laid down a track about 400 
 feet long, on which a single car was operated for several 
 weeks by means of a 3 horse-power motor. Later in the same 
 year a little overhead line was constructed by Mr. Van Depoele 
 for the Chicago State Fair, and remained in operation six or 
 seven weeks. 
 
 Late in the autumn of 1883 still another effort at an electric 
 road was made by Mr. Leo Daft. The experiments were tried 
 on the Saratoga and Mt. McGregor Railroad in November, 
 1883, and, as in the case just mentioned, an electric locomo- 
 tive was used. The motor was of about 12 horse-power, and
 
 346 THE ELECTRIC RAILWAY. 
 
 its best actual performance consisted in hauling a lo-ton 
 ordinary railway car loaded with 68 persons, at 8 miles per 
 hour, over a track having a grade of 93 feet to the mile. 
 The current was supplied by a third and central rail composed 
 of iron rails of 35 pounds to the yard laid on blocks of hard 
 wood saturated with resin. The current was taken off this 
 rail by a pair of phosphor-bronze contact wheels pressed 
 lightly down upon the rail by springs. The regulation of 
 speed was effected by commutating the field coils. 
 
 The next year, 1884, saw still other inventors in the field, 
 and before the end of the year considerable advance had been 
 made. During the summer Mr. Daft equipped a short line 
 on one of the piers at Coney Island, which ran during the 
 latter part of the season and carried in the aggregate 38,000 
 passengers. At about the same time, however, two notable 
 short roads were opened one of them on the underground 
 conduit principle, the other adopting the overhead trolley 
 wire now generally used. 
 
 The former road, installed in Cleveland, Ohio, by Bentley 
 and Knight, was actually thrown open to the public on 
 July 27, 1884, and was the first electric system to be actu- 
 ally operated in competition with horses on street railway 
 lines. The road thus equipped was about two miles long. 
 The motor employed was placed between the wheels, being 
 supported from the car body, and connection to the axles 
 was made by means of wire cables. The slotted conduit was 
 of wood placed between the rails, and of rather meager di- 
 mensions. The line was operated with quite uniform success 
 for about a year, when the Bentley- Knight Company trans- 
 ferred its headquarters to Providence, R. I. 
 
 A little later in the autumn of 1884 the first practical over- 
 head line on this side of the water was erected, the place 
 being the suburbs of Kansas City, Mo. ; and the inventor 
 whose ideas were embodied in the short experimental road 
 of a half mile in length was Mr. J. C. Henry. 
 
 This little road is notable as embodying some radical 
 changes from previous methods. It was known as the West- 
 port Electric Road, and was building during the summer of 
 1884. The overhead line consisted of two bare, hard-drawn
 
 HISTORICAL NOTES. 347 
 
 copper wires supported from the top, as is usually the case 
 with all trolley w r ires to-day. The size was No. i B. & S., 
 and at that time such hard-drawn wire was only put on the 
 market in 6o-foot lengths, so that the experimenters were 
 obliged to straighten it and braze the joints on the ground. 
 
 The motive power was supplied by a 5 horse-power Van 
 Depoele motor, which was connected to the car of standard 
 gauge by means of differential gearing not unlike that used 
 on lathes. There were five changes of speed. The motor 
 and gearing were supported together in a frame, one end of 
 which was connected to the car axle, the other to the car plat- 
 form. There was a bevel gear at the car axles and a clutch 
 connecting it to the motor, so that the armature could run 
 free if necessary. 
 
 Various experiments were tried in the way of different 
 systems of current supply and different sorts of trolleys. 
 The rails were temporarily bonded by driving horseshoe 
 nails between the fish plates and the rails themselves, and in 
 some cases one overhead wire was used merely as a feeder 
 for the other and the rail as return. The two overhead 
 wires were also sometimes used for separate tracks at turn- 
 outs and the rails again used as a return. Various descrip- 
 tions of trolleys supported by poles, and otherwise, were 
 tried, but the form finally adopted was a trolley running on 
 the line wires and with contact wheels bearing against their 
 sides, the whole being connected to the car by a flexible cable. 
 
 The generator was a series- wound dynamo intended for 
 arc lighting, of 10 horse-power capacity ; and the electromo- 
 tive force was several hundred volts in some of the experi- 
 ments ; where a dynamo of double the capacity was employed, 
 probably nearly 1,000. The steepest grade was 7 per cent., 
 and the car was handled upon it without any special diffi- 
 culty. 
 
 A few experiments were also tried with a motor of double 
 the horse-power, fitted on an open street car, which was used 
 for dragging a freight car on a short section of a neighboring 
 steam railroad fitted up for the purpose. The highest speed 
 attained was said to have been at the rate of 20 miles an hour. 
 
 This little Kansas City road was a nearer approach to the
 
 348 THE ELECTRIC RAILWAY. 
 
 forms familiar to us now than anything that had gone 
 before it. 
 
 Unfortunately, however, nothing^ important came of the 
 experiments, owing to the difficulty that inventors always 
 experience in getting the public to adopt their ideas. 
 
 The fundamental difficulty with the electric road at that 
 time was that it lacked public confidence ; even electricians 
 turned up their noses and looked askance at the new-fangled 
 substitute for the horse. The demonstrations made on a 
 small scale from time to time were not sufficiently convinc- 
 ing, so that it was an up-hill fight for those who were pos- 
 sessed of the courage of their convictions and determined to 
 demonstrate the practicability of electric traction. 
 
 Nevertheless, the work went on, and in 1885 very percep- 
 tible headway was made. In the first place, Mr. Daft, in 
 the early summer of that year, equipped electrically a portion 
 of the lines of the Baltimore Union Passenger Railway Com- 
 pany. The branch selected for this purpose was that con- 
 nectyig Hampden and Baltimore. It includes considerable 
 grades and curves, the heaviest grade being nearly 7 per 
 cent., while the radii of some of the curves were as low as 50 
 feet. The system of supply selected was the third rail, as 
 in Mr. Daft's earlier roads. 
 
 To this end a 2 5 -pound T-rail was laid on special insulators 
 and formed the outgoing lead, the return being through the 
 two outer rails together with the ground. The electromotive 
 force used was but 125 volts, which enabled a tolerable degree 
 of insulation to be kept up. The electric locomotive contained 
 a single 8 horse-power motor weighing 1,000 pounds. A 
 single gear reduction was employed, and the 3 -inch phosphor- 
 bronze armature pinion engaged an internal gear 27 inches 
 in diameter keyed on the axle of the driving wheels. The 
 total weight of the locomotive was about 4,200 pounds. The 
 regulation, as in Mr. Daft's previous motors, was secured by 
 the commutated field method. 
 
 On August 8, 1885, the road went into operation, and 
 the original motors were in service until quite recently. 
 Four electric locomotives were installed at various times 
 during the first year of operation. At street crossings the
 
 HISTORICAL NOTES. 349 
 
 third rail was abandoned, and a bare overhead wire carried 
 the current, from which it was obtained by a trolley pole 
 placed on top of the car much as at present, carrying on its 
 end the contact brush. 
 
 About the same time that this road went into operation 
 Mr. Van Depcele equipped a short line at Toronto, Canada, 
 with a single motor car drawing three trailers. Here an 
 overhead line was employed all the way, the bare wire being 
 supported over the car from brackets at the side of the track. 
 An underrunning contact much like that just mentioned in 
 connection with the Baltimore road was used. 
 
 Meanwhile Mr. F. J. Sprague entered the field, and, 
 profiting by the experience of others, spent a large amount 
 of time in working up the details of electric traction ap- 
 paratus. His first effort was the application of electric- 
 ity to the New York Elevated Railroad system, and in 
 December, 1885, he publicly discussed the matter. In the 
 early part of the succeeding year he was busily engaged on 
 the problem and carried out an extensive series of experi- 
 ments on a section of the elevated track. 
 
 Though this attempt at the elevated railroad came to noth- 
 ing, it nevertheless taught many a valuable lesson regarding 
 the handling of suitable power units and the precautions that 
 would have to be taken in railway work. 
 
 During 1886 and 1887 Mr. Sprague was steadily at work on 
 the electric traction problem, and the features which he 
 brought into special prominence were the method of sus- 
 pending the motors now generally used and the development 
 of the overhead underrunning trolley into something like its 
 present shape. The method of suspension employed was, as 
 is well known, to pivot one end of the motor directly upon 
 the car axle, sustaining the other by springs, so that the 
 center of the armature shaft would preserve an invariable 
 distance from the center of the axle, thereby securing the 
 proper meshing of the gears. At first only a single gear 
 reduction was used, but afterward the intermediate shaft 
 made its appearance, to be displaced only in the practice of 
 the last year. 
 
 This particular method of support was new in electric
 
 350 
 
 THE ELECTRIC RAILWAY. 
 
 motors, although it is said to have been anticipated abroad by 
 the same method patented and used for gas motors. 
 
 It was not until almost the beginning of 1888 that the first 
 Sprague road was actually put into operation a small line 
 at St. Joseph, Mo. 
 
 By January i, 1888, there were in operation in the United 
 States and Canada 13 electric roads, operating 95 motor 
 cars over 48 miles of track. They were as follows: 6 Van 
 Depoele roads, 3 Daft, i Fisher, i Short, i Henry, and the 
 above-mentioned Sprague road at St. Joseph, the most ex- 
 tensive of these being the Daft road at Asbury Park, N. J., 
 where eighteen cars were in operation. Electricians had 
 not been idle, but the electric road was not yet fairly 
 established. 
 
 During the spring of 1887 the Union Passenger Rail- 
 way, of Richmond, Va., determined to adopt electricity as a 
 motive power, and Mr. Sprague undertook the equipment 
 of the line. This was successfully accomplished, and the 
 road went into operation, after a brief period of experi- 
 mental use, early in 1888. It was the first of the important 
 electric roads equipped in this country or elsewhere, and 
 gave an impetus to electric traction that is unique even in 
 the progress of this nineteenth century. 
 
 Thirteen miles of line were equipped, and during the 
 spring of iSSS, on an average, 20 cars were employed. 
 There were six 4O,ooo-watt, 5oo-volt Edison dynamos; three 
 125 horse-power engines, and three 125 horse-power boilers. 
 Although the apparatus was crude and the road underwent 
 many vicissitudes, the present success of the electric road is 
 very largely due to the perseverance and the energy that put 
 the Richmond road in operation in the face of every sort of 
 discouragement and gloomy prediction of failure. From that 
 time on electric railways have grown in number and size so 
 rapidly as to almost defy any attempt to record them. Dur- 
 ing the succeeding year a number of additional roads were 
 installed, and the Thomson-Houston Company, which had 
 acquired the Bentley-Knight and Van Depoele patents, en- 
 tered the field. 
 
 Since that time other electricians have been busy. The
 
 HISTORICAL NOTES. 351 
 
 Sprague Electric Railway and Motor Company came into the 
 possession of the Edison General Electric Company, new 
 companies have arisen, and electric roads have been installed 
 so rapidly and in such numbers as to occupy to the fullest the 
 resources of the several companies that have been engaged 
 in the work. 
 
 This brief record of the history of the electric railway 
 makes no claim to completeness; indeed, a complete history 
 would be hard to write until the position of the various 
 inventors shall be officially determined by the courts. The 
 name of electric railway patents is legion, but their value is 
 as yet an indeterminate quantity with a tendency, however, 
 toward a zero value, in view of the early anticipations both 
 of some of the most essential features of modern electric 
 traction and the fact that many of the patented devices have 
 existed only upon paper. Perhaps it has been well for the 
 development of this branch of the electrical transmission of 
 energy that such is the case. 
 
 We have made no attempt to mention all those who have 
 taken a part in the development of the electric railway, or 
 all the various attempts that inventors have made to solve 
 the many difficulties that have been encountered. It has 
 been our endeavor, however, to sketch out, with more atten- 
 tion to general accuracy of perspective than to minuteness of 
 detail, the growth of the art that has been already productive 
 of such important results and promises an almost immeasura- 
 ble future.
 
 APPENDIX A. 
 
 ELECTRIC RAILWAY VERSUS TELEPHONE DECISIONS 
 IN THE SUPREME COURT OF OHIO. 
 
 JANUARY TERM, 1891. 
 
 THE CINCINNATI INCLINED PLANE 
 RAILWAY COMPANY, 
 
 Plaintiff in Error, 
 
 against 
 
 THE CITY AND SUBURBAN TELE- 
 GRAPH ASSOCIATION, 
 
 Defendant in Error. 
 
 ELECTRIC STREET RAILWAYS SINGLE-TROLLEY OVERHEAD SYSTEM 
 RIGHTS OF TELEPHONE COMPANIES. 
 
 Syllabus. 
 
 1. The dominant purpose for which streets in a municipality are 
 dedicated and opened is to facilitate public travel and transportation, 
 and in that view, new and improved modes of conveyance by street 
 railways are by law authorized to be constructed, and a franchise 
 granted to a telephone company of constructing and operating its lines 
 along and upon such streets is subordinate to the rights of the public 
 in the'streets for the purpose of travel and transportation. 
 
 2. The fact that a telephone company acquired and entered upon the 
 exercise of a franchise to erect and maintain its telephone poles and 
 wires upon the streets of a city, prior to the operation of an electric 
 railway thereon, will not give the telephone company, in the use of 
 the streets, a right paramount to the easement of the public to adopt 
 and use the best and most approved mode of travel thereon, and if the 
 operation of the street railway by electricity as the motive power tends 
 to disturb the working of the telephone system, the remedy of the 
 telephone company will be to readjust its methods to meet the con- 
 dition created by the introduction of electromotive power upon the 
 street railway. 
 
 23 353
 
 354 THE ELECTRIC RAILWAY. 
 
 3. Where a telephone company, under authority derived from the 
 statute, places its poles and wires in the streets of a municipality, and 
 in order to make a complete electric circuit for the transmission of 
 telephonic messages, uses the earth, or what is known as the " ground 
 circuit," fora return current of electricity, and where an electric street 
 railway afterward constructed upon the same streets is operated 
 with the " single-trolley overhead system" so called of which the 
 ground circuit is a constituent part, if the use of the ground circuit in 
 the operation of the street railway interferes with telephone communi- 
 cation, the telephone company, as against the street railway, will not 
 have a vested interest and exclusive right in and to the use of. the 
 ground circuit as a part of the telephone system. (Decided Tuesday, 
 June 2, 1891.) 
 
 But it is urged that the franchise of the telegraph association to 
 construct lines of telephone is greatly impaired by reason of the single- 
 trolley railway using a grounded circuit, whereby a large part of the 
 electric current flows off from the rails to the surrounding earth, and 
 to and upon all telephone wires which may be connected with the 
 earth in proximity to the railway. The action is described as conduc- 
 tion, causing more or less of electric current to be poured into the 
 earth and into all electric conductors connected with the earth, thereby 
 reaching telephone wires in a grounded circuit, and creating loud and 
 continuous noises upon the wires, which disturb telephonic communi- 
 cation. This disturbance, however, results not solely from the earth 
 circuit of the railway company, but also from the fact that the defend- 
 ant in error likewise relies upon the earth for its return circuit, by 
 connecting with the earth the end of its wire furthest from its electric 
 batteries. 
 
 It is claimed that in addition to this conduction or leakage disturb- 
 ance the single-trolley electric railway introduces serious disturb- 
 ances on telephone lines by induction, for the reason that such electric 
 railways employ large wires to convey the current used for the pro- 
 pulsion of their cars, and this current is constantly and rapidly chang- 
 ing its strength ; that these rapidly changing currents in the electric 
 railway wires induce disturbing currents in parallel telephone wires 
 near which the electric railways have been built, and thus prevent a 
 successful transmission of telephonic messages. 
 
 These interferences with the telephone service may be obviated, 
 it is stated, by the railway company giving up the single-trolley system 
 with the ground circuit, and substituting the double-trolley system 
 with its two trolley wires, two trolley wheels, and electric current 
 passing from one wire through one trolley, through the motor, back 
 through the other trolley to the other wire, and so back to the genera- 
 tor, without escaping to the earth. The grounded circuit, it is insisted, 
 should be abandoned and surrendered to the sole use and service of 
 the defendant in error. But it is admitted that other remedies of
 
 APPENDIX A. 
 
 355 
 
 the telephone disturbances may be easily obtained by constructing the 
 telephone with a complete metallic circuit, or by resort to what is 
 known as the McCluer device, consisting of a single return wire, to 
 which a number of telephone wires are attached. 
 
 Conceding that the mode adopted by the railway company of pro- 
 pelling its cars by electricity is an interruption to the telephone service 
 of the defendant in error and calculated to impair its franchise in the 
 manner contended, the inquiry is suggested whether the railway com- 
 pany must yield up a useful franchise that the same may be exclusively 
 enjoyed by the telegraph association, or whether the association shall 
 adapt its system to existing conditions ; whether the company shall 
 change from the single to the double trolley system, from the 
 grounded to the metallic circuit, or whether the association shall 
 either use a complete metallic circuit or resort to the McCluer device. 
 
 It is immaterial on which party the expense of the change may fall 
 the more heavily. It is a question of legal right. 
 
 When the telegraph association erected its poles and lines in 1881 
 and 1882, with the design of conducting the business of a telephone 
 company, it found the railway company operating its street railway, 
 with authority under the statute to use other motive power than ani- 
 mals, to wit, electricity, cable, or compressed air, upon obtaining the 
 consent of the Board of Public Works. The telephone business was not 
 among the probabilities when the streets of Cincinnati now made use 
 of by the telegraph association were dedicated or condemned for the 
 public use. The primary and dominant purpose of their establishment 
 was to facilitate travel and transportation ; they belong from side to 
 side and end to end to the public, that the public may enjoy the right 
 of traveling and transporting their goods over them. The telephone 
 poles and wires and other appliances are not among the original and 
 primary objects for which streets are opened, for they may be placed 
 elsewhere than on the highways and yet accomplish their purpose. 
 In Taggart v. Street Railway Co., 16 R. L, it was said by Durfee, C. J., 
 that telephone poles and wires are not used to facilitate the use of 
 the streets for travel and transportation ; " whereas the poles and 
 wires of the railway company are directly ancillary to the use of the 
 streets as such, in that they communicate the power by which the 
 street cars are propelled." As a general rule, an occupation of the 
 streets otherwise than for travel and transportation is presumptively 
 inferior and subservient to the dominant easement of the public for 
 highway purposes, for if not so, the primary object of their dedication 
 or appropriation might be largely defeated. And the fact that per- 
 mission is granted to occupy the streets or highways for a purpose 
 other than travel, does not confer a prior and paramount right to 
 occupy them to the exclusion of their use for travel in a mode different 
 from what obtained when such permission was given. 
 
 To those improved agencies, devised for the convenience and ad-
 
 356 ' THE ELECTRIC RAILWAY. 
 
 vantage of the community in general, the franchise of the telephone 
 company to occupy the streets for carrying on its business must be 
 secondary and subordinate. Whether all who go upon the streets shall 
 have the most convenient and expeditious passage and carriage of 
 person and goods, has not been made dependent upon the manner in 
 which the defendant in error has preferred to locate its poles, stretch 
 its telephone wires, or form the electric circuit. 
 
 It is in recognition and maintenance of the superior easement of the 
 public in the streets that city councils are required to " cause the same 
 to be kept open and in repair and free from nuisance ;" that the streets 
 are graded and paved and proper regulations of police provided to 
 govern the actions of persons using them ; that the abutting owner, 
 though having a peculiar interest and easement in the adjacent street 
 appendant to his lot, has no right to place permanent obstructions in 
 the street, nor do any act on his own land outside the limits of the 
 street, that will make the way inconvenient or hazardous or less secure 
 than it was left by the municipal authorities. 
 
 This paramount easement or estate which the public acquires in the 
 streets, carrying with it a special interest in the adoption of the most 
 approved systems of modern street travel, cannot be made subservient 
 to the telegraph or telephone when admitted on the highway, without 
 the clearest expression of the legislative will. 
 
 The demand made by the telegraph association is, not that the rail- 
 way company shall so modify its existing electrical apparatus as not 
 to interfere with the telephone service, but shall forever abandon the 
 use of an essential part of its electromotive system, or be perpetually 
 enjoined. In other words, the association claims the exclusive use of 
 the grounded circuit, inasmuch as the mechanism of the telephone is. 
 so complex and the electric currents employed so delicate and sensitive 
 that they cannot be used without disturbance from the heavier cur- 
 rents employed by neighboring electrical enterprises that operate with 
 the grounded circuit. We find no foundation for such an exclusive 
 franchise or right. 
 
 When the telegraph association began its operations under the tele- 
 phone system, neither the statute authorizing it to erect and maintain 
 poles, wires and other necessary fixtures, nor the ordinance under 
 which it obtained the power to extend its lines in the streets, gave an 
 exclusive right either to use the earth for a return circuit or a com- 
 plete metallic circuit formed by double wires. The Legislature did not 
 grant the right by general enactment, nor was the municipal corpora- 
 tion empowered by the Legislature to give the telegraph association 
 the exclusive right to make use of its streets so as to create a monopoly. 
 
 It is contended, however, that the defendant in error, by virtue of 
 its grants, acquired, before the railway company had a right to use 
 electricity as a motive power, a vested interest in the telephone sys- 
 tem as it now operates it, with a grounded circuit, and that not even
 
 APPENDIX A. 357 
 
 the Legislature of the State could take away from it or injure this fran- 
 chise, on the faith of which it has expended its capital and labor. 
 Special privileges or immunities are under the control of the Legislat- 
 ure. If granted they may be altered, revoked, or repealed by the 
 General Assembly. Art. i, Sec. 2 of the Constitution. And while cor- 
 porations with valuable franchises may be formed under general laws, 
 all such laws may, from time to time, be altered or repealed. Consti- 
 tution, Art. 13, Sec. 2. In view of these constitutional provisions, it is 
 clearly within the power of the General Assembly to authorize one 
 class of corporations to use, in the streets, electricity with the 
 grounded circuit as a motive power, and another class to employ the 
 same or a similar agency for the transmission of telegraphic or tele- 
 phonic messages. And if the proper exercise of the rights granted to 
 the one class under general law is irreconcilable and plainly interferes 
 with a prior grant to a corporation of the other class, it may be con- 
 strued as the intention of the Legislature to deny an exclusive fran- 
 chise, if not to repeal the antecedent grant. 
 
 It is contended, however, in behalf of the defendant in error, that 
 conceding the railway company and telegraph association to be upon 
 an equal footing on the streets and highways in the enjoyment of their 
 respective franchises, the company is bound to conform to the rule sic 
 utere tuo ut alienum non laedas. In the view which we take of the rela- 
 tion to each other of the parties to the action, we deem it unnecessary 
 to inquire whether there has been a want of conformity, and to what 
 extent, if any, on the part of the railway company, to the requirements 
 of the legal maxim. Nor do we think it necessary to determine how 
 far an incorporated company making a lawful and careful use of its 
 own property, or of a franchise granted to it by the proper municipal 
 authorities, may be held liable for damages incidentally caused to 
 another. 
 
 From the undisputed facts in the case, as disclosed in the record and 
 printed arguments of counsel, it is evident, as we have already seen, 
 that the railway company acquired from the State and from the city of 
 Cincinnati authority to erect and maintain poles and wires in the 
 streets or highways, and to use electricity as a motive power for its 
 cars. Clothed with such authority, we have, upon weighing the- alle- 
 gations .in the original petition and applying to them the well-settled 
 principles governing the legal rights of the pfcblic in the highways, 
 reached the conclusion that the facts set forth in the petition are not 
 sufficient to constitute a cause of action. We are of the opinion that 
 there has been no invasion of the rights of the telegraph association 
 by the plaintiff in error, and that the telegraph association is not en- 
 titled to the relief prayed for in its petition. The judgment, therefore, 
 of the Superior Court at general and special term must be reversed 
 and the original petition dismissed. Judgment accordingly.
 
 THE ELECTRIC RAILWAY. 
 
 SUPERIOR COURT OF CINCINNATI. 
 
 THE CITY AND SUBURBAN TELE- 
 GRAPH ASSOCIATION 
 
 THE CINCINNATI INCLINED PLANE 
 RAILWAY COMPANY. 
 
 Taft, J. This is an action to enjoin the defendant from using a sys- 
 tem of electric railway propulsion, known as the single-trolley system, 
 in running its cars upon the streets of this city and the roads of the 
 county, because, as plaintiff claims, it does a great injury to the con- 
 duct of the public business in which it is engaged i.e., the maintenance 
 for profit of telephone communication between a large number of 
 subscribers in the city and county. 
 
 Plaintiff alleges that it has been conducting its business under lawful 
 grants of the Legislature of the State and the municipal authorities, 
 and by virtue thereof has lawfully erected many lines of telephone 
 wires on the streets upon which the street railway of defendant is 
 constructed and operated ; that the defendant has no lawful authority 
 to use electricity as a propulsive power for its cars ; that the single- 
 trolley system in use by defendant conducts upon and induces upon 
 the wires of the plaintiff erected on the same streets currents of elec- 
 tricity, which makes it impossible to use those wires for the purpose 
 of telephone communication ; that the result is that many of the sub- 
 scribers of plaintiff have made complaint and threatened to discontinue 
 the use of the telephone, and some have refused to continue payment 
 for the same, all to the great injury of the plaintiff , that the damage 
 is irreparable and continuous, and that plaintiff notified defendant be- 
 fore the erection of the single-trolley system that it would result in a 
 serious injury to plaintiff, and of the objection to the system. 
 
 The defendant answers, denying plaintiff's lawful right to. operate 
 telephone lines; avers its own lawful right to conduct a system of 
 electrical railway propulsion ; avers that the single-trolley system is 
 the only practicable one in use and secures advantages to the defendant 
 not afforded by any other; that plaintiff's difficulties arise from the 
 defects of its own telephone system, in that, for economy, it uses the 
 earth as a return conductor, and so invites onto its lines many disturb- 
 ing currents, both natural and artificial, although the mechanism of 
 the telephone is so delicate that any proper system would make the 
 return circuit metallic; that plaintiff's claim amounts to a claim to
 
 APPENDIX A. 359 
 
 \ 
 
 the exclusive use of the earth as a conductor, which is without war- 
 rant of law ; that the use by the defendant of the earth as a return 
 circuit is a material assistance in propelling cars up heavy grades; 
 that it could not alter its system so as to avoid the use of the earth as 
 a return circuit except at great outlay and loss of efficiency. The 
 answer denies that the operation of its road produces interference 
 with the telephone lines of plaintiff, and says that if any injury has 
 been produced for which defendant is liable, the damages are capable 
 of exact ascertainment, because they can all be remedied by an addi- 
 tional expenditure by the plaintiff, and that therefore there is no 
 ground for equitable relief. 
 
 Th^re can be no doubt that' the plaintiff has a lawful right to occupy 
 the streets upon which defendant's track is laid, with the poles and 
 wires erected for the conduct of its business, and that the use of the 
 earth as a return is not a violation of its franchise. It was organized as 
 and still is called a telegraph company. It began its telephone busi- 
 ness before Section 3,471 was passed, and perhaps some of its poles and 
 lines along the streets occupied by defendant were erected for such 
 use. But this circumstance is immaterial, for it has been held, both 
 in England and in this country, that the telephone system, which is 
 communication over long and short distances through the agency of 
 currents of electricity on metal wires, is really a magnetic-telegraph 
 system, and that an exclusive right to operate the telegraph includes 
 the exclusive right to operate telephone lines. Attorney General v, 
 Edison Telephone Co., 6 Q. B. D. , 244; Wisconsin Tel. Co. v. Oshkosh, 
 62 Wis., 32. 
 
 Having determined, then, that plaintiff's use of the streets on which 
 defendant operates its road is lawful, the next question which arises 
 is whether defendant has the right to use electricity for the propulsion 
 of its cars. If it has not and it thereby interferes with plaintiff's 
 franchise, it is very clear that such interference gives plaintiff a right 
 of action. 
 
 On the whole, then, I am of opinion that the Legislature conferred 
 the right upon defendant to use any other motive power than animal 
 whenever the Board of Public Works should consent. Now, the board 
 did consent, on October 24, 1885, that defendant should use either a 
 cable, compressed air, or electricity. It has chosen electricity, and 
 has produced the necessary authority to erect its poles and string its 
 wires. 
 
 Such being the condition of the franchises which the plaintiff and 
 defendant are lawfully entitled to enjoy, considered each without ref- 
 erence to the other, it becomes necessary to inquire, first, whether any 
 loss has been inflicted upon plaintiff by defendant, and if so, how has 
 it occurred ; second, whether such loss, if any, is justified by defend- 
 ant's franchise, so as to be damnum absque injuria, which involves the 
 preliminary question whether the Legislature, after having given the
 
 5 6o THE ELECTRIC RAILWAY. 
 
 plaintiff the right to construct its telephone system, on the faith of 
 which right it has expended large amounts, can confer a franchise on 
 another, the exercise of which will seriously impair the plaintiff's 
 franchise as heretofore enjoyed. 
 
 First, a current of electricity cannot be produced without a circuit 
 that is, unless the negative and positive poles of the generating battery 
 or instrument are connected by a continuous substance capable of 
 conducting the current. Such a substance may be a metal wire, or 
 if both poles of the battery be connected with the earth by metal 
 wires the current will find a circuit through the wires and the earth. 
 The earth, by reason of its immense mass, makes an excellent con- 
 ductor. By what paths the current discharged into it over the wire 
 from one pole finds its way throughout to the wire from the other pole 
 is not capable of determination. 
 
 The telephone is a mechanism by which the sound of human speech 
 is reproduced over long distances. Without attempting to describe the 
 exact mode in which this result is brought about, I may say that the 
 sound waves of the human voice produce vibrations in a thin ferrotype 
 plate, which, by means of a magnet and an induction coil, are con- 
 verted into corresponding vibrations in an electric current on the con- 
 necting wire, and these vibrations are, in turn, by means of the 
 inducing coil and magnet at the other end, converted into exactly 
 corresponding vibrations on a plate there, reproducing the sound 
 waves of the voice of the speaker in such a manner as to enable the 
 receiver to understand. The current on the connecting wire is a 
 slight one, and the circuit is completed, not by a return wire, but by 
 a ground wire brought into contact with the earth. This contact is 
 usually made by attaching the wire from the negative pole of the 
 single cell battery in each telephone to a gas pipe or water pipe 
 running down into the earth. 
 
 In the Sprague system of electric railways, which is the kind used 
 by the defendant, the electricity used to operate the motors under the 
 cars is conveyed to them by a single overhead wire suspended over 
 the middle of the track, along the under side of which runs a trolley 
 wheel on a single mast attached to the car, making electric connection 
 between the overhead wire and the motor of the car, and allowing the 
 current to pass down through the motor and onto the track, whence 
 some of it returns directly to the dynamo generator at the power 
 house. A large part of the electricity leaves the track, however, and 
 by other and various paths also finds its way through the earth back 
 to the dynamo. 
 
 In addition to the overhead trolley wire, which is supported by guide 
 wires from iron posts erected on the curb at regular intervals, there 
 is what is called a feed wire strung along on these posts for the pur- 
 pose of keeping up the required quantity of electricity on the trolley 
 wires. On the streets where are telephone wires and electric railway
 
 APPENDIX A. 361 
 
 wires their general course must be parallel. The evidence in the 
 case establishes beyond a doubt that since the defendant began the 
 operation of its line by electricity in June, 1889, down to the time of 
 trial, the usefulness of all the telephones along the line of the railway 
 has been more or less impaired; that in many cases the buzzing 
 noise which seems to be the chief form of the disturbance has been 
 so loud and continuous that communication over the lines has been 
 impossible. Nor has the disturbance been confined to telephones on 
 the line of the railway. Telephones several miles from the city whose 
 connecting lines ran parallel for any distance with the railway were 
 similarly affected and the buzzing noise in many of these seems to 
 have been quite as deafening as in telephones along the line. Alto- 
 gether more than two hundred lines have been affected in this way, to 
 a greater or less degree. The cause of the trouble is undoubtedly 
 the operation of the electric railways, and the way by which it is 
 brought about is twofold. First, the escape of the electric fluid from 
 the rails, which is called earth distribution or leakage, near where the 
 wire from the telephone is connected with the earth, brings upon this 
 earth-connection wire of the telephone varying currents of electricity 
 of much greater quantity than that necessary for the telephone current, 
 and produces upon the magnet and inducing coil an effect which results 
 in vibrations of a very different character from those produced by the 
 . human voice, and makes a noise like the buzzing of a saw. Second, a 
 similar noise is made by induction. It is a physical fact of much im- 
 portance in electric mechanism that where two wires of two circuits 
 are parallel to each other, and there is a current of varying intensity 
 on one of them, this will produce in the other, in the opposite direc- 
 tion, a current of electricity of similar variation. The insulation of 
 the wires has no effect to reduce the current produced in this way. 
 The amount of induction depends upon variation in the current, the 
 distance of the wires from each other, and the length of the parallelism 
 of the wires. The current upon the trolley wire and the feed wire of 
 the railway is quite variable in quantity and intensity, owing to the 
 drain upon the store of electricity by the moving and stopping of the 
 car. Nor is the electricity, as generated, exactly uniform in its flow 
 from the dynamo. The result is that wherever the telephone wire 
 is parallel with the trolley wire and the feed wire, there is induced 
 upon the telephone wire a current whose variations correspond with 
 the variations of the electrical current on the electric railway wires, 
 and this, acting upon the inducing coil and magnet, produces vibra- 
 tions of the pin plate, which makes the buzzing sound. It is not possi- 
 ble, in listening to the sounds produced by the electric railway, to 
 say whether it is the result of induction or earth leakage. Some of 
 plaintiff's subscribers, notably Proctor & Gamble, have loud buzzing 
 sounds upon their telephones, the ground wires of which are at such a 
 distance from the railway track as to make it quite unlikely that the
 
 362 THE ELECTRIC RAILWAY. 
 
 disturbance could have been caused by earth leakage, while, on the 
 other hand, their wires are for some distance parallel with the trolley 
 and feed wire of the railway company. Other telephones are dis- 
 turbed on the line of the road, though their wires are not parallel with 
 the trolley wire. Expert evidence attributed the disturbance about one- 
 half to induction and one-half to conduction or earth leakage. This, 
 of course, is only a rough estimate, and the fact may vary much in 
 particular instances. The injury to plaintiff's service is produced, then, 
 speaking in a general way, one-half by the defendant pouring elec- 
 tricity into the earth, which finds its way into the property of plain- 
 tiff's subscribers, and thence into the wire of the telephone, and one- 
 half by a creation of a current on plaintiff's wires in the street by the 
 parallelism of defendant's wire and the varying character of the cur- 
 rent. The result has been a very substantial interference with the 
 plaintiff's business, and if the present condition continues it will 
 end in a serious loss. 
 
 Is it a loss for which defendant is liable? The contention on behalf 
 of the defendant is that because it has full power to operate by elec- 
 tricity under the law, the loss resulting to the plaintiff is dammun 
 absque injuria, and if the plaintiff wishes to avoid the loss, it must 
 adopt safeguards in the shape of a metallic circuit to avoid the diffi- 
 culty. To this plaintiff replies that by virtue of its grants it ac- 
 quired, before the defendant had a right to use electricity as a motive, 
 power, a vested interest in the telephone system as it now operates it, 
 with a grounded circuit, and that not even the Legislature of the State 
 could take away from it or injure this franchise, on the faith of which 
 it has expended so much capital and labor. 
 
 I am inclined to think that under the constitutional provision that 
 all laws for the forming of corporations may be altered or repealed 
 (Sec. 2, Art. 13), it would be in the power of the Legislature to grant 
 a right to other corporations for a public use to so use the street as to 
 require the plaintiff company, if it wished to continue in the telephone 
 business, to change its system, and that without any right of action 
 against such corporations. The case of Ry. Co. r. Ry. Co., 30 Ohio 
 St., 604, shows that where one railway company condemned a right of 
 way across the track of another, that other cannot recover for an 
 injury to its franchise as a railroad or for the increased expense en- 
 tailed upon it in obeying the laws of the State with reference to rail- 
 road crossings. However this may be, I am very clear that no intention 
 on the part of the Legislature to abridge the granted rights of one cor- 
 poration by a new grant to another will be recognized by the courts 
 unless such intention plainly appears in the law. In England, the 
 power of Parliament is unlimited, and it may even confiscate private 
 property, and a fortiori may abridge and destroy chartered rights and 
 franchises. Nevertheless, we find in that country that where one cor- 
 poration is granted a right which may be so exercised as to injure or
 
 APPENDIX A. 363 
 
 interfere with a right previously granted to another, the presumption 
 of law is that Parliament intended only such uses as were consistent 
 with the rights of the first corporation. In Gas Light and Coke Co. v. 
 Vestry, 15 Q. B. D., i, plaintiff was a gas company which had laid its 
 gas pipes, by virtue of a public grant, under a street which the defend- 
 ant, a public corporation, was charged with keeping in repair, and 
 upon which it used such heavy rollers as to injure the pipes of the 
 plaintiff. The rolls used were economical and well fitted for the pur- 
 pose, but it was held that unless the defendants were expressly 
 authorized by statute to use rollers of the size and weight of those 
 which did the injury, the defendant could not justify under a duty to 
 keep in repair which might be discharged by rollers of less weight 
 and without breaking the pipes. 
 
 This case is peculiarly applicable to the case at bar, because here 
 was a case of a public grant to the gas company enjoyed in a certain 
 way, followed by a grant to the defendant to exercise another right, 
 which, if exercised in one way, would injure plaintiff's enjoyment of 
 its right, and which, if exercised in another, would not. The same 
 principle is here applied that courts recognize wherever private prop- 
 erty is injured by the exercise of authority granted by the Parliament. 
 See Hammersmith, etc., Ry. Co. v. Brand, L. R. 4 H. L., 171 ; Queen 
 v. Bradford Navigation Co., 6 B. & S., 631 ; Geddis v. Proprietors of 
 Bann Reservoir, 3 App. Cases, 430; Att'y Gen'l r. Colney Hatch Lu- 
 natic Asylum, L. R. 4 Ch. App., 416; Att'y Gen'l r. Gas Light and Coke 
 Co., 7 Ch. D., 217; Managers Metrop. Asylum District v. Hill, 6 App. 
 Cases, 193. 
 
 Even, then, if franchises to occupy the public streets conferred by the 
 Legislature may be subject to modifications of their use and enjoyment 
 by other public grants of the Legislature, it is certain that unless the 
 legislative intent to make such modification clearly appears, either by 
 express words or by necessary implication arising from the impossibil- 
 ity of enjoying the second grant without such modification, it will not 
 be inferred. 
 
 But it is said that this principle can have no place here because the 
 right to occupy the street for the purpose of travel is a superior right 
 to that of using it for the telephone poles. Defendant's counsel, to 
 establish this, rely on The Spring Grove Ave. v. Cumminsville, 14 
 Ohio St. : Smith?'. Tel. Co., 2 C. C. R., 259; Mt. Adams & E. P. R. Co. v. 
 Winslow, 3 C. C. R., 425; R. R. Co. v. Williams, 35 Ohio St., 171; 
 Pike's Ex'rs v. Western U. Tel. Co., decided by this Court. 
 
 These cases have no bearing upon this case at all, as it seems to me. 
 They involved a discussion of whether the erection of certain struct- 
 ures in the street could be considered a new burden and use, not 
 included in the easement which the abutting property holder had 
 originally granted to the public, and whether therefore he was entitled 
 to compensation. It is unquestionably within the power of the Legis-
 
 264 THE ELECTRIC RAILWAY. 
 
 lature, so far as the public is concerned, to enlarge the benefit to be 
 derived from the streets, so as to include other public purposes than 
 those of mere travel (Ry. Co. v. Lawrence, 38 Ohio St., 45). If this 
 takes more from the abutting property holder than he originally gave, 
 then he may have compensation, but the public cannot complain, for 
 their representative, the Legislature, has spoken and granted the use. 
 When the telephone company is granted the right to use the streets, 
 its right is as well founded as that of a street railway company, and in 
 the absence of express legislative direction to the contrary, there is 
 to be no yielding to any other. The provision that the telephone and 
 telegraph lines shall not incommodate the public in the use of the 
 street, in Section 3,461, does not help the defendant. The inconven- 
 ience must be determined when the enjoyment of the franchise is 
 entered upon. 
 
 After rights have been acquired by the outlay of capital and user, 
 there must be express legislative sanction, at least, to warrant a 
 court in finding a use of the street to be an interference with public 
 travel which was not so when it began. It should be noted in this 
 connection, too, that the plaintiff performs a very important quasi 
 public duty, and is in fact a common carrier of messages. It is given 
 the power of condemnation on that ground alone. Pierce z 1 . Drew, 136 
 Mass., 75; Hackett r. State, 105 Ind., 250. 
 
 Coming now to apply the principle just under discussion to the case 
 at bar, what do we find ? For ten years the plaintiff has exercised the 
 franchise of occupying the streets along defendant's lines with its 
 poles and wires, conducting a telephone business with a single wire 
 circuit and an earth return. This mode of return was universally 
 employed when it began, and is to-day in general use. It has con- 
 structed a valuable plant, many parts of which will have to be changed 
 at great expense if it is to adopt the only system which will obviate 
 the difficulty it now encounters from the operation of defendant's rail- 
 way. I refer to the metallic circuit. It is admitted by every one that 
 if the telephone company were to make every one of its lines a com- 
 plete metallic circuit, with the return wire parallel with the outgoing 
 wire, the disturbance both from induction and earth leakage would 
 be completely removed. It is obvious that if the circuit never came 
 into contact with the earth, the electricity dumped into the ground by 
 defendant could not reach the telephone wire, and so no disturbance 
 could arise there. And it is also a well-ascertained fact that if the 
 two wires of the circuit, the outgoing and the return, are parallel an<- 
 the same length, no effect will be perceptible from induction by r, 
 third parallel wire of another circuit, however variable may be the 
 current on that wire. This is because the induction which actually 
 takes place upon each of the wires of the circuit results in currents 
 of equal intensity and variability in opposite directions, which being 
 on the same circuit exactly neutralize each other.
 
 APPENDIX A. 365 
 
 To construct such a circuit for every subscriber would entail an 
 expense upon plaintiffs, perhaps, even greater than the original out- 
 lay. Nor does it seem practicable to have metallic circuits in disturbed 
 districts and allow the other lines to remain grounded, because of the 
 difficulty in arranging the switchboards at the exchange by which sub- 
 scribers are connected with each other, and for the further reason that 
 in order to connect a metallic circuit with a grounded circuit, the re- 
 turn wire of the metallic circuit must be grounded at the exchanges. 
 The result is that the circuit then made between the two subscribers 
 is a grounded circuit with a long leg and a short leg of wire parallel 
 in the disturbed district. The difference in the length of the two 
 wires prevents the neutralizing effect of the induced current in one 
 wire upon the opposite induced current in the other, because, by 
 reason of the different lengths, that part of the induced currents said 
 to be produced by static induction is unequal. Moreover, to put up 
 metallic circuits through the disturbed districts and to make the 
 necessary alterations in the switchboard would entail an expense, not 
 nearly so heavy, of course, as a complete conversion into a metallic 
 circuit, but which would be quite substantial. There remains to con- 
 sider what is called the McCluer device, which consists in a large wire 
 carried into the disturbed district with which are connected all the 
 return wires of the telephones. This main is grounded at a point 
 away from danger of earth leakage. It is likely that such a device 
 would get rid of earth leakage, but it would have no practical preven- 
 tive effect upon the disturbances by induction, which, as I have said, 
 make up about half of the troubles. The expense of such a device 
 would not be large, still it would be something. It follows, from 
 what has been said, that if the defendant is permitted to use its pres- 
 ent single-trolley system, it will require the plaintiff either to lose the 
 business of many subscribers, or it will have to go to an enormous 
 expense to obviate all difficulty or to a less expense to get partial 
 relief. 
 
 The plaintiff invested large capital on the faith of its present mode 
 of enjoyment of its franchise. It has continued in such enjoyment 
 for some ten years. To change involves loss and expense. Clearly, 
 then, it cannot be deprived of what so nearly approximates a vested 
 right without clear legislative intent. 
 
 Is it impossible, then, for the defendant to run an electric railway 
 along its streets without making these disturbances? It is practically 
 conceded, although there was some slight evidence to the contrary, 
 that if instead of a single trolley wire and an earth return two trolley 
 wires were used, one for the positive and the other the negative cur- 
 rent, the difficulties would be just as completely obviated as if the 
 telephone company used a metallic circuit. There is such a system 
 in use in a number of cities of the country. There are two in opera- 
 tion in this city and two or three more are projected here. In such a
 
 366 THE ELECTRIC RAILWAY. 
 
 system the electricity is carried from one wire down through one trolley 
 wheel and mast to the motor of the moving car, and returns from the 
 motor to the other wire by means of a second trolley wheel and mast. 
 It is said it is wholly impracticable. 
 
 The single-trolley system is in use on nine-tenths of the electric rail- 
 way systems in the United States. This arises doubtless from two 
 facts : First, that it is perhaps one-fourth cheaper in its outside con- 
 struction; and, second, because in single-track railways, of which 
 there are many more than double track, where it is necessary to have 
 many switches and turn-outs, the complications of wires overhead 
 increase much more rapidly with a double trolley than a single trolley. 
 But in the case at bar we have a double track from one end of the 
 road to the other, so that no switches are required. Mr. Short, who is 
 a skilled mechanician engaged in a company which bears his name 
 and which erects both the single and the double trolley systems, says 
 that he recommends the single trolley for single tracks and the double 
 trolley for double tracks. He seems the only witness on either side 
 that is free from suspicion of bias. 
 
 It is admitted by many of defendant's witnesses that electrically 
 the double-trolley system is exactly as good as the single-trolley, and 
 this must be so, because a return by wire makes quite "as good a cur- 
 rent as by the earth. The claim that electricity in the wheels passing 
 to the track gives additional traction is conjectural, and is not ap- 
 parent in a practical comparison of the grades ascended by the cars on 
 the two systems. The objection to the double-trolley system is the 
 mechanical difficulty in making proper switches and in supporting 
 the superstructure. In the double-track road, like the defendant's, the 
 first difficulty does not arise, and the second difficulty has in fact been 
 overcome in every double-trolley system which has been erected. On 
 the whole, then, it seems to me that there is no serious obstacle to 
 defendant using the double trolley. 
 
 The defendant cannot rely on an estoppel. It was notified by the 
 plaintiff as early as March 12, 1889, some three months or more before 
 its electrical plant was put into position, that the use of the single- 
 trolley system would interfere with and injure the use of the plain- 
 tiff's telephone lines. 
 
 But it is said that the injuries here occasioned are not cognizable in 
 the courts, even if the telephone company is to be regarded as a pri- 
 vate property owner in the enjoyment of the telephone franchise, and 
 the case of Frazier v. Siebern, 16 Ohio St., 614, is relied on in this 
 connection. That was a case between adjoining proprietors, where 
 the defendant dug a well on his premises and knowingly diverted per- 
 colating waters which made a spring on the plaintiff's premises and 
 dried the spring. This was held to be damnum absque injuria. There 
 is no parallel between that case and the one before us. There the 
 spring owner had. no ownership in the percolating waters until they
 
 APPENDIX A. 367 
 
 appeared on his property, and the injury he suffered arose simply 
 from his neighbor making his own that to which, while it was on his 
 premises, he was lawfully entitled. In the case at bar, the disturb- 
 ances, in their twofold origin, present slightly different circumstances, 
 but call for the application of the same principle. The earth leakage 
 arises from the defendant pouring electricity into the earth, which, 
 following a course as natural for it as the seeking of a lower level is 
 for water, comes upon the premises of plaintiffs subscribers, and by 
 getting upon the wires of the telephone does harm. The plaintiff, by 
 contract with its subscribers, has a right to have its wire where it is, 
 and for the purposes of this discussion has exactly the same right to 
 object to the presence of electricity on its premises caused by defend- 
 ant and resulting in damage as would the subscriber himself. It is 
 impossible to distinguish such an injury in principle from the case of one 
 discharging filthy water into the ground so that it shall percolate into 
 another's premises and there do damage, or of one causing smoke 
 and cinders to be constantly thrown into one's windows and injuring 
 one's enjoyment of one's property. These are all examples of nui- 
 sances which have been held to give a right of redress. 
 
 Ballard v. Tomlinson, 29 Ch. D., 116; Ry. Co. v. Gardiner, 45 Ohio 
 St., 309. 
 
 As was said in Reinhardt r. Mestasti, 42 Ch. D., 685, " the principle 
 governing the jurisdiction of the court in cases of nuisance does not 
 depend on the question whether the defendant is using his own, reason- 
 ably or otherwise. The question is, Does he injure his neighbors?" 
 
 The disturbance by induction is of similar character, except that 
 the force is applied from one wire to the other through the air instead 
 of the earth. If, as we have found, defendant is not entitled to inter- 
 fere with plaintiff's enjoyment of the telephone franchise, here is a 
 direct act of interference. 
 
 Then it is said no claim can arise for disturbances by the defendant, 
 because there are similar disturbances from other causes, for which the 
 defendant is not responsible, and Woodz>. Sutcliff, 2 Simons N. S. , 163, 
 is cited upon this point. That was a case where the injury sought to 
 be enjoined was the pollution of a stream from which plaintiff had, in 
 time past, drawn water for his business. One of the reasons given by 
 the vice-chancellor for refusing the injunction was that others than the 
 defendant were polluting the stream. The real basis of the decision 
 was the laches of the defendant in asserting his right, for he had pro- 
 vided a new supply of water and had not used the river for some 
 time. This case presented no difficulty in considering the case at bar, 
 first, because there are no substantial disturbances shown other than 
 those caused by defendant, and, second, because the case itself is not 
 sound authority. (See Crump v. Lambert, 3 Eq. Cas. ,414; Crosky v. 
 Lightowler, 2 Ch., 478; Walter v. Selfe, 4 DeGex & S., 315.) In the last 
 case the Court said that because one man was maintaining a nui-
 
 368 THE ELECTRIC RAILWAY. 
 
 sance against the plaintiff was no reason why defendant should set up 
 an additional nuisance. 
 
 We find, then, that defendant is inflicting a legal injury upon the 
 plantiff in the nature of a nuisance from which has already arisen 
 loss, and which must inevitably cause loss in the future, constantly 
 recurring. It is said that the damage is not irreparable, because 
 the plaintiff can expend money and avoid it, and in the same way 
 can arrive at its exact loss, and that, therefore, its remedy is not 
 by injunction, but at law. Neither of these claims can be sus- 
 tained. The most frequent exercise by a court of equity of the power 
 of injunction is to prevent the continual recurrence of injuries from 
 nuisance. The ground is that the plaintiff should not be put to a 
 multiplicity of suits and endless litigation. To 'say that, in order 
 to entitle a man to obtain an injunction against such injury, he 
 should not be able, by the expenditure of even vast amounts, to avoid 
 the injury, is to say that no injunctions can ever issue for such a 
 cause. The point is that he is not obliged to expend money to protect 
 himself when a neighbor injures him, but he may appeal to the courts; 
 and when upon the law side he finds that the remedy there afforded 
 is inadequate because of the necessity for bringing a separate suit for 
 each injury, he may fairly say that by appeals to legal tribunals he 
 cannot repair his damage, and therefore his loss will be irreparable. 
 As to the ascertainment of damages, it is by no means true that in 
 each suit the entire cost of introducing a metallic circuit or a McCluer 
 device would be the measure of damages for this sort of interference, 
 and the very reason for going into a court of equity is to get into a 
 forum where all the injuries can be considered together. 
 
 The order of the court will therefore be that the defendant be enjoined 
 perpetually from the use of the system of electric railway propulsion 
 as now operated by them, or any other which will occasion similar 
 disturbances to those now caused by defendant's single-trolley system. 
 The order of injunction will not take effect until six months from this 
 day, with leave to the defendant to apply to the court hereafter for 
 further time, if necessary, in a bona fide effort to make the necessary 
 changes. The costs of the action must be paid by the defendant. 
 Decree accordingly.
 
 APPENDIX B. 
 
 INSTRUCTIONS TO LINEMEN. 
 
 For the trolley wire, hard drawn copper is used to a very great 
 extent, its size depending on the service required. No. o B. & S. 
 gauge is a very convenient size and is generally used, although the 
 wire may be smaller with sub-feeders as explained. Silicon-bronze 
 wire has also been largely used on account of its great tensile strength. 
 Copper is quite strong enough, however. 
 
 Span wires should be No. i B. & S. gauge, galvanized, wiped, Swedish 
 iron wire if the trolley wire to be used is No. o. They may be of 
 smaller size if a smaller trolley wire is used. 
 
 Pull-offs and anchor guys should be the same size as the span wires. 
 
 For guard-wire spans, use No. 8 B. & S. gauge, galvanized, wiped, 
 Swedish iron wire. 
 
 For longitudinal guard wires, use No. 10 B. & S. gauge, galvanized, 
 wiped, Swedish iron wire. 
 
 In case of extra heavy or long stretches, two or more strands of the 
 above iron wire may be twisted together. Stranded wires of galvanized 
 iron have been used for ordinary lengths, and are preferred by some. 
 
 POLE CLAMPS. 
 
 The poles should be fitted with suitable pole clamps, so that the 
 wire may be readily adjusted to the required weight. It is very im- 
 portant that the height of the trolley wire from 
 the track should be uniform, and by using pole 
 clamps this uniformity can easily be secured. 
 Great care should be taken in this respect. 
 Fig. 127 shows one style of pole clamp which 
 may be used, and Figs. 128-132 several insula- 
 tor holders for substantially the same purpose. 
 There should be two clamps on each pole, one FJG I27 ~IJoLE~ CLAMP. 
 for trolley spans, the other for guard spans. 
 
 In certain cases where a number of crowfoot wires diverge from 
 one pole it may be found necessary to make use of more than two 
 clamps on the pole. The clamp to which the guard span is to be 
 attached may be smaller and lighter than that intended for the 
 trolley span. 
 
 24 369
 
 370 THE ELECTRIC RAILWAY. 
 
 After the wires have been erected and lined up they should be more 
 closely adjusted by loosening or tightening the bolts in the clamps 
 which hold the spans. 
 
 In some cases where the spans are extremely long or heavy it may 
 
 FIGS. 128-132. INSULATED HOLDERS 
 
 be advisable to use a third span fastened to clamps and allowed to 
 sag toward the trolley span, which should be secured to it for addi- 
 tional support. 
 
 SPANS. 
 
 In the erection of a line the trolley spans should be the first wires 
 stretched, and should be pulled up into their permanent position. Or- 
 dinarily, they should be as tight as two men can pull them with the 
 help of a single 3-inch block and fall. The tension should be from 
 300 to 600 pounds. The guard spans should be as nearly perfectly 
 insulated as possible from the trolley spans and the poles. This is 
 accomplished by stretching them from pole clamps to which nothing 
 is attached but the guard spans. Anchor guys should be attached to 
 the trolley-span pole clamps and not to the guard-span clamps. 
 
 Next, guard spans should be stretched, and should ordinarily be 
 pulled tight with the vise and strap, and should have about 100 pounds 
 tension on them. In extra long and heavy stretches more tension 
 should be placed upon both trolley and guard spans than that above 
 mentioned, but care shoiild be taken that the spans in every case shall 
 be as loose as steadiness and appearance will allow, since the more 
 tension there is on them the greater liability is there of breakage. 
 The attachment of the wires to the eyebolt of the clamps should be 
 made by a long and close half-connection.
 
 APPENDIX B. 
 
 371 
 
 EAR BODIES, OR SPAN-WIRE CLAMPS. 
 
 After the spans are up the next work is to locate and put on the ear 
 todies. On straight line work the ear bodies, if similar to the ones 
 shown in Figs. 133-146, are " sprung on " the wire. Figs. 147-149 show 
 several of the many kinds of ear bodies that may be used on curve 
 
 
 FIGS. 133-137. SPAN-WIRE CLAMPS. 
 
 work. These should be " cut into" the span, that is, the proper point 
 on the span wire having been determined by means of a plumb line, 
 the wire is cut at this point and the ear inserted, as shown in Fig. 150. 
 Sometimes insulating holders, such as shown in Figs. 151-153. ar 
 used in this work.
 
 372 
 
 THE ELECTRIC RAILWAY. 
 
 Figs. 154-157 show some of the types of " pull-offs," which are also 
 intended for curve work, and Figs. H7-H9 several bracket clamps 
 used in special cases. 
 
 " RUNNING" TROLLEY WIRE. 
 
 Next, the trolley wire should be anchored securely at one end of the 
 line and reeled out for a distance of 1,000 feet, or as far as is found 
 convenient; temporary slings of iron wire should be placed over the 
 
 FIGS. 138-146. SPAN-WIRE CLAMPS. 
 
 span wires and the trolley wire raised up on a tower wagon and hung 
 in these slings. 
 
 Hauling clamps should be placed on the trolley wire and a tempo- 
 rary anchor made, the wire being pulled up tight by four or five men 
 with double "block and tackle." The permanent anchor should then 
 be pulled up tight and fastened to a pole so as to securely hold the 
 tension placed upon the trolley wire. 
 
 On reaching a curve the wire should be pulled up tight to this point,
 
 APPENDIX B. 
 
 373 
 
 anchored, and then run around the curve, allowing as much slack as 
 may be needed to make the curve. On the curves the slings should 
 be attached to the span wires or pull-offs in such a way that they can- 
 not slip along them. The trolley wires should be placed a little inside 
 
 FIGS. 147-149. CLAMPS FOR CURVE WORK. 
 
 FIG. 150. "CUTTING IN ' A SPAN-WIRE CLAMP 
 
 FIGS. 151-153. INSULATOR HOLDERS 
 
 of their usual position, as judgment may dictate, and pulled compara- 
 tively tight. The wire should then be anchored both ways at the end 
 of the curve, so as to make a fixed point. 

 
 374 THE ELECTRIC RAILWAY. 
 
 The construction may be continued in the same manner as before 
 from this point on. 
 
 After running and hanging up the trolley wire the ears or clamps 
 should be put on. 
 
 " EARS" OR INSULATOR CLAMPS. 
 
 The same clamps should be used on curves as on straight lines, 
 excepting when anchors or splices are introduced. Some ears or clamps 
 have sides or points, which, being bent over the wire, clamp them 
 to it. (See Figs. 133-146 with paragraph on " ear bodies" for " ears.") 
 If clamps are used which require soldering they should be fitted to 
 the wire upside down, closely pressed around with great care, so as not 
 to cut or damage the trolley wire, and then carefully soldered with 
 the ordinary lineman's soldering iron for railway work. All the 
 clamps should be soldered on and the line then turned over so that the 
 clamps will be in their proper position. It is not possible to solder 
 on one clamp, turn it upright, and then go on to the next, and at the 
 same time have a straight, untwisted line. In repairing a line on 
 
 FIGS. 154-157. " PULL-OVER " BRACKETS. 
 
 which soldered insulator clamps are used it is necessary to do one of 
 two things : First, to give the wire a half twist, then solder on the 
 clamp in a reversed position, and when the soldering is done allow the 
 wire to spring back into its original position, thus turning the clamp 
 into its proper place ; or, second, to solder the clamp on when in its 
 proper position, which is an exceedingly difficult thing to do. 
 
 The clamps used for splicing should fit any of the insulators in use 
 on the line. The method of using splicing clamps is explained in 
 another place. 
 
 The strain clamps for anchoring should fit any of the insulators or 
 ear bodies in use on the line. They should be similar to the ordinary 
 clamps so far as the groove for holding wire is concerned, and should 
 be put on in the same way. 
 
 GUARD-SPAN HANGERS. 
 
 In the same way that insulators are used the guard-wire hangers 
 should be employed, several styles of hanger being shown in Figs.
 
 APPENDIX B. 
 
 375 
 
 158-160. For straight line work the straight line guard hanger should 
 be " sprung on" the wire with the insulator down. 
 
 The longitudinal guard should be suspended from below, not over 
 the insulator, by means of the common insulator tie of No. 1 2 wire. 
 
 FIGS. 158-160. 
 
 On curves single and double guard-wire curve suspensions should 
 be employed, the longitudinal guard wire being secured to the insula- 
 tor by a stout tie. 
 
 ANCHORS. 
 
 Double anchors should be put in every few hundred (500 to 1,000) 
 feet on straight line work, and should be put in at each end of every 
 
 curve. By double anchors are meant anchors which hold the line both 
 ways. Figs. 161-162 show how these anchors should be installed. 
 
 A single anchor must never be put in except at the end of a line. 
 Anchoring is half the line construction, and if not well done will cause 
 endless trouble. 
 
 The following instructions for anchoring should be carefully noted : 
 
 For the immediate connection to the trolley wire, use a short strain 
 clamp. 
 
 The same wire that is used for spans should be used for anchors, 
 
 and should be pulled about one-half as tight. Some insulator should 
 be put into each anchor wire about 2 feet from the anchor clamp. 
 For double-track construction of anchors, see Fig. 162 above.
 
 376 
 
 THE ELECTRIC RAILWAY. 
 
 When the trolley wire requires splicing a " splicing ear" should be 
 used in connection with an insulator. If a splicing ear like the one 
 shown in Fig. 163 is used, the end of the wire after being passed up 
 
 FIG. 163. SPLICING "EAR." 
 
 through the proper holes and soldered should be turned back over 
 the top, as shown in the figure, and not left sticking up straight. 
 
 FEEDERS. 
 
 In " cutting in " feeders, measurements should be made of the width 
 of street, distance between trolley wires, etc., and the feeder span 
 made up in the shop ready to install, since trouble will be found in 
 making up a span on the street. All feeder spans should be put up as 
 represented in Fig. 164. 
 
 The wire used for connecting the feeder with the clamp or ear 
 should be of the same size and material as the trolley wire. The wire 
 on the dead end of the feeder span should be the same size and kind 
 as used for span wires. The insulator on the dead end should be close 
 to the ear and not next to the pole. 
 
 The insulator at the other (live) end should be close to the pole, 
 and the side feed- fastened to the insulator by a half connection. The 
 
 FIG. 164. ARRANGEMENT OF FEEDER. 
 insulator should then be attached to the pole clamp by the regular 
 wire used for spans, allowing about two feet between the pole cap 
 and the insulator. Side feeds should not be put in on curves. 
 
 INSULATING JOINTS. 
 
 It is sometimes necessary to divide the trolley wire into sections 
 insulated from each other. This is done by the insertion of " insulat- 
 ing joints." 
 
 Insulating joints should be inserted at spans, in the same manner 
 as a splicing ear.
 
 APPENDIX B. 
 
 377 
 
 If such an insulating joint as the one shown in Fig. 165 is used, the 
 span wire should be attached to its central segment. 
 
 Connection can be made with a feeder by the ordinary side feed 
 
 FIG. 165. INSULATING JOINT. 
 
 simply by clamping the side feed wire into the tongue of the insulat- 
 ing joint, by means of a screw provided for this purpose. 
 
 It is sometimes advisable in the construction of overhead lines to 
 run a wire from each side of an insulating joint to a switch on the 
 pole, so that the insulating joint may be short-circuited in case it is 
 desirable to throw the two sections of line 
 together. One method of doing this is 
 shown in Fig. 166. 
 
 The kind ot insulating joint shown in Fig. 
 165 may be used as a lightning arrester, if 
 supplemented by a special device having a 
 series of small breaks of about one-sixteenth 
 of an inch each, which should be " cut into " 
 the span wire attached to the ear and dead 
 grounded at the other end on the pole, as- 
 suming that iron poles are used. If wooden 
 poles are used this ground should be made 
 by leading a wire down the pole and fasten- 
 ing it to the rails. 
 
 LOCATION OF TROLLEY WIRE. 
 
 The trolley wire should be located as fol- 
 lows : 
 
 On straight line work the wire should be 
 over the center of the track. 
 
 At curves the wire should be placed on 
 the inside of the curve, its distance from the 
 center of the track depending on the degree 
 of curvature. 
 
 \ 
 
 FIG. 166. SWITCH AROUND IN- 
 SULATING JOINT. 
 
 Where lines branch from each other frogs 
 are used for directing the trolley wheels to 
 the proper line. When the line branches off to the right use a right- 
 hand frog, and when to the left, a left-hand frog, and when the track 
 splits symmetrically a symmetrical frog should be used. 
 
 In case the angle between the diverging tongues is not the same as
 
 378 
 
 THE ELECTRIC RAILWAY. 
 
 that which the trolley wires make, the tongues should be slightly bent 
 to make the two angles the same. Some frogs are made with lugs on 
 the sides for the attachment of an insulating truss, so that guy wires 
 may be fastened thereon to steady the frog and prevent i-ts tilting. 
 
 FIG. 107. -PLACING FROGS. 
 
 Frogs should not be secured to the trolley wire until the proper loca- 
 tion is assured. With the line at its usual height, 18' 6" to 19' 6", the 
 proper place may be found as follows : 
 
 In Fig. 167 let R X Z Y S represent the position of the track. 
 
 Let o represent the center of the triangle, X Y Z; that is, the 
 point of intersection of right lines bisecting the three angles X, Y, Z. 
 The position of the frog should be directly over the spot (o). Observe 
 
 FIGS. 168-175. FROGS. 
 
 that this throws the main line a few inches out of center, and like- 
 wise the branch line a little out of center. This is all right, however, 
 and is the secret of making the frog work successfully. Figs. 168-175 
 show a few types of frogs.
 
 APPENDIX B. 
 
 379 
 
 DIAGONAL AND RIGHT-ANGLE CROSSINGS. 
 
 Where trolley wires cross there should be used : First, a diagonal 
 crossing, if the lines make an acute angle of less than 60 degrees; 
 second, a right-angle crossing, if the acute angle is between 60 and 
 
 FIGS. 176-179. TROLLEY-WIRE CROSSING. 
 
 90 degrees. The method of putting up diagonal and right-angle cross- 
 ings is the same as that for frogs. See Figs. 176-179. 
 
 GUARD WIRES. 
 
 Guard wires should be insulated from everything else. They should 
 be cut into sections at the same points as the trolley wire, and at every 
 i, ooo feet besides. The guard wires should be directly over the trolley 
 wires, that is, one guard wire to each trolley wire. 
 
 Before the line is left as completed all insulators, pole clamps, and 
 other parts of the line material needing it should be painted neatly 
 and with care, that the parts may be kept from rusting. 
 
 POLE SPECIFICATIONS. 
 
 If wooden poles are used they are generally of cedar, chestnut, or 
 Georgia pine, the latter being usually squared and pointed at the 
 top. Round poles should be trimmed, smoothed and pointed, and the 
 butts of all wooden poles should be painted with some preserving 
 compound. The length should be not less than 28 feet, with mean 
 diameters of at least 7 and 9 inches at the top and bottom respectively. 
 
 Iron poles should be at least 28 feet long, and if made in sections, as 
 the Walworth pole, the respective diameters of the sections should be 
 generally 6, 5, and 4 inches. There are many pole manufacturers, but 
 the Walworth Co. is perhaps most widely known. Iron poles should 
 have insulated caps and cast-iron trimmings, though these latter are not 
 absolutely necessary. All poles should be set six feet in the ground,
 
 380 THE ELECTRIC RAILWAY. 
 
 with tops on a level, about 20 feet above the level of the rails. For iron 
 poles, the rake should be about five-sixteenths of an inch to one foot of 
 length, and for wooden poles at least twice as much, this depending of 
 course upon circumstances. The object aimed at is that when the strain 
 of the span and the other wires is upon the poles they shall stand 
 straight. All poles should be well grouted with broken stone and ce- 
 ment, and it is well to place a large stone at the heel of the pole, also 
 one in front of the pole just below the surface of the earth, so that the 
 pole shall be well braced. The distance apart of the poles will depend 
 upon circumstances; usually about 125 feet apart will answer. All 
 poles should be painted once before being set, and again afterward, 
 so that they may present a good appearance.
 
 APPENDIX C. 
 
 ENGINEER'S LOG-BOOK. 
 
 January 17, 1892. 
 COAL USED, 5,650 LBS. WATER USED, 36,360 LBS. 
 
 CONDITION OF TRACK. 
 
 Muddy. 
 
 CARS IN USE. 
 Standard, 8. 
 Double truck, i. 
 Trailers, 2. 
 
 NOTE TIME AND NUMBER OF 
 EXTRA CARS IN LOG. 
 
 Started, 5.50. 
 Shut down, 11.20. 
 
 MACHINES IN USE. 
 Engines, A. 
 Boilers, / and 2. 
 Dynamos, /, 2, 3. 
 
 NOTE ANY MACHINES THROWN 
 INTO USE IN LOG. 
 
 TIME. 
 
 INDICATOR 
 
 CARDS. 
 
 AMPERES 
 
 ON LINE. 
 
 VOLTS ON 
 
 LINE. 
 
 WEATHER. 
 
 SPECIAL REPORTS. 
 
 
 H 4 8.2 
 
 
 
 
 
 8.00 
 
 C 47-9 
 
 112 
 
 550 
 
 Cold 
 clouds. 
 
 
 0.17 
 
 
 182 
 
 505 
 
 
 Heavy current 
 for about 
 
 
 
 
 
 
 six minutes. 
 
 
 
 
 
 
 Dy n a mo I 
 sparked bad- 
 
 
 
 
 
 
 ly . Cut it 
 
 
 
 
 
 
 out ana 
 
 12.50 
 
 
 
 
 Snow. 
 
 threiv in No. 
 
 1.32 
 
 
 125 
 
 540 
 
 
 4- 
 
 2.40 
 
 
 117 
 
 546 
 
 
 
 2-45 
 
 
 H3 
 
 548 
 
 
 
 4-50 
 
 
 
 
 Snow 
 
 
 
 
 
 
 ceased. 
 
 
 6.10 
 
 
 95 
 
 550 
 
 
 
 10.00 
 
 
 80 
 
 550 
 
 Clear. 
 
 
 381
 
 APPENDIX D. 
 
 CLASSIFICATION OF EXPENDITURES OF ELECTRIC 
 STREET RAILWAYS. 
 
 COMPILED BY H. I. BETTIS. 
 (y permission ) 
 
 An authority upon railway accounts has said that " the most remark- 
 able diversity exists among railway companies in reference to the 
 classification of their operating expenses." 
 
 Undoubtedly the statement was correct as applied to steam railroad 
 accounts, but could the same writer have had his experience in street 
 railway management he probably would have declared that there is 
 no similarity whatever in their treatment of disbursements. 
 
 The electric street railways of to-day are in their infancy, and we 
 have but a limited knowledge of the practical .utility of the various 
 parts and the manner in which they are operated, but with careful 
 attention to the accounts the benefits derived from the various com- 
 binations and methods are readily seen. 
 
 Also comparative statements can be made from month to month, by 
 which it can be seen at a glance wherein we have not paid the proper 
 attention to certain parts of our system, and demonstrate that our neg- 
 lect results in dollars and cents on the wrong side of the income 
 account. 
 
 At the time of the Buffalo Convention our managers and superin- 
 tendents made statements concerning their operating expenses which 
 actually staggered one another. One would state that on his line 
 operation cost so much per mile ; another would state that it cost his 
 company a much different sum per mile. 
 
 The first had included, perhaps, the repairs of cars with wages of 
 motor-men and conductors, while the second might have omitted 
 repairs of cars and included the cost of operating the power house, 
 with or without the cost of repairs on the dynamos. 
 
 There was, and is now, no way in which a comparison can be made 
 between the costs of operating our various street railways, and it is 
 with this object in view that this manual has been compiled, hoping 
 that although it might not suit all, it might at least be a guide to those 
 who have not yet decided upon any classification, and for those who 
 are at all desirous of having a standard or universal system. 
 
 Such a classification should be as simple as possible, and yet with 
 382
 
 APPENDIX D. 383 
 
 sufficient detail to afford the management and owners an easy means 
 for determining in what manner the business of the road could be con- 
 ducted with better success, where the failures had been, and what had 
 been the causes. 
 
 The classifications as given here are not subdivided to the extent that 
 some might think desirable, but sufficiently, I think, for any practical 
 purposes, as any further division would be purely statistical. 
 
 It has been the author's intention to follow the idea of the Interstate 
 Commerce Commission when designing the standard classification for 
 steam roads, to separate the expenses which add to the present or 
 future value of the property from those which do not, as the reader 
 will see by comparing the general expenses and transportation ex- 
 penses with the maintenance of way and buildings and maintenance 
 of equipment, the latter accounts adding materially to the value of the 
 property. 
 
 There is a short classification of construction expenses added, but 
 too much care cannot be exercised in charges to these accounts, bear- 
 ing in mind, however, that nothing should be charged to construction 
 and equipment except that which adds to the first or original cost of 
 the property. 
 
 I wish to express my thanks to Mr. W. E. Baker, Superintendent of 
 the West End Street Railway contract for the Thomson-Houston 
 Company, Boston, whose generosity has enabled me to profit by his 
 experience and advice upon the subject treated. 
 
 OPERATING EXPENSES. 
 GENERAL EXPENSES. 
 
 1. SALARIES OF GENERAL OFFICERS. In this account are included 
 the salaries of the general officers; the heads of departments connected 
 with the supervision and management of the general business of the 
 road. Salaries of division superintendents and assistants may also be 
 charged to this account. By general officers are meant officers in 
 charge of departments and whose jurisdiction extends over the entire 
 road. 
 
 2. SALARIES OF CLERKS IN GENERAL OFFICES. This account em- 
 braces the salaries of all clerks in the general offices, clerks for heads 
 of departments, and all clerks not hereinafter mentioned. 
 
 3. MISCELLANEOUS EXPENSE GENERAL OFFICES. The expense of 
 heating, lighting the general offices; wages of porters, messengers, 
 etc. ; telephone service, and all miscellaneous supplies and expenses of 
 the general offices are charged to this account. 
 
 4. STATIONERY AND PRINTING. Includes the cost of all stationery, 
 books, paper, stamps, pens, pencils, etc. ; also cost of all printing of 
 blanks, circulars, statements, tickets, etc., and the cost of advertising.
 
 384 THE ELECTRIC RAILWAY. 
 
 5. INSURANCE. Includes cost of insurance on property of the com- 
 pany, and against injuries to employees, and all expense of collection. 
 
 6. LEGAL EXPENSE. In this account are included the salaries, fees 
 and expenses of attorneys, witnesses' fees and other court expenses. 
 
 7. INJURIES AND DAMAGES. Expenses on account of persons injured 
 and property damaged, with 'payments of claims, are all chargeable to 
 this account. Wages of persons while disabled, medical attendance, 
 and funeral expenses; also wages of claim agent and others connected 
 with the claim department. Lawyers' fees and other court expenses 
 are not chargeable to this account; nor are damages to property be- 
 longing to the company. 
 
 8. CONTINGENT EXPENSES. This account includes the miscellaneous 
 expenses not otherwise provided for; traveling and other expenses 
 of general officers and assistants, etc., etc. 
 
 TRANSPORTATION EXPENSES. 
 
 1. CAR SERVICE. This account includes the wages of conductors, 
 motor-men, starters, aids, inspectors, and switchmen ; cost of punches, 
 ticket registers, sign sticks, switch sticks, and miscellaneous supplies 
 for car service. The wages of the superintendent of time-tables and 
 chief of conductors, with such clerks as may be under them, should 
 also be charged to this account. 
 
 2. CAR-HOUSE EXPENSE. This account includes the wages of shed 
 foremen, shifters, cleaners, oilers, wipers, laborers, inspectors and 
 watchmen, except such as are employed on repairs of cars. The cost 
 of fuel and lighting the car houses and sheds, lanterns and oil for 
 watchmen, and tools used by workmen on cars (cleaning and oiling, 
 other work except repair) are chargeable to this account. 
 
 3. LUBRICANTS AND WASTE FOR CARS. Oil, grease, tallow and 
 other lubricants, with waste used upon car journals and motors, are 
 included in this account. 
 
 4. ELECTRIC SUPPLIES. This account includes such supplies as are 
 constantly needed for the operation of the electric cars, but cannot be 
 charged to repairs, such as lamps, fuses, carbon brushes, trolley 
 cord, etc. 
 
 5. WRECKING. Wages of those employed in getting derailed cars 
 on the track and removing obstructions and wrecks; tools used and 
 all other expenses incurred on the same account. Expense of getting 
 cars back to the car house when broken down on the line is also 
 chargeable to this account. 
 
 6. OPERATION OF POWER HOUSES. This account includes wages of 
 engineers, firemen, coal shovelers, dynamo-men, oilers, cleaners and 
 others employed in the power houses, except when employed upon 
 repairs. Also the cost of water, water rates, or cost of pumping 
 where the company furnishes its own water-works; carbon brushes,
 
 APPENDIX D. 385 
 
 fuses, lamps and other supplies necessary for the daily operation of 
 the power houses, and not otherwise provided for ; cost of heating and 
 lighting power houses. Repairs and renewals of engines, boilers, 
 dynamos, switchboards and station fixtures are not chargeable to this 
 account. Fuel and lubricants are also chargeable to separate accounts. 
 
 7. FUEL. This account includes the cost of all fuel used in the power 
 houses, with transportation charges on the same. 
 
 8. LUBRICANTS AND WASTE FOR POWER HOUSES. Oils, greases, 
 tallow and waste for use in the power houses, for engines, shafting, 
 dynamos, pumps, etc. 
 
 MAINTENANCE OF WAY AND BUILDINGS. 
 
 1. REPAIRS OF ROADWAY AND TRACK. This account includes all 
 expenditures on account of the road-bed and track, except the cost of 
 rails and ties used, and the cost of repairs and renewals of paving and 
 the supplementary wire. It includes tracks laid in buildings, yards, 
 on turn-tables, wharves, and over bridges ; wages of roadmasters, track 
 foremen, laborers, watchmen, and others, while engaged in track 
 repairs and renewals. It includes cleaning, oiling and sanding track, 
 repairs and renewals of drains under the track, repairs and renewals 
 of planking ove'r bridges, repairs and renewals of frogs and switches, 
 joint fastenings, etc., etc., removing snow and ice, repairs of snow 
 plows and sweepers. It also includes repairs of rails and all work on 
 rails, cutting and drilling, except drilling for tie wires; also labor ex- 
 pended in taking up track. The cost of tools, implements and all 
 supplies used in connection with the track is included in this account. 
 The expense of removing snow and ice, with the cost of repairs on 
 snow plows and sweepers, may be made a separate account if so de- 
 sired, but comes under the head of Maintenance of Way. 
 
 2. RENEWALS OF RAILS. This account includes the cost of new rails 
 laid in the track, with the transportation charges on the same, less 
 the value of old rails taken up. The expense of loading, unloading, 
 drilling, cutting, laying and repairing rails is not included in this account. 
 
 3. RENEWALS OF TIES. This account includes the cost of new ties 
 laid in the track and the freight on the same. The expense of loading, 
 unloading and laying ties is not included in this account. 
 
 4. REPAIRS AND RENEWALS OF PAVING. This account embraces all 
 expenditures on account of the paving. It includes the cost of paving 
 blocks and sand and the cost of transportation of the same ; the wages 
 of pavers, laborers and others engaged in repairs and renewals of pav- 
 ing; also the cost of tools and other supplies for the same work. The 
 expense of taking up and relaying paving when necessitated by the re- 
 pairs on the road-bed, the track and the supplementary wire is not 
 chargeable to this account, but to the account for which such expense 
 was incurred. 
 
 2 5
 
 3 86 THE ELECTRIC RAILWAY. 
 
 5. REPAIRS AND RENEWALS OF THE SUPPLEMENTARY WIRE. This 
 account includes all expenditures on account of the supplementary 
 wire and its connections. It includes the cost of the wire, tie connec- 
 tions, channel pins, solder, and other supplies, also tools and imple- 
 ments used in connection with the work of repairing the supplementary 
 wires; wages of solderers, laborers and others engaged upon this 
 work. The expense of drilling rails for channel pins and tie wire 
 rivets is also chargeable to this account. Expense on the supple- 
 mentary wires, necessitated by the taking up of rails, ties, switches, 
 frogs, etc. , is not chargeable to this account, but to repairs of roadway 
 and track. Expense on the supplementary wires necessitated by the 
 taking up or laying of paving should be charged to the account of 
 paving. 
 
 6. REPAIRS AND RENEWALS OF BUILDINGS, DOCKS AND WHARVES. 
 This account includes the cost of repairs and renewals of all buildings, 
 docks and wharves and of the stationary fixtures and furniture of the 
 same not otherwise provided for ; car houses and sheds, store houses, 
 car shops, repair shops, blacksmith and machine shops, power houses, 
 coal sheds and bins, stations, etc., etc. Repairs of* pits in car houses 
 and shops, cranes in power houses and coal sheds, etc., are embraced 
 in this account. Repairs of tracks in buildings and on wharves are 
 not chargeable to this account. 
 
 7. REPAIRS AND RENEWALS OF POLES AND OVERHEAD LINES. This 
 account includes the cost of repairs and renewals of poles and brackets, 
 with trolley, span, guard and feed wires, with all appliances for sus- 
 pension and insulation of the same. 
 
 MAINTENANCE OF EQUIPMENT. 
 
 i. REPAIRS OF CARS. This account includes the cost of all repairs 
 on car bodies, painting, varnishing, upholstering, relettering cars 
 and car signs ; repairs and renewals of the trucks, brakes, brake shoes, 
 axle boxes, springs, track brushes, snow scrapers, pilots, sand boxes, 
 etc. ; repairs and renewals of wheels and axles. The cost of new cars 
 taking the place of old to make the number good is also chargeable 
 to this account. On roads using both motor and tow cars it is ad- 
 visable to keep each kind separately. 
 
 2. REPAIRS OF ELECTRICAL EQUIPMENT. 
 
 A. Armatures and Fields. This account includes the cost of repairs 
 and renewals of armatures and fields, labor of removing from the cars 
 those damaged, also cost of replacing them and making all connec- 
 tions. New armatures and fields put in to take the place of old, to 
 keep the number good, are chargeable to this account. 
 
 B. Gears and Pinions. This includes the cost of repairs and renewals 
 of gears and pinions, labor removing and replacing in the cars; also
 
 APPENDIX D. 387 
 
 the cost of new gears and pinions put in to take the place of those 
 damaged or destroyed. 
 
 C. Trolleys. This account embraces all repairs and renewals of 
 trolleys and their parts, labor of taking off and replacing, and the cost 
 of new trolleys replacing those damaged or destroyed. 
 
 D. Sundry Repairs. This account embraces the repairs of motors 
 and their connections, excepting armatures and fields, gears and pin- 
 ions, and trolleys. It includes the repairs of wiring, cables, repairs 
 and renewals of lightning arresters, reversing switches, cut-out 
 switches, main motor switches, rheostats, and pans, brush holders, 
 motor pans, etc., etc. 
 
 3. REPAIRS OF STEAM PLANT. To this account should be charged 
 all repairs and renewals of the steam plant in the power house, includ- 
 ing the boilers, engines, pumps and shafting, repairs and renewals of 
 belts, piping, steam. fitting, etc., etc. 
 
 4. REPAIRS OF ELECTRICAL PLANT. 
 
 A. Dynamos. Repairs and renewals of dynamos and their parts; 
 armatures, fields, pulleys, commutators, oilers, bearings and boxes, 
 brush holders, etc., are all chargeable to this account; also labor re- 
 moving and replacing damaged parts. 
 
 B. Switchboard Equipment. Repairs and renewals of the switchboard 
 equipment are charged to this account, such as repairs and renewals 
 of station switches, rheostats, circuit breakers, ammeters, wiring and 
 connections. 
 
 5. TOOLS AND MACHINERY. Repairs and renewals of tools and 
 machinery, shafting, boilers, engines, etc., in the shops of the com- 
 pany; also cost of lubricants for the same. Small tools (not shop 
 fixtures) are chargeable to the account most benefited by them. 
 
 6. MISCELLANEOUS EXPENSE. To this account should be charged all 
 miscellaneous expense of maintenance of equipment not otherwise 
 provided for. 
 
 CONSTRUCTION AND EQUIPMENT EXPENSES. 
 
 Too much care cannot be exercised in charges to this account. 
 Nothing should be charged to construction and equipment except that 
 which adds to the first or original cost of the property. 
 
 1. SUPERINTENDENCE AND GENERAL EXPENSE. Salaries of superin- 
 tendent of construction, assistants, wages of clerks and others em- 
 ployed in the offices of this department. Expense of the office, fur- 
 niture, fuel, lighting, supplies for office, miscellaneous and personal 
 expense of superintendent and assistants while on business. Includes 
 stationery and printing for this department. 
 
 2. ENGINEERING. Wages and expenses of engineers and draughts- 
 men on preliminary work and construction.
 
 388 THE ELECTRIC RAILWAY. 
 
 3. RIGHT OF WAY. Salaries and expenses of right-of-way agent, 
 together with payments for rights of way, easements, franchises, pay- 
 ments for land for buildings, shops, etc. 
 
 4. BUILDING CONSTRUCTION. Cost of buildings car houses, sta- 
 tions, offices, store houses, power house, repair shops, wharves, coal 
 sheds, etc., etc. ; also furniture and fixtures for the same. 
 
 5. TRACK AND ROADWAY CONSTRUCTION. Includes the expense of 
 grading, surfacing, ballasting, ditching and paving ; the cost of rails, 
 rail chairs, ties and stringers, tie rods, joint fastenings, track spikes, 
 frogs and switches, supplementary wire, tie wires, channel pins, solder 
 and miscellaneous track material; also the cost of distributing and 
 laying the same, with the supplementary wire and its connections. 
 
 6. OVERHEAD CONSTRUCTION. Cost of poles and setting, putting up 
 trolley, feeder and guard wires, including cost of wire and all devices- 
 for overhead construction. 
 
 7. CAR EQUIPMENT. Cost of cars built or purchased, including the 
 cost of trucks, motors, upholstering, painting, lettering, varnishing, 
 etc. 
 
 8. SNOW PLOWS AND SWEEPERS. Cost of snow plows and sweepers 
 built and purchased, including the electrical equipment of the same. 
 
 9. POWER STATION EQUIPMENT. Cost of steam plant, engines, 
 boilers, pumps, piping, shafting and belting, dynamos and switchboard 
 equipment, together with installation of the same. 
 
 10. TOOLS AND MACHINERY. Cost of tools and machinery for repair 
 shops, car houses, etc. , and expense of setting and placing in running 
 order. 
 
 11. IMPROVEMENTS AND BETTERMENTS. All expenditures which im- 
 prove the original plant, and of which a portion should be charged to- 
 operating and a portion to construction expenses. 
 
 ACCOUNTING FORMS. 
 Form A. 
 
 Form A is a sheet 9^ inches wide by 12 inches long, and printed on 
 one side only. All the solid lines in the form, as shown below, both 
 horizontal and perpendicular, are ruled in red ink. The dotted hori- 
 zontal lines are all faint blue ruled. The upper set of dotted lines are 
 crossed over each side of the double perpendicular lines by eleven 
 perpendicular heavy blue lines, making thereby twelve columns, each 
 set of twelve columns being separated by double perpendicular red 
 ink hues. The figures under the columns entitled Date run from r 
 :o 31, inclusive. The space above the double horizontal lines at the 
 the page is i# inches; the space devoted to Summary of Mile- 
 age is 2^ inches high. The dotted lines, nine in all, are faint blue 
 
 ied. The short lines at the bottom of the page are printed heavy 
 black, being in the original form one inch long.
 
 APPENDIX D. 
 
 389 
 
 FORM A. 
 
 Daily account of trips run and monthly report of revenue mileage of 
 Motor Electric Car No 
 
 
 
 
 
 Revenue Motor Trips. 
 
 Revenue Towed Trips. 
 
 Date. | 
 
 
 
 1 
 
 
 
 2 
 
 
 
 3 
 
 
 
 29 
 
 
 
 3 
 
 
 
 3i 
 
 
 
 Total. 
 
 
 
 
 
 
 Motor trips on Route N( 
 
 SUMMARY OF MILEAGE. 
 
 Towed trips on Route No 
 
 Total motor mileage, 
 
 Total towed mileage, 
 
 Form C. 
 
 Form C is printed on a four-page sheet 14 inches high by 8^ inches 
 wide. The width of each column of the first page appears under its 
 respective title, the Endorsement taking one-quarter of the page, and 
 both sections entitled Pay Rolls appear on the first page. The fol- 
 lowing two subdivisions appear on the second page, and the next two 
 on the third page. The section entitled Earnings, and all that follow, 
 are on the fourth page. The dotted horizontal lines through the entire 
 form as printed are ruled faint blue, spaced } inch apart. All the 
 solid lines are ruled in red, except the heavy perpendicular lines, 
 which are ruled dark blue. Each of the four sections on the second
 
 390 
 
 THE ELECTRIC RAILWAY. 
 
 and third pages is given half a page, each separate item being placed 
 on a single line. The columns are uniform, being, on the second and 
 third pages, # inch and i inch wide alternately, while on the fourth 
 page they are i inch and X inch wide alternately. On all three pages 
 there are four sets of columns. 
 
 FORM C. 
 
 RECAPITULATION. 
 EARNINGS, EXPENSES AND NET EARNINGS. 
 
 
 Motor 
 Mile- 
 age. 
 
 Earnings. 
 
 Expenses. 
 
 Per cent, 
 of Exp. 
 to Earn. 
 
 Net 
 Earnings. 
 
 Exp. 
 per Car 
 Mile. 
 
 Exp. 
 per 
 Pass. 
 
 Month of 
 
 i 
 
 r 
 
 r 
 
 i 
 
 )'/4 
 
 K 
 
 % 
 
 
 
 
 
 
 
 :;. 
 
 
 
 Decrease.... 
 
 
 :::::::: :r 
 
 
 
 :: 
 
 
 
 EARNINGS, EXPENSES AND NET EARNINGS. 
 Current Year Compared with Corresponding Period of Last Year. 
 
 
 Motor 
 Mile- 
 age. 
 
 Earnings. 
 
 Expenses. 
 
 Per cent, 
 of Kxp. 
 to Earn. 
 
 Net 
 Earnings. 
 
 Exp. 
 per Car 
 Mile. 
 
 Exp. 
 
 per 
 Pass. 
 
 Jan. ist to 
 
 i 
 
 , r, 
 
 \ 
 
 ' 1 
 
 ' 1* 
 
 y\ 
 
 % 
 
 Increase 
 Decrease 
 
 
 ::::::!::[: 
 
 :::::::'::l: 
 
 :::::::[: 
 
 ::::::::{ 
 
 
 
 
 PAY ROLLS. 
 
 
 18.. 
 
 18.. 
 
 Increase. 
 
 Decreas 
 
 General Office 
 
 ' 
 
 '4 
 
 i 
 
 i 
 
 1 
 
 '. 
 
 i 
 
 Motor Men and Conductors 
 Power House 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Maintenance of Roadway and Track 
 
 
 
 
 
 
 
 
 " Equipment 
 
 
 
 
 
 
 
 
 Total 
 
 
 
 
 ] 
 

 
 APPENDIX D. 
 
 391 
 
 PAY ROLLS. 
 Current Year Compared with Corresponding Period of Last Year. 
 
 
 18.. 
 
 18.. 
 
 Increase 
 
 Decrease 
 
 
 1 
 
 % 
 
 1 
 
 K 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Maintenance of Roadway and Track 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Total 
 
 
 
 
 
 
 i 
 
 
 
 
 
 GENERAL EXPENSES. 
 
 Salaries of General Officers 
 " " Clerks in General Offices 
 
 K 
 
 : _- 
 
 % 
 
 y 4 
 
 y^ 
 
 Jl K 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - 
 
 
 
 :: 
 
 
 Contingent Expenses 
 
 
 Total 
 
 
 
 
 \ 
 
 
 TRANSPORTATION EXPENSES. 
 far Service 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 Lubricants and Waste for Cars ! 
 Electric Supplies 
 
 
 
 Wrecking 
 
 
 
 
 
 Fuel 
 Lubricants and Waste for Power House 
 
 
 
 
 
 Total 
 
 
 
 
 
 DETAILS OF EXPENSE ACCOUNT. 
 
 
 
 
 
 
 
 
 
 MAINTENANCE OF WAY AND BUILDINGS. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 " " Ties 
 
 
 
 
 
 
 
 
 Repairs and Renewal of Paving 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ** " u " O * h H T inf* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Total, 
 
 
 
 1 
 
 
 

 
 392 
 
 THE ELECTRIC RAILWAY. 
 DETAILS OF EXPENSE ACCOUNT Contimied. 
 
 MAINTENANCE OF EQUIPMENT. 
 
 Repairs of Cars 
 
 " " Electrical Equipment.. 
 " " Armatures and Fields. 
 
 " " Gears and Pinions 
 
 " " Trolleys 
 
 Sundry Repairs 
 
 Repairs of Steam Plant 
 
 " " Electrical Plant 
 
 Dynamos 
 
 Switch Board Equipment 
 
 Tools and Machinery 
 
 Miscellaneous Expenses 
 
 Total, 
 
 Comparative Statement of Earnings \ Expenses, Net Earnings and Mileage. 
 
 Statistics 
 
 and for Current Year Compared with Corresponding Period of Last Year 
 
 EARNINGS. 
 
 Increase. Decrease. 
 
 Passenger 
 
 Tickets 
 
 Mail 
 
 Miscellaneous . 
 
 Tot 
 
 EXPENSES. 
 
 Increase. Decrease 
 
 I 
 
 Transportation 
 
 Maintenance of Way and Buildings 
 " Equipment 
 
 Total, 
 
 MILEAGE STATISTICS. 
 
 18.. ij Increase. Decrease. 
 
 Double Motor Box. 
 Single " " . 
 Double " Open 
 Single 
 
 Tow Car 
 
 Long Box 
 
 " Open 
 
 Total,
 
 APPENDIX D. 
 
 393 
 
 EARNINGS. 
 Current Year Compared with Corresponding Period of Last Year. 
 
 18.. 
 
 Increase. 
 
 Decrease. 
 
 Passenger I I 
 
 Ticket 
 
 Mail ! I I I. 
 
 Miscellaneous | I.. 
 
 Total, | 
 
 EXPENSES. 
 Current Year Compared with Corresponding Period of Last Year. 
 
 Increase. Decrease 
 
 General 
 
 Transportation 
 
 Maintenance of Way and Buildings 
 " Equipment 
 
 Total, 
 
 MILEAGE STATISTICS. 
 Current Year Compared with Corresponding Period of Last Year. 
 
 Increase. Decrea 
 
 ' ' 
 
 Double Motor Box. . . 
 Single " " .. 
 Double " Open. 
 Single " " . 
 
 Tow Car 
 
 Long Box 
 
 " Open 
 
 Total, 
 
 Form D. 
 
 Form D is a sheet 12 inches wide by 17 */ inches long, and is printed 
 on both sides, Form E being on the reverse side, so that when the 
 book lies open, the Material 7$^ page is opposite the Material Received 
 page. Single red ink perpendicular ruled lines divide the columns, 
 and the width of each column is given under its respective title. 
 There are 59 horizontal faint blue lines on the page. The height 
 above the double horizontal red lines at the top is 1 1 /2 inches. 
 
 FORM D. 
 MATERIAL USED. 
 
 Charge Where To whom 
 
 unt - i account. used. delivered.
 
 394 
 
 THE ELECTRIC RAILWAY. 
 
 Form E. 
 For explanation, see remarks descriptive of Form D. 
 
 FORM E. 
 MATERIAL RECEIVED. 
 
 Date. 
 
 Kind and amount 
 of material. 
 
 Price. 
 
 Amount. 
 
 From 
 whom. 
 
 How 
 received. 
 
 Bill 
 Checked. 
 
 *" 
 
 3. 
 
 % 
 
 X 
 
 * 
 
 
 
 * 
 
 
 
 Form F. 
 
 Form F is a slip 7% inches wide by 4 inches high, and printed on 
 one side only. The word Foreman is at the foot of the slip on which 
 appears 8 lines running the full cross-width of the paper; with a 
 double line above and below the single lines. There is no colored 
 ruling on the slip. 
 
 Storekeeper: 
 
 Please furnish the following material for use on 
 
 Foreman. 
 
 Form P. 
 
 Form P is a book 4 inches wide by 6^ inches long, bound with a 
 stout paper cover. It contains six sheets, which when folded in the 
 centre make a book of 24 pages. 
 
 The blank space above the double ruled line at the top of the page 
 is about i inch high. 
 
 The widths of the several columns appear under their respective 
 titles. The dotted lines in the form below are ruled faint blue in the 
 original, and are spaced % inch apart, of which there are 19 in all. 
 All the solid lines are ruled in red. 
 
 TIME WORKED by.... 
 for week ending 
 
 Occupation, 
 
 Rate per day or month, $. 
 
 Specification of Work. 
 
 a 
 
 ^ 
 
 CO 
 
 | 
 
 
 i 
 
 d 
 
 o 
 
 
 
 H 
 ^ 
 
 
 
 1 
 
 
 
 r. 
 
 ?, 
 
 IS 
 
 . 
 
 Amount 
 
 
 Charge 
 
 Account. 
 
 3^ in. 
 
 '4 
 
 '4 
 
 YA, 
 
 ', 
 
 K 
 
 '. 
 
 * 
 
 ?i 
 
 % 
 
 X 
 
 3 
 16 
 
 3. 
 
 16 
 
 !-- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Total, 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
 APPENDIX D. 
 
 395 
 
 On the title-page of cover appears the following, properly spaced, 
 of course : 
 
 TIME BOOK. 
 
 week ending. . , 
 
 certify the time and rates as shown herein are correct. 
 
 Foreman. 
 
 On the last page of cover appears the following : 
 
 The time of each man must be entered in this book at the close of 
 each day's work, and the book forwarded to the General Office at the 
 end of each week. Do not fill in the last two columns. 
 
 Form G. 
 
 Form G is printed on a sheet 8 inches wide by 7 inches high. An 
 endorsement appears on the back of the sheet. All of the solid lines 
 in the following form are ruled .in red ink, and the width of the col- 
 umns appears under each heading. There are 22 ruled lines, in faint 
 blue, on the sheet, and the height of the space at the top is i inch. 
 
 FORM G. 
 
 Department. 
 
 Foreman's report of material used in month of , 189. . 
 
 Date. 
 
 Location and de- 
 scription of work. 
 
 Quantity and kind 
 used. 
 
 Foreman \ 
 columns 
 
 vill leave these 
 blank. 
 
 i in. 
 
 2K 
 
 *X 
 
 Cost. 
 
 Total cost. 
 
 X IX KIM 
 
 On the back of Form G appears the following endorsement : 
 
 Department. 
 
 Foreman's Report of material used 
 
 in month of , 189. . 
 
 I certify the within report to be correct. 
 
 Foreman, 
 
 Foreman will keep a careful and accurate account of all materials 
 used or issued, and make report of same on this blank daily. 
 
 Form S. 
 
 Form S is a sheet 9^ inches wide by 12 inches long, and is printed 
 on one side only. There are 22 horizontal lines on the page, ruled
 
 396 THE ELECTRIC RAILWAY. 
 
 faint blue. The lines that appear in the following printed form are 
 red. The width of each column appears under its title. 
 
 FORM S. 
 
 THOMSON-HOUSTON ELECTRIC COMPANY, RAILWAY DEPARTMENT. 
 On account West End Street-Railway Company. 
 
 Daily Report of Condition of Motor Car 
 
 
 rRr, 
 
 
 Div Clerk 
 
 
 Car 
 
 Number. 
 
 In Shops. 
 
 Out Shops. 
 
 Remarks. 
 
 Date. 
 
 Time. 
 
 Date. 
 
 Time. 
 
 iK in. 
 
 j 
 
 ' 
 
 ' 
 
 1 
 
 4
 
 APPENDIX E. 
 
 CONCERNING LIGHTNING PROTECTION. 
 
 BY PROFESSOR ELIHU THOMSON. 
 
 The discharges of lightning as affecting railway and other installa- 
 tions may be divided into four classes. 
 
 The first and probably the most dangerous of all discharges of 
 lightning is a stroke falling on the line from the clouds and seeking 
 ground through the conductor itself and through the cars connected 
 to it. 
 
 Second, inductional discharges produced by discharges in the 
 clouds moving in a parallel or approximately parallel direction of the 
 line of the track. These discharges are of course less severe than the 
 former, but may introduce complicated actions owing to the induction 
 on the track being the same in direction as that on the line. 
 
 Third, leakage discharges, which may be of the nature of lateral 
 branchings from a portion of earth which has received a heavy stroke, 
 and which lateral branchings will of course divide themselves over the 
 return conductors, and may reach the station. 
 
 Fourth, electrostatic induction from cloud layers, which when dis- 
 charged of course give rise to changes of electric potential in all 
 metallic masses under them. This latter is perhaps the most common 
 action occurring during thunder storms and the one which gives rise to 
 the minor discharges tending to equalize the potentials over the sys- 
 tem and which have to be guarded against by the arrester devices. 
 The character of the discharges which reach the line may be that of a 
 sudden or instantaneous action or may occupy a slight period of time, 
 and may therefore be characterized as slower inductive discharges. 
 
 As to the character of the discharge, of course the more sudden and 
 violent the action the more dangerous it is likely to be to the insulation 
 of the plant. The slower discharges having time to reach an equaliza- 
 tion are not so apt to produce those great differences of potential 
 between one point and another which lead to perforation of insulation, 
 etc. It is questionable, indeed, whether any devices whatever will be 
 sufficient to take care of the plant and avoid all damage where the 
 lightning falls directly on the trolley line. 
 
 The only safeguard in such a case is to seek out the exposed sections 
 of such a line places, for example, where the line passes over high 
 ground and stands alone unsurrounded by metallic wires or high or 
 
 397
 
 39 8 'THE ELECTRIC RAILWAY. 
 
 metallic buildings and protect such places by the introduction of 
 lightning rods on the suspending poles run into the ground and extend- 
 ing high enough above the general trolley line to divert any discharge 
 therefrom. 
 
 It is pretty safe to say that, so far as the station itself goes, damaging 
 discharges will mostly enter on the feeder lines; though discharges 
 which do damage may of course come in on the ground return, espe- 
 cially if the stroke should fall to the ground not far away from the 
 station and if the ground itself be not of a very moist and conducting 
 character. It is conceivable that in the case assumed the discharge 
 from cloud to earth may find the earth so poor a conductor as to seek 
 the rails and scatter itself along them. As the trolley line and the 
 whole structure of the electrical installation is practically only a 
 metallic extension of the rail conductor, there may, of course, be dan- 
 ger from the current scattering itself over the whole system and break- 
 ing down the insulation in its path. The remedies which seem most 
 applicable to the case of lightning discharges reaching the stations by 
 the feeders, whether these discharges are of an inductive character or 
 direct strokes, are the employment of lightning arresters on the feeder 
 lines in such a manner that the earth connection from them is as short 
 and direct and of as little self-induction as possible. 
 
 To this end each station should have heavy, broad ground connec- 
 tions, such as good-sized piping led to the best metallic grounding con- 
 ductor within reach, arid this ought to be as short as possible, the 
 feeder lines being brought down to it rather than that it be brought 
 up to the feeders. The type of arrester which seems to be the best 
 adapted to use here is one in which the spark gap is as small as it can 
 be made safely and in which there is a means for disrupting or blow- 
 ing outthe arc, which will, of course, form to ground and short circuit 
 the system when a lightning discharge takes place. 
 
 There should also be a heavy self-induction coil on the dynamo side 
 of this arrester, placed in such a manner that it tends to divert any dis- 
 charge to ground and oppose a reaction of great force to any heavy 
 discharge moving in the direction of the dynamo. A self-induction of 
 this kind is also useful in preventing the short-circuiting action through 
 the arrester from attaining undue proportions in an instant. 
 
 It acts, in other words, to secure a little time for the blowing-out 
 action to be accomplished before the current in the ground branch has 
 become equivalent to a short circuit on the system. It would seem in 
 all cases preferable wherever the grounded dynamo terminal exists to 
 bring that terminal back to the ground plate of the arrester and make 
 the ground at that spot. 
 
 In this way the spark gap in the lightning arrester becomes a shunt 
 to the dynamo terminals, the gap possesses the less opposition to 
 currents going to ground, and this tends to divert the discharges from 
 the dynamo loop. A counter-inductive protector should be introduced
 
 APPENDIX E. 399 
 
 into the circuit to protect the dynamo by the counter-inductive effect 
 set up in it by the discharge coming to earth. This arrangement of 
 devices in ordinary cases it would seem ought to be sufficient, and 
 it is only conceivable that in the case of very heavy discharges they 
 might not serve their purpose fully. 
 
 A station constructed with an iron framework which in itself was 
 made the ground for the arrester would, of course, be an ideal arrange- 
 ment. An open iron cage or framework of iron around the dynamo 
 would be an addition of value, the openings in the cage being made 
 of convenient size, such as a foot square, and therefore made so as not 
 to interfere with the manipulation of the machine. In case such a cage 
 were employed it should have a spark gap and blow-out device be- 
 tween each terminal leading from the commutator to the cage. The 
 self-induction of the blowing-out apparatus should be on the side of 
 the dynamo circuit. This arrangement might be an inconvenient one 
 to adopt in stations, and while effective, the objections to it might be 
 sufficient to rule it out. 
 
 An effective substitute for such an arrangement may be to use a 
 spark gap between the field magnet and the conductors leading to the 
 commutator brushes of the dynamo, so that a stroke tending to jump 
 through the insulation of the armature to the core would be prevented. 
 The principle of this arrangement would be to establish a weak point, 
 as it were an artificial weak point, in insulation, between the iron frame 
 of the dynamo and its wiring such that the discharge on reaching the 
 dynamo from either terminal would be led to take its path across the 
 discharge plates (or weak point provided) instead of through any of the 
 insulation of the machine, which insulation should, of course, be su- 
 perior at every point to the insulating power of the spark gap. 
 
 This, it would seem, ought to be a very effective arrangement in 
 saving injury to dynamos, and if its action could be coupled with 
 self-induction in the lines beyond the spark gap on the dynamo side, 
 of course the effectiveness would be increased. 
 
 There still remains the condition of the machine acting as a con- 
 denser, taking in a certain static discharge, suddenly reaching the 
 limit of its capacity, backing up the discharge and causing perforation 
 an action which occurs in condensers and frequently leads to their 
 destruction by perforation. A counteracting condenser might possibly 
 be introduced, which would cause the spark gap to be called into 
 service as the weak point, instead of leaving it to the condensing 
 action of the machine. If this reasoning be correct, and it appears 
 that it must be, then the spark gap should be supplemented by a 
 certain amount of condenser surface adjoining it, and leading, as it 
 were, the discharge in a sort of blind alley for building up the po- 
 tential across the spark gap. 
 
 As to continuance of the use of the pipe arrangements for dis- 
 charges tending to reach earth at stations, it would seem that if the
 
 40 THE ELECTRIC RAILWAY. 
 
 precautions above mentioned were caried out consistently there would 
 be no real reason for the pipe being retained, though without question 
 it does no harm if present, and may do some good. 
 
 Concerning the protection of dynamos from discharges reaching 
 them from a ground line, it would seem that the precautions above 
 stated would be sufficient to protect them, the greater danger being 
 from the discharges reaching the machines from the overhead lines, 
 for which all the precautions are taken ; while the discharges coining 
 from earth would necessarily be of rather lower intensity, and with the 
 device of the spark gap attached to the dynamo it would seem that 
 these discharges would not necessarily puncture the insulation. 
 
 It may be well to reiterate here that the best station arrangement 
 would be one constructed of an iron framework connected to the 
 ground, and it is further to be understood that if the station be itself ex- 
 posed to strokes of lightning none of the wiring leading from the station 
 overhead should exceed in height the highest point of the station 
 which has a conductor, but that, on the other hand, a solid conducting 
 rod should be placed in connection with the framework of the station 
 and pass upward to a sufficient height to cover electrically, or at least 
 protect from discharges, a considerable extent of area of the surround- 
 ing territory. This would prevent any lightning stroke falling on the 
 lines very near the station, the object being to interpose as much of 
 the self-induction of the lines as possible between the point of stroke 
 and the station. 
 
 Whether the lightning-arrester devices be put upon the feeders or 
 upon the wires leading to the dynamo separately does not seem to 
 matter very much. This remark applies to the general lightning 
 arresters which are applied to the line's entering the station. If applied 
 to the feeders alone it would seem to be sufficient when supplemented 
 by the dynamo spark gap shunting arrangements above described. As 
 to the line protection, it is recommended that a survey of the trolley 
 line be made in relation to the chances of its being struck by lightning, 
 or in relation to its exposure to cloud effects directly. Note if it should 
 pass over a hill more or less open and which is not built upon, as the 
 line would be exposed in an exceptionally favorable way to strokes of 
 lightning, being, as it is, an excellent conductor of electricity and 
 having connection to ground though the cars, the ground also being 
 an excellent ground on account of the extent of surface covered by the 
 connected track. 
 
 A survey of the line, then, should discover the points of danger from 
 lightning discharge, and precautions should be taken along the line 
 and particularly at those points liable to receive strokes of lightning. 
 This protection might be accomplished by connecting the foot of the 
 poles to the track and extending the poles upward by a rod to a suffi- 
 cient height properly to protect the trolley line. It is not to be ex- 
 pected, however, that in such case a lightning discharge falling upon
 
 APPENDIX E. 40! 
 
 any pole or poles would fail to disturb the electrical equilibrium of the 
 line. It would, indeed, set up a complicated set of inductive disturb- 
 ances, if it did not leak directly to the line. The line, on account of 
 its extent, is also subject to extraordinary inductive actions from 
 cloud discharges and to electrostatic induction from discharging cloud 
 masses. 
 
 To take care of these as far as possible there should be placed along 
 the line a number of grounded lightning arresters, well protected from 
 rain so as to avoid leakage, whereby a free discharge to earth may be 
 secured without allowing the line current to follow. These arresters 
 should be placed, perhaps, every few hundred feet apart and in ex- 
 posed locations more frequently. They should be found, also, at every 
 angle of the line where, for example, the line takes a turn at right 
 angles or in some other direction, the inductive action of the clouds 
 being expected to accumulate or be more pronounced at such places, 
 and therefore more apt to seek earth through arresters placed at these 
 points. 
 
 The track connections of the line may of course be regarded, in a 
 sense, as extensions of the general conducting system, and the track 
 itself may become subject to inductive disturbances giving rise to 
 trouble. To avoid this there does not appear to be any better means 
 than to have a thorough grounding at points along the track, such 
 grounding being obtained by driving tubes into the ground, iron bars 
 or pipes, until a water stratum is reached, or by making attachments 
 to existing pipes leading deep into the ground. 
 
 In regard to the protection of motors on cars, the problems are of a 
 similar nature to those existing in stations, and the car itself may be 
 regarded for the time being as a small moving station. The light- 
 ning arresters already existing may be retained with advantage and 
 arranged so that they become efficient spark gap shunts of the motor. 
 I am inclined to think that the field-magnet cores in motors will be 
 best protected by being insulated thoroughly from the ground as well 
 as the armature cores, as by that means there will be less tendency for 
 discharges of lightning to leave the wire and jump through the iron of 
 the machine. Where they are not so protected by being insulated, 
 that is, where they are really grounded, it would seem that the light- 
 ning arresters should be able in most cases to take care of the motors. 
 
 There comes in, however, in this connection a condenser action 
 between the field-magnet coils and their cores and between the arma- 
 ture winding and its core which might divert at low potentials a dis- 
 charge insufficient to jump the spark gap, but which by the condenser 
 action existing would suddenly give rise to local potentials at weak 
 points, breaking through the insulation and spoiling the machine. The 
 only remedy available seems to be such a one as has been suggested 
 for the dynamos, to establish, as it were, a sort of blind alley for the 
 discharge, and force it to leap an artificial weak spot which virtually 
 26
 
 402 THE ELECTRIC RAILWAY. 
 
 takes the place of the insulation between the winding and the cores of 
 the dynamo. 
 
 Upon this point experiments are needed to determine just what the 
 danger is and how to meet it. I am inclined to think that here again 
 the precautions used on the dynamo would be effective, that counter- 
 inductive devices of simple character might be introduced along with 
 the ordinary self-induction of the arresters, and that possibly a properly 
 disposed condenser and spark gap combined might be introduced. In 
 fact the spark gap of the lightning arrester itself might be utilized as 
 the protecting spark gap, especially should the sparking be assisted 
 by the condenser action just indicated.
 
 
 APPENDIX F. 
 
 MOTORS WITH BEVEL GEAR AND SERIES-MULTIPLE 
 CONTROL OF MOTORS. 
 
 SINCE the first edition of this book was published there have been 
 two noticeable improvements in the street railway apparatus, some- 
 what antagonistic in their result, it is true, but decidedly ingenious, 
 and each having an important field of action. 
 
 One of these is the improved arrangement of bevel gear transmis- 
 sion that appears in the Sperry street railway motor shown in Figs. 
 
 FIG. 180. SPERRY MOTOR WITH BEVEL GEAR. 
 
 1 80 and 1 8 1. Its essential principle consists in supporting bevel gears 
 by rigid bearings in oil-tight cases, so that the teeth always mesh 
 accurately, and in driving one or both of these gears from the motor 
 shaft through a peculiar form of flexible clutch which seems to give 
 excellent results in practice. It is shown in section in Fig. 181, and 
 needs little explanation, consisting as it does of a pair of concentric 
 spiders with projections interlocking through rubber cushions. The 
 outer spider has teeth inwardly projecting which engage a toothed 
 wheel on the motor shaft. These two sets of teeth admit of consider- 
 able play, which is permissible because there is no marked rubbing 
 action as in ordinary gears only a slight sliding and steady pressure. 
 The result of this device is to give the bevel-geared street-car 
 403
 
 404 
 
 THE ELECTRIC RAILWAY. 
 
 motor a much better place in the art than it has ever taken before, 
 rendering it possible to drive both axles effectively from a single 
 motor flexibly supported. In certain classes of work the advantage 
 of this procedure is very decided. 
 
 The other improvement noted is the evolution of a successful series 
 multiple controller. The scheme has already been mentioned and some- 
 what discussed ; the device for accomplishing it, however, has neces- 
 sitated considerable experimentation, and only very recently has it 
 come into a thoroughly practical form. The difficulties that had to be 
 overcome have been of a mechanical rather an electrical nature, the 
 details hardly being suited for a close description in this place. The 
 result is, that the motors are started in 
 series and thrown into multiple as part 
 of a regular speed-varying regulation. 
 By thus halving the electro-force neces- 
 sary to be generated in each motor it be- 
 comes possible to run at low speeds, with 
 a fair degree of economy and without 
 any considerable use of a rheostat. When 
 high speeds are desirable, the motors are 
 thrown into multiple and work efficiently 
 at the new and greater speed. The net 
 result is an increase of general efficiency 
 of probably not less than 15 per cent., 
 with a corresponding decrease in the amount of power necessary to be 
 furnished. This is most emphatically true in cases where a large 
 range of speeds is necessary, as in roads which start in the centre 
 of a city and then run out into the suburbs, where the limitations im- 
 posed by street traffic no longer exist. 
 
 This series multiple control appears to avoid most of the difficulties 
 which have heretofore been met in running motors economically at 
 low speed. It may be mentioned in passing that this method would 
 perhaps find a very favorable application in enabling the gearless 
 motors to be used with a much greater degree of economy than has 
 been heretofore possible. It goes without saying that to a certain 
 extent the series multiple control decreases the advantages to be 
 gained from the use of a single motor, by raising the efficiency of the 
 ordinary combination to a considerably higher point than heretofore. 
 It seems probable, however, that the single motor geared to both axles 
 may prove very serviceable where speeds are comparatively uniform, 
 as in certain classes of tramway work, while the series multiple con- 
 trol secures for street railway purposes a good efficiency at very widely 
 varying speeds. 
 
 FIG. 181. FLEXIBLE CLUTCH.
 
 APPENDIX G. 
 
 METHOD OF MEASURING INSULATION RESISTANCE OF 
 OVERHEAD LINES. 
 
 IN operating an electric railway plant efficiently, it is necessary 
 that the loss of current through those devices supporting the over- 
 head construction should be a minimum, but no method is generally 
 known by which the insulation resistance of any individual insulator 
 
 FIG. 182. CONNECTIONS FOR MEASURING INSULATION RESISTANCE. 
 Depress key i to measure x. Depress key 2 to measure y. 
 
 can be measured without removing it from the line. Such a method 
 is here described. 
 
 Its use depends upon the line being insulated at each point of sup- 
 port by two resistances in series, of which there is commonly one 
 between the trolley wire and the span wire, and the insulation of each 
 pair of poles between the span wire and the ground. 
 
 The method consists in measuring the voltage from trolley wire to 
 405
 
 4 06 THE ELECTRIC RAILWAY. 
 
 span wire, and also voltage from span wire to ground. These meas- 
 urements should preferably be made in immediate succession, in order 
 
 To determine Leakage on Unea=R(2-x-y) b=^(Z-x-y) R =72700 
 
 MOO 20000 3000U 40000 ' 50000 60000 ,0000 . i 
 
 8QOOO ohms 
 
 Volts 
 FIG. 183. DIAGRAM FOR COMPUTING INSULATION RESISTANCE. 
 
 that the voltage of the line shall not have changed in the mean while. 
 Indicating the respective quantities by the following letters, these 
 equations are true : 
 
 Let R equal resistance of the volt meter. 
 
 A be the insulation resistance of the line insulator. 
 B pole or poles. 
 
 x measured voltage from trolley to span wire. 
 
 y span wire to ground, 
 
 trolley wire to ground. 
 
 For rapidity in securing these results, it is convenient to have a bam- 
 boo pole about fifteen feet long (weighing only about 18 oz.), with 
 two metallic hooks on the upper end thoroughly insulated from each 
 other. One hook should be caught over the trolley wire and the other 
 hook over the span wire, and, by means of suitable commutating keys 
 in metallic connection with each of these hooks, the volt meter, and 
 the ground, the three necessary measurements of voltage at which
 
 APPENDIX G. 407 
 
 point of support can be made within say five seconds. The adjoining 
 cut shows the diagrammatic connections. 
 
 These measurements can be systematically entered in a note-book, 
 and the required value of insulation resistance secured without calcu- 
 lating, by the use of a chart with radial lines drawn thereon, as shown 
 in the cut. 
 
 By proper adding or dropping of ciphers from either or both ordi- 
 nates, and the radial lines, results can be secured from this chart with 
 as great accuracy as if it were drawn on a scale ten or one hundred 
 times greater. This method is due to Mr. Theodore Stebbins.
 
 INDEX. 
 
 ACCIDENT charges, 320. 
 Accumulator, Alkaline zincate, 243. 
 
 alkaline zincate, Use of, 244. 
 Classes of, 239. 
 defined, 230. 
 
 Efficiency of the, in electric traction, 247-. 
 Lead, 231 
 
 lead, Chemical actions in, 231. 
 lead, Electromotive force of the, 231. 
 lead, Mechanical defects of the, 242. 
 lead. Weight of, 243. 
 Life of, 250. 
 Accumulators, Classification of, 237. 
 
 Effect of forced output on, 246. 
 energy, Losses of, in, 240. 
 First use of, for traction, 344. 
 Weight of, necessary for electric traction, 245. 
 Ampere-turns, 15. 
 Anchoring line, 375. 
 Armature, 8. 
 
 Diagram of connections of, 25. 
 Directly connected, 95. 
 shaft, Manner of connecting, 87. 
 Siemens, 18. 
 
 Armatures, Short circuits in, 197. 
 Testing, in shop, 317. 
 
 BATTERIES, Primary, 7, 8. 
 Belting, 183. 
 Blasting, Cost of, 310. 
 Boiler scale, 192. 
 
 scales, Results of, 193. 
 Boilers, 174, 192. 
 
 Proper firing of, 174. 
 
 Means for feeding, 193. 
 
 Water-tube, 174. 
 Brakes, Magnetic, 302. 
 Brush, Carbon, 188, 189. 
 
 CAR bodies, Cost of, 311. 
 
 bodies, Cost of maintenance of, 318. 
 houses, 152. 
 
 Sixteen-feet, work per car mile, 245. 
 wiring, 111. 
 Cars, Double-track, 179. 
 
 open and closed, Trucks for, 107. 
 409
 
 410 
 
 INDEX. 
 
 Chains, Sprocket, 91. 
 Circuit, Field, 10. 
 
 Working, 10. 
 
 City and South London Railway, 273. 
 Coal per car mile, 213. 
 Conductors, Cost of, 284. 
 
 area of, Formula for, 137, 139. 
 Conducting system, Determination of area of, 135. 
 
 Economy law, 285, 286. 
 Coefficient, Riding, 323. 
 
 Traction, 124. 
 Conduit, Closed, 256. 
 
 Modifications of, 259. 
 
 Siemens & Halske, for electric railway, 256. 
 slotted, Commercial importance of the, 258. 
 Slotted, for electric railways, 255. 
 Slotted, with flexible walls, 260. 
 Conduits, Modified forms of, 271. 
 Connecting rods versus gearing, 222. 
 Current rentals, 315, 316. 
 
 density of induction of motor field, 23. 
 
 DROP of potential, 126. 
 
 Average, versus total, 126, 140. 
 Dynamo, Action of, 26. 
 
 Compound-wound, u. 
 equipment, Average output of, 213. 
 first used as a motor, 339. 
 Series- wound, 11. 
 Shunt-wound, 9. 
 
 Dynamos, Connections of, for railway work with earth return, 187. 
 efficiency of, Commercial, 211. 
 efficiency of, Electrical, misleading, 211. 
 Efficiency of, at various loads, 212. 
 General care of, 188. 
 Laws of operation of, 8. 
 Probable faults in, 195. 
 railway, Proper care of, 194. 
 Running, in parallel, 189. 
 self-exciting, Invention of the, 339. 
 
 EARNINGS to pay 5$, 322. 
 
 Earth, The, considered as a conductor, 138. 
 
 Efficiency, Central station, 214. 
 
 Effect of countershaft on, 215. 
 Motor and gear, 224. 
 Electric hoist, The first, 341. 
 
 plant for railways, Cost of a, 311. 
 
 railway, Accounting rules for, 382. 
 
 railway plants, Irregular output of, 163, 164. 
 
 traction by storage batteries, Commercial efficiency of, 247. 
 
 traction, Efficiency of, with one and two motors, 223. 
 Electrics, 14. 
 Electromotive force, Counter, of motors, 65.
 
 INDEX. 411 
 
 Engine, Compound, 39. 
 
 and dynamos, Foundations for, 176. 
 
 Efficiency of an, 31. 
 
 as a heat converter, 210. 
 Engines, 32. 
 Engineer's log-book, 190, 191, 381. 
 
 FEED wire, Cost of, 310. 
 Feeders, 376. 
 
 Cost of running, 137. 
 Sub-feeders, 132. 
 
 Field circuits, Arrangement of, for commutation, 74. 
 Commutation of, 84. 
 coil, Magnetizing of a, u. 
 Strength of, 15. 
 Force, Electromotive, 13, 15. 
 
 Electromotive, generated by unit field at unit speed, 20. 
 Field of, 9. 
 Lines of, 9. 
 
 Magneto-motive, as relating to induction, 15. 
 Magneto-motive, 12, 
 Frogs, 378. 
 
 GAUGE, steam, Standard, 185. 
 Gear journals, Lubrication of, 90. 
 Gearing, Efficiency of, 220. 
 Gearing, Bevel, 403. 
 
 Hydraulic, 83, 227. 
 
 Worm, 222. 
 Gears, Beveled, 93. 
 Spur, 87. 
 
 with " staggered " teeth, 89. 
 Governor for engines, 40. 
 Guard wires, 143, 379. 
 
 HANGERS, Span, 371. 
 
 Guard iron, 375. 
 Heating in magnet coils or armature, 198. 
 
 of bearings, 199. 
 Horse-power, 44, 45. 
 
 per car, 171. 
 
 INDICATOR cards, 49, 51, 52, 53, 54. 
 
 card from railway power station, 55. 
 Indicators for engines, 46. 
 
 Use of, 47. 
 Induction, 15. 
 
 Total, 23. 
 
 Inequality of work between two motors, 80. 
 Insulating joint, 376. 
 Insulator clamps, 374. 
 
 LAW, Ohm's, 14. 
 Lightning arrester, 113, 185. 
 
 arrester, Diagram of, 113.
 
 412 INDEX. 
 
 Lightning arrester, Principles of, 1 14. 
 
 arrester, Self-induction coils for, 390. 
 arrester, Westinghouse, 115. 
 arrester, Wirt, 115- 
 Damage done by, 185. 
 Protecting dynamos against, 391. 
 Protecting motors against, 393. 
 Life of motor parts, 318. 
 Line, Construction of, 369. 
 
 Cost of maintenance of, 318. 
 Drop of, 285. 
 Efficiency of, 216. 
 Height of, 378. 
 Insulation resistance of, 216. 
 resistance of, Calculation of, 123. 
 
 resistance of, Calculation of, with "bunched" cars, 140. 
 Load, Predetermination of maximum, 124. 
 Ratio of maximum and average, 169. 
 Locomotive, electric, Weight efficiency of the, 281. 
 Locomotives, Cost of a horse-power in steam and electric, 282. 
 electric, Repairs of, 288. 
 steam, Repairs of, 288. 
 steam, Weight efficiency of, 281. 
 
 MAGNETIC circuit, Resistance of the, 322. 
 
 properties of various materials, 17. 
 Magnetization, Density of, 17. 
 Magnets, Field, 9. 
 
 McCluer telephone return circuit, 365. 
 Motive power, Cost of, by steam and electricity, 293. 
 Motor, Action of, 26. 
 
 and gearing, Efficiency of, 224. 
 
 as dynamo on down grade, 248, 294. 
 
 care of, Instructions for, 116. 
 
 Commercial efficiency of a single, 220. 
 
 Drehstrom, for electric traction, 263. 
 
 equipment, Cost of, 311. 
 
 equipment, Cost of repairs of, 318. 
 
 gearless, Conditions for, 226. 
 
 Maximum rate of work of a, 69. 
 
 Methods of suspension for the, 89. 
 
 Regulation of a, by varying its field, 65. 
 
 repairs, 316. 
 
 Short gearless, 96, 225. 
 
 single-reduction gear, Efficiency of, 222. 
 
 speed, Regulation of, by varying E. \M. F., 70. 
 
 truck, Eickemeyer gearless, 94. 
 
 truck, Rae, 93. 
 
 Westinghouse gearless, 96, 225. 
 Motors changed from series to multiple, 79. 
 
 Commercial efficiency of a pair of, without gear loss, 220. 
 
 Eickemeyer gearless, 225. 
 
 for high-speed service, 274, 298, 299, 300. 
 
 gearless, Conditions for, 226. 
 
 gearless, Efficiency of, 226. 
 
 One versus two, 220.
 
 INDEX. 413 
 
 Motors, Practical care of, 116. 
 Repairs of, 316. 
 Shunt versus series, 85, 866, , 
 street-railway, Sources of loss in, 218. 
 Typical modern, 97-99. 
 
 OHM'S law, 14. 
 Operation, Cost of, 258. 
 
 as percentage of gross earnings, 326. 
 
 PLANS for small power station, 174. 
 
 for power station of moderate size, 176. 
 
 for very large power station, 178. 
 Pole clamps, 369. 
 
 specifications, 379. 
 Poles, Cost of, 310. 
 Portelectric system, The, 269. 
 Power, Average, 171. 
 
 plant, Material of, 175. 
 
 plant, Proper capacity of a, 167. 
 
 plant, Rule for subdivision of a, 181. 
 
 required for a railway system, 166. 
 
 required for electric cars on grades, 170, 171. 
 
 required for hauling one ton at various speeds, 277, 278, 279, 280. 
 
 station, Arrangement of a, 165, 184. 
 
 station, Cost of, 312. 
 
 station, Efficiency of, at various loads, 214. 
 
 station, Locating a, 162. 
 
 station, Position of a, 161. 
 
 stations, 161. 
 
 stations of various capacities, Cost of a horse-power hou,r in, 284. 
 Pressure, Mean, in engine cylinder, 45. 
 
 mean, Method of computing, 50. 
 
 RAIL sections, 150, 151. 
 
 Railroading, electric, High-speed, 272. 
 
 electric, Limiting possibilities of, 275. 
 high-speed electric, Resistance to motion in, 276. 
 Rails, Bonding, 138. 
 
 Detecting trouble in connections of, 139. 
 Railway crossings, 147. 
 
 dynamos, electromotive force of, Danger to life from, 123. 
 
 Electric, at Richmond, Va., 350. 
 
 electric, Bentley and Knight's, 346. 
 
 electric, Bessolo, Patent involving the, 338. 
 
 electric, Davenport's experiments on the, 334. 
 
 electric, Farmer's experiments on the, 336. 
 
 electric, Field's, 342. 
 
 electric, First commercial, 343. 
 
 electric, Green's experiments on the, 339. 
 
 electric, Henry's, 246. 
 
 electric, History of, 333. 
 
 electric, Lilley and Colton's experiments on the, 337. 
 
 electric, Page's experiments on the, 335. 
 
 electric, Pinkus, Patent involving the, 337.
 
 4 1 4 INDEX. 
 
 Railway, electric, Sprague's work in the, 349. 
 
 electric, Van Depoele's, 349- 
 
 statione fficiency, 214. 
 Railways, electric, Accounting rules for, 329. 
 
 electric, Cost of motive power for, 314. 
 
 electric, Gross receipts necessary for, 322. 
 
 electric, Operating expenses of, 313. 
 
 electric. Patents on, 351. 
 Repair shop, 1 54. 
 
 shop, Special tools for, 156. 
 
 Resistance of rail and earth return, 137. . '.,- 
 
 of iron and copper, 138. 
 Specific, 14. 
 Rheostat, 76. 
 Rods, Connecting, instead of gears, 94. 
 
 SECTION hangers, 376. 
 Series, motors, Connection of, in, 78. 
 Series-multiple control of motors, 403. 
 
 Short-circuiting versus cutting out damaged motor armature coil, 270. 
 Snow, General treatment of, 160. 
 machines, 157. 
 Removal of, 157. 
 Span wires, Stretching of, 370. 
 Splicing gears, 376. 
 Ste*am, Absolute pressure of, 206. 
 
 engine, 29. 
 
 engine, Condensing, 36. 
 
 engine, Losses in underloading the, 39. 
 
 engine, Practical efficiency of the, 210. 
 
 engine, Practical results from the, 56. 
 
 engine, Throttling, 35. 
 
 engine, Working of a, 34. 
 
 engines, compound, Economy of, 207. 
 
 engines, Conditions for maximum efficiency of, 207. 
 
 engines, Friction in, 203. 
 
 engines for large power stations, 168. 
 
 engines, Kind of, suitable for given plant, 169. 
 
 engines, Relation between maximum and average load of, 208. 
 
 Most economical point of cut-off of, 206. 
 
 Most economical ratio of expansion of, 206. 
 
 plant for electric railway, Cost of, 311. 
 
 turbine, Results obtained from the, 215. 
 Street-railway service, Variation of, with size of city, 323. 
 Strength of field in electric motors, 65. 
 Storage battery cars, Regulation of, 77. 
 battery, Definition of, 230. 
 battery, Variations of capacity in a, 241. 
 batteries, Condition of electric traction by, 235. 
 batteries, Cost of maintenance of, 250. 
 batteries, Grids for, 236, 237. 
 
 TELEPHONE circuits, Interference of railways with, 143, 353, 355. 
 Telpherage, 264. 
 
 Electric arrangements of, 266.
 
 INDEX. 
 
 4'5 
 
 Testing line, Method of, 405. 
 Tolls, 328. 
 
 Tonnage coefficient, 178. 277. 
 Track, 147. ' ,' 
 
 bonding, Cost of. 310 
 Construction of, 149. 
 Cost of maintenance of, 318. 
 for electric railways, Cost of, 309. 
 Traction by storage batteries, 230. 
 coefficient, 124. 
 electric, Actual cost of, 328. 
 electric, Alternating currents in, 261. 
 
 electric, by storage batteries, Commercial availability of. 248 
 electric, Commercial consideration of, 309. 
 electric, Commercial efficiency of, 228. 
 electric, Efficiency of, 202. 
 electric, High-speed automatic, 268. 
 electric, high-speed automatic, Siemens', 343. 
 electric, high-speed, Commercial aspects of, 305. 
 electric, high-speed, Efficiency of, 290. 
 electric, high-speed, Experiments on, 296. 
 electric, high-speed, Limiting curvatures for, 304. 
 electric, high-speed, Line voltage for, 304. 
 electric, high-speed, Conclusions regarding, 293. 
 electric, high-speed, Power required for, 298. 
 electric, Probable maximum efficiency of, 228. 
 electric, Three-rail system of, 254. 
 electric, Total cost per car-mile of, 320. 
 electric, Transmissions and transformations in. 202. 
 Trailers, 321. 
 
 Trail-car-mile, Cost of, 321. 
 Train mile versus H. P. hour, 291. 
 Transit, Electric rapid, for large cities. 295. 
 Trolley spans, 370. 
 
 system, Single versus double, 142. 
 wire, 369. 
 
 wire, Anchoring, 375. 
 wire, Arrangement of, on curves, 373. 
 wire clamps, 371. 
 wire, Cost of, 310. 
 wire, Location of, 371. 
 wire, Running, 372. 
 Trolleys, 107,377. 
 
 and bases, 1 10. 
 Early experiments on, 347. 
 Truck, Maximum traction, 105. 
 
 Radial, 106. 
 Trucks, 100. 
 Turbines, 58. 
 Typical electric railway. 101, 103. 
 
 VALVE. 33. 
 
 gear, Corliss, 38. 
 
 gears, 36. 
 Voltage, Danger to life from, 123.
 
 416 INDEX. 
 
 WATER power, 163. 
 
 power, Difficulties with, 163. 
 Water-wheel, Difficulty of governing, 61. 
 Water-wheels, 57. 
 Winding, Compound, 12. 
 Series, 12. 
 Shunt, 12. 
 
 Wire, Supplementary, 139, 385. 
 Wire-tie, 385. 
 Wiring line distance with i-o B. & S.. 142. 
 
 line, Overhead system of, 137. 
 
 Standard car, of Edison Co., 109. 
 
 Standard car, of Thomson-Houston Co.. 109. 
 Watt meters, 315. 
 
 meter, L'se of, on cars, 220.
 
 PUBLICATIONS OF -THE W, J. JOHNSTON CO., LTD., 
 
 41 PARK RO V (Times Building), NEWYORK. 
 
 THE ELECTRICAL W LD. An Illustrated Weekly Review of Current 
 
 tJGER&rsaqt^ 
 
 TORY, Containing LMRtOTRral Electric Light Stations, Isolated Plants. 
 Electric Mining Plants, Street Railway Companies-Electric. Horse and Cable 
 with detailed information regarding each ; also Lists of Electrical and Street 
 Railway Manufacturers and Dealers, Electricians, etc. Published annually. ... 5.00 
 
 THE TELEGRAPH Ii AMERICA. By JAS. D. REID, 894 royal octavo 
 
 pages, handsomely illustrated, Russia ........... 7 oo 
 
 DICTIONARY OF ELKCTICAt, WORDS, TERMS AND PHLHASICSl 
 By EDWIN J. HOUSTON, A. M. Second edition. Entirely rewritten. 5,000 defi- 
 nitions, 562 double column octivo pages, 570 illustrations ....... 500 
 
 THE ELECTRIC MO I'OR AND I TS APPLICATIONS. By T.C. "MAR- 
 TIN and Jos. WKTZLER With an appendix on the Development of the Electric 
 
 Motor since 1888, by DR. Louis BELL. 315 pages, 353 illustrations. ____ 300 
 
 E ELECTRIC RAILWAY IN THEORY AND PRACTICE. The 
 First Systematic Treatise on the Electric Railway. By O. T. CROSBY and DR. 
 
 . . . . 
 
 Louis BELL. Second edition, revised and enlarged. 416 pages, 182 illustrations. 2.50 
 ALTERNATING CURRENT*. An Analytical and Graphical Treatment for 
 
 Students and Engineers. By FKEDEKICK BEDELL, Ph. D., andALBKRTC. CRE- 
 
 HORB, Ph. D. 325 pages. 112 illustrations ................. , .. 2.50 
 
 PRINCIPLES OF DYNAMO-ELECTRIC MA H I NES and Practical Di- 
 
 rections for Designing and Constructing Dynamos. By CARL HERING Seventh 
 
 thousand, 279 pat^es, 59 illustrations ..................................... '. ____ ....... 2.5!X 
 
 ELECrRIO LIGHTING SPECIFICATIONS for the iueof Engineersand 
 
 Architects. By E. A. MERRILL, 175 pages ...................................... 1.50 
 
 THE QUADRUPLEX. By WM. MAVER, Jr., and MINOR M. DAVIS. With 
 
 Chapters on Dynamo-Electric Machines in Relation to the Quadruplex, Tele- 
 
 graph Repeaters, the Wheatstone Automatic Telegraph, etc., 126 pp., 6 till .. .. 1.50 
 THE ELE.fi KN TS OF STATIC ELEC TRICIT Y, with Full Descriptions of 
 
 the Holtz and Topler Machines. By PHILIP ATKINSON. Ph. D. 228 pp., 64 ill ____ 1.50 
 LI(, :i I'.M\<; FL SHE. A Volume of Short, Bright and Crisp Electrical 
 
 Stories and Sketches. 160 pages, copiously illustrated ............................ 1.50 
 
 ELECTRICITY A^D .'lAG^ETISM. Being a Series of Advanced Primers. 
 
 By EDWARD J. HOUSTON. A. M. 308 pages, 116 illustrations ..................... 1.00 
 
 ELECTRIC AL MEASUREMENTS AND OTHER ADVANCED PRIM- 
 
 ERS OF ELECTRICITY. By EDWIN J HOUSTON, A . M . 429pp.. 169 ills... 1.00 
 THE ELECTRICAL TRANSMISSION OK INTELLIGENCE ANO 
 
 OTHER ADVANCED PRIMERS OF ELECTRICIT*. By EDWIN 
 
 J. HOUSTON. A. M ............................................................ 1.00 
 
 ALTERNATING CURRENTS OF ELECTRICITY. Their Generation, 
 
 Measurement. Distribution and Application. Authorized American Edition. 
 
 By GLSBEKT KAPP. 164 pages, 38 illustrxtions ................................... 1.00 
 
 RECE\r PROGRESS i\ ELECTRIC RAILWAY**. Being a Summary 
 
 of Current Advance in Electric Railway Construction, Operation, Systems, Ma- 
 chinery, Appliances, etc . Com pi led by CARL HERING . 386 pp., 120 illus ......... 1.00 
 
 ORIGINAL PAPERS ON DYNAMO MACHINERY AM* ALLIED 
 SUBJECTS. Authorized American Edition. By JOHN HOPKINSON, F. R. S. 
 249 pages, 98 illustrn tions .............................................. ........ 1.00 
 
 DAVIS' STANDAHD TABLES FOR ELECTRIC WIHEUEN. With 
 Rules for -afe Wiring, Instructions for Wiremen and Linemen, Diagrams of 
 Circuits and Useful Data, Third Edl. ion. Revised by W. D. WEAVER ........ 100 
 
 UNI VRAL WIRING COMPUTER for Determining the Sizes of Wires for 
 Incandescent Electric Lamp Leads and for Distribution in General Without 
 Calculation, with Some Notes on Wiring and a Set of Auxiliary Tables. By 
 CARLHKRING. 44 pases ....................................................... 1-00 
 
 EXPERIMENTS WITH ALTERNATING CURRENTS OF HIGH 
 POTENT! \L AMD HIGH FREQUENCY:. By NIKOLA TESLA. 146pp., 
 30 illustrations ...................................................................... 1-00 
 
 LECTURES ON THE ELECTMOMAGNET. Authorized American Edi- 
 
 tion. By Prof SILVANUSP THOMPSON. 287 pages, 75 illustrations ............... 1.00 
 
 PRACTICAL INFORMATION *OR TELEPHONISTS. ByT.D.LocK- 
 
 WOOD. 192p-iges .................................................................... 1 - 00 
 
 WHkiELEK'S CHAR TOP WIRE GAUGES ............................... 1.00 
 
 PROCEKDINGS OF THE NATIONAL CONFERENCE OF ELEC- 
 
 TRICIAN-*. 300 pages. 23 illustrations ....................................... -J 5 
 
 WI<tEil LOVE: A RitmaiiOK of Dots and D <sbe. 256 pages .............. 75 
 
 TA iLEs OF EQUIVALENTS OF UNITS OF MEASUREMENT. By 
 
 CARL HERING ..................................... ...................... * 
 
 CO NTINUOUS CURRENT DVN AMOS AND MOTORS. An Elementary 
 
 Treatise for Students and Engineers. By FRANK P. Cox ................... In press. 
 
 Copies of the above, or any other electrical book published, will be mailed to any 
 < ildresa in the world, POSTAGE PRKPAID. on receipt of urine. Address 
 
 THE W. J. JOHNSTON CO., LTD., 41 ark Row ( Times Building), New York.
 
 3 1158 00490 2549 
 
 ipi 
 
 AA 000697266 5