> f ^ ■ UNIVERSITY OF ILLINOIS ? j LIBRARY ■ Class Book Volume J' - j K\"Ve5 .]■- I iPEMO'j E STORAGE ^ I ii 7 .) It > 7 I ■li ¥ i Spau LlI S^^.%r/7.£4i7#d Digitized by the Internet Archive ,in 2016 https://archive.org/details/electrictransmis00kapp_0 ELECTRIC TRANSMISSION OF ENERGY. THE SPECIALISTS’ SERIES. Frontispiece. ELECTRIC TRANSMISSION OE ENERGY AND ITS TRANSFORMATION, SUBDIVISION, AND DISTRIBUTION. A PRACTICAL HANDBOOK BY GISBERT 'KAPP, C.E., Associate Mevther of the Institution of Civil Engineers y Memher of the histitution of Electrical Engmeers, WITH 130 ILLUSTRATIONS. THIRD EDITIONy REVISED, LONDON : WHITTAKER & CO., 2, WHITE HART STREET, PATERNOSTER SQUARE; GEORGE BELL & SONS, YORK STREET, COVENT GARDEN. 1891. ■ v-i j i i ill I i CHISWICK press: C. WHITTINGHAM and CO., TOOKS COURT, CHANCERY LANE. 4iJ uJ ^>4 REMOTE PREFACE. With a view to bringing the book up to date, the author made some changes in the second edition, the most im- portant of which may be summarized as follows : — The theoretical part has been slightly extended by the addition of the author’s method for the predetermination of the characteristics of dynamos. He had already made use of 'I this method when the first edition of this book was pub- c lished, but as at that time no other engineer had used it, the author did not feel sufficiently confident of its general ^\\ applicability to admit it into his book. Since then, how- ever, the method has been used by many of his colleagues, and has been found to give, on the whole, fairly accurate ^ results, so that the reason which prompted him to exclude ‘S it from the first edition is no longer valid, and the ^ method has therefore been inserted. The practical part ^ of the book has also undergone some alterations primarily ^ due to the progress made within the last three years in the construction of dynamos. The description of obsolete machines has been omitted, and that of more modern U machines inserted, and, wherever practicable, data have been given comprising the leading features of dynamos actually made, and the results of practical tests ; the ^ author holding that precise and numerical information g regarding a few characteristic features in the design of @ successful machines will be of greater value to the reader VI PREFACE. than any amount of general description. The chapter on electric railways has been only very slightly extended, but an exhaustive treatment of the subject would require a volume to itself, larger than the present book, and could therefore not be attempted. Another subject which might have found a place in the present volume is that of power transmission by means of alternating currents. It is quite possible that for the transmission of very large powers over very long distances, the alternating current may eventually prove more convenient than the continuous current, but as up to the present no such transmission plant has yet been erected, the subject must still be regarded as in the experimental stage, and not ripe for discussion in a book which, being intended for practical men, should deal with accomplished facts rather than with possibilities. As regards long distance transmission by means of continuous currents, the account of M. Marcel Deprez’ experiments has been entirely omitted from the present edition. When the first edition was published these experiments formed the only example of any magnitude which could be cited to show that the idea of carrying mechanical power, by means of an electric current over a considerable distance, had been translated into practice. Since then, however, much good work has been done by other engineers, and large numbers of very successful transmission plants are now at work. The examples selected for more detailed description will prove that the electric transmission of energy is not only theoretically, but also technically and commercially, a complete success. The third edition has also been revised and slightly enlarged. Wimbledon, October^ 1890. CONTENTS. PAGE Introductory . . . .1 — 9 CHAPTER I. General Principles — Lines of Force — Relations between Mechanical and Electrical Energy — Absolute Measurements — Ideal Motor and Transmission of Energy — Practical Units .... 10 — 48 CHAPTER II. First Electro-Motor — Professor Forbes’ Dynamo — Ideal Alternating Current Dynamo — Siemens’ Shuttle -Wound Armature — Effect of Self-Induction — Experiments with Electro-Motors — Hefner-Alteneck Armature — Gramme Armature — Pacinotti Armature — Electro- Motive Force created in any Armature . . . . 49 — 83 CHAPTER III. Reversibility of Dynamo Machines — Different conditions in Dynamos and Motors — Theory of Motors — Horse-power of Motors — Losses due to Mechanical and Magnetic Friction — Efficiency of Conversion — Electrical Efficiency — Formulas for Dynamos and Motors . 84 — 99 CHAPTER IV. Types of Field Magnets — Types of Armatures — Exciting Power — Magnetic Circuit — Magnetic Resistance — Formulas for Strength of Field — Single and Double Magnets — Difficulty in Small Dynamos — Characteristic Curves — Pre- determination of Characteristics — Arma- ture Reaction — Horse-power Curves — Speed Characteristics — Appli- cation to Electric Tramcars 100 — 144 Vlll CONTENTS, CHAPTER V. Graphic Treatment of Problems —Maximum External Energy — Maxi- mum Theoretical Efficiency — ^Determination of best Speed for Maxi- mum Commercial Efficiency — Variation of Speed in Shunt Motors — The Compound Machine as Generator — System of Transmission at Constant Speed — Practical Difficulty .... 145 — 159 CHAPTER VI. Classification of Systems according to Source of Electricity — Trans- mission at Constant Pressure — Motors mechanically governed — Self- Regulating Motors — Transmission at Constant Current — Difficulty of Self-Regulation — Motor for Constant Current made Self-Regula- ting — Application to Transmission over large Areas — Continuous Current Transform ator — Transmission between two Distant Points — Loss of Current by Leakage — Theory — Commercial Efficiency — Conditions for Maximum Commercial Efficiency — Self-Regulation for Constant Speed — Practical Example .... 160 — 200 CHAPTER VII. The Line — Relation between Capital Outlay and Waste of Energy — Most Economical Size of Conductor — Formula for Maximum Cur- rent — Formula for Mean Current — Tables for Finding Most Econo- mical Size— Heating of Conductor — Table for Rise of Temperature 201—213 CHAPTER VIII. Circuits for Electric Transmission — Circuits for Electric Distribution — Relative Importance of Insulation — Aerial Lines — Insulators — At- tachment of Conductor to Insulator — Joints— Couplings — Material for Aerial Lines — Estimate for Aerial Line — Protection from Light- ning — Underground Lines — Edison Mains — The Three-Wire System — Various Systems of Underground Conduits — Lead-covered Cables 214—240 CHAPTER IX. Possible Applications of Electric Transmission of Energy — Best Field for it is Long Distance Transmission — Comparison with other Sys- tems — Herr Beringer’s Investigation — Hydraulic Transmission — Pneumatic Transmission — Wire-Rope Transmission — Comparative Tables of Efficiency and Cost — Practical Conclusions . 241 — 259 CONTENTS. IX CHAPTER X. Classification of Dynamo Electric Machines — The Edison -Hopkinson Dynamo — The Thomson-Houston Dynamo — The Immisch Dynamo — The Laurence, Paris, and Scott Dynamo — The Manchester Dynamo — The Elwell- Parker Dynamo — The Crompton Dynamo — The Andrews Dynamo — The Siemens Dynamo — The Goolden Dynamo — The Phoenix Dynamo — The Kapp Dynamo — The Brown Dynamo — The Victoria Dynamo — The Giilcher Dynamo 260 — 311 'CHAPTER XL Historical Notes — Fontaine’s Discovery — Figuier’s Explanation — Early Patent of Pinkus — Early Electro-Motors-^Page’s Electric Railway — Ploughing by Electricity at Sermaize — Electric Cranes — Venti- lating and Pumping by Electricity — Modern Electric Railways — Different Systems — Comparative Merits of Battery System and Conductor System — ^The Besshrook-Newry Electric Railway — The Blackpool Electric Tramway — The Telpher Line at Glynde — Reckenzaun’s Electric Tramcar — Comparative Estimates for Horse Traction and Electric Traction — Progress of Electric Railways and Tramways in America 312 — 339 CHAPTER XIL General character of work done in England and Abroad — Development of Electric Transmission of Energy in Switzerland — List of Installa- tions — The Kriegstetten-Solothurn Plant — Official Tests of Same — The Aichherg Plant — Latest type of Brown’s Dynamo . 340 — 354 h ELECTRIC TRANSMISSION OF ENERGY. INTEODUCTOEY. The transmission of energy and its transformation is the fundamental problem of mechanical engineering. No piece of mechanism yet devised is able to create energy, but all mechanism has for its object the transmission and transformation for useful purposes, of energy already existing in nature in a more pr less inconvenient form. The more perfect our mechanical appliances, the better are they fitted to direct the forces of nature to do useful work ; and in this sense the electric transmission of energy must be regarded simply as an improvement on purely mechanical methods already existing. But it is something more. It not only improves mechanical methods, but extends the field for their application, inas- much as it can, in many instances, reach nearer to the sources of power than any mechanical means. The most important natural sources of power are fuel, wind, and water. As regards the first-named, electric transmission can hardly be considered of any great im- portance for the purpose of reaching the source of power, for fuel, especially in its most useful form of coal, is so easily portable, that in most cases it is more convenient to B 2 ELECTRIC TRANSMISSION OF ENERGY. carry the fuel to the place where the energy is required than to transform it into energy where found and transport the energy to the place of application. It has been sug- gested to erect large generating stations for electricity close to the pit’s mouth, and work the dynamos by steam- power obtained from the small coal which is not worth being carried by rail. The current generated could then be sent along wires to places where power is required, and thus the energy contained even in the refuse of our coal- fields could be utilized. As yet this suggestion has not been carried into practice, except on a very limited scale, namely, in providing motive power for underground rail- ways in coal mines, and in one case for an electric railway on the surface. The other two great natural forces, wind and water, especially the latter, offer a larger field for the applica- tion of electricity. Water-power is only portable in a very limited sense. The great cost of channels, and the difficulty of providing elevated reservoirs close to those places where the power would be of greatest use, compel us in most cases to establish our factories close to natural waterfalls ; in other words, we cannot carry water-power to the work, but must take the work to where the water- power is. Where that is impossible or inconvenient, the power cannot be directly utilized. It is in these cases that electric transmission of power is of greatest value, inasmuch as it enables us to get at many sources of enery which would otherwise be wasted. The amount of energy contained in waterfalls all over the world is enormous. To cite only one or two cases. According to Herr Japing, the hourly weight of water falling in the Niagara is one hundred million tons, representing about sixteen million horse-power, and the total production of INTRODUCTORY, 3 -coal in the world would just about suffice to pump the water back again. M. Chretien^ a French engineer, has in a paper read at the Paris Electrical Exhibition in 1881, given the total water-power in France as seventeen million horse-power, and has suggested, that if by electric transmission only a part of this vast amount of energy were made available for useful purposes, an enormous economy in the consumption of fuel in France would be effected, and, at the same time, the hydraulic works neces- sary would also have the beneficial result of preventing, or at least mitigating droughts and inundations. It is, however, in Switzerland that the better utilization of water-power by electric transmission has the most pro- mising field. Not only is this country rich in waterfalls, which are not liable to any great reduction in volume in summer time, as is the case with water-power in non alpine regions, but coal is dear, wood is scarce, and the population is dense and industrious. We have here con- ditions all favourable to the adoption of any new and imjDroved method of transporting power. It was in Switzerland that thirty or forty years ago teledynamic transmission was first brought to a practical issue, and it is in the same country that electric transmission has most rapidly developed. The total energy represented by all the different transmission plants at Avork at the begin- ning of this year in S^vitzerland, may be roughly esti- mated at 1,500 horse-poAver. In Great Britain there is of course not so much scope for electric transmission of Avater-power, principally be- cause there are but few Avaterfalls of any dynamic magni- tude Avhich are not akeady utilized. The tAvo most important examples of electric transmission are in con- nection Avith electric railways. 4 ELECTRIC TRANSMISSION OF ENERGY. At Portrush, in Ireland, the energy of a waterfall is by means of a turbine and dynamo converted into electri- cal energy, which is conveyed to the line and along the rails into the motor of the car. There it is reconverted into mechanical energy and utilized in propelling the car. A similar installation exists at Newry, where power ob- tained from the Camlough stream is utilized to work an electric railway between that town andBessbrook. There are also several other examples where either water or steam-power is transmitted electrically, but generally speaking progress in this direction has been slow. The reason lies in this, that installations of this kind are necessarily of some magnitude, and cannot be undertaken as mere experiments. If a small installation of electric lighting were to turn out a failure in any particular case, the loss to the contractor would not be so very serious. The dynamo, the wire, and the lamps have all their fixed market value, and if they have to be removed from one installation, they can be utilized in another. Not so with the transmission of energy from some hitherto inaccessible source. The dynamo and the motor have to be built specially for each particular case, and the probability that they can be used elsewhere is small. The line and supports are expensive items, which have only value in that particular locality where they have been erected^ and the works necessary for transforming the crude energy of nature so as to be applied for driving the gene- rating dynamo, have also only a local value. In such cases the installation must be a complete success, or else most of the plant and work is a dead loss ; and it is but natural that capitalists shrink from rushing into enter prises as long as there is the least taint of an experimental nature about them. INTRODUCTORY. 5 Another reason which has, in England at least, operated to delay the electric transmission of energy from natural and inaccessible sources to more convenient places, is that in this country coal is cheap and water-power scarce. In Switzerland and France the case is different, and accord- ingly we find that the first experiments on a large scale have been undertaken in the latter country. Although it is quite incorrect to say, as is frequently stated in French j)apers, that M. Marcel Deprez has invented the electric transmission of energy, or has even invented any special system by which the electric transmission of energy is made practicable, he has been the first to demonstrate that energy can be transmitted electrically over long distances. That his experiments were simply costly failures does not detract from his merit of having early recognized the importance of high voltage, and by draw- ing public attention to this branch of electrical engineer- ing, he has stimulated others^to devote their attention to it. Those who followed him were practical men, not afflicted with any particular system, and in their hands long distance transmission has become a complete success. Broadly speaking, there are two purposes for which the •electric transmission of energy is of great value. The one comprises all cases where, as has been shown above, hitherto inaccessible sources of natural energy are by its means rendered accessible, and the other comprises all those cases where the source of energy itself is accessible, but where it is desired to distribute it to a number of in- dependent small working centres. In the first case we have to transmit a large amount of energy, so to speak, in one lump from the distant source to the place of opera- tion ; and, in the second case, we have to split up the energy of a source close at hand into a number of small 6 ELECTRIC TRANSMISSION OF ENERGY. fragments, and distribute them within a limited area to do useful work. In this case electric transmission of energy comes into competition with the more mechanical means of belts, shafts, wire-ropes, and pneumatic or hydraulic tubes, and the question whether one or the other of these systems is preferable, depends on the amount of energy transmitted, and the distance over which it is transmitted, as well as on many local circum- stances. Electricity has the great advantage of being extremely portable, and capable of having its direction and intensity changed with greatest ease. No mechanical force can be detected in the conductor carrying the elec- trical energy such as appears during purely mechanical transmission with shafting, belts, wire-ropes, or in pipes conveying steam, water, or air. The conductor is clean^ cold, does not move, and altogether appears inert. It can be bent, moved, or shifted in any manner while trans- mitting many horse-power. It might be brought round sharp corners, and, having little weight, it can be fixed with greater ease than any mechanical connection. It is- thus possible to bring the energy into rooms and places awkwardly situated for mechanical transmission, and there is no noise, smell, dirt, or heat during the transit,, nothing to burst or give way. The power is, moreover, under perfect control, and its application exceedingly elastic. The same circuit which may be tapped to give many horse-power can, at the same time, and as con- veniently be used to work a sewing-machine, or other small domestic implement, and the power consumed at the generating dynamo is always in proportion to the power obtained from all the motors, so that there is no^ waste of energy if some of the motors are standing still or are working with less than their full load. In addition INTRODUCTORY. 7 to these advantages^ electrical distribution of energy has also the merit of being exceedingly economical. The commercial efficiency of dynamos and electro-motors sel- dom falls below 80 per cent., and is in many cases as high as 95 per cent., so that even if we make a liberal allowance for loss of energy in the conducting wires, 60 per cent, of the power of the prime-mover at the gene- rating station can be recovered from the motors distri- buted over a limited area. For instance, a steam-engine of 100 horse-power, driving a generating dynamo in the centre of a two-mile circuit, could deliver an aggregate of sixty horse-power in as many separate points within that circuit. Apart from all considerations of nuisance and cost of attendance in the case of sixty separate small steam-engines placed throughout the district, which might be used instead of the sixty electro-motors, it is evident that we can generate one hundred horse-power in one single engine at a far less cost of fuel than could be done in small engines, and although the double conversion necessitated by electrical distribution of energy entails some loss, there is still a large margin in the general economy of the system. In some cases it is found convenient to transmit the energy from the generating dynamo, not directly to the motors, but to interpose between the two a set of accumu- lators or secondary batteries. This is in reality an ex- tension of the system, and has the double advantage of providing motive power even at those times when the generating dynamo is standing still, and also of giving to the motor a certain amount of portability. Electric transmission of power is thus actually carried beyond the limits of a fixed conductor, or is even effected without the aid of a conductor at all. As a case in point, may 8 ELECTRIC TRANSMISSION OF ENERGY. be cited the propulsion of street tramcars by means ot secondary batteries. Here we have a charging station at some place near the line containing some prime mover and dynamos, the current from which is sent by a pair of cables to the secondary batteries in the car which are to be charged. This is the first stage in the electric trans- mission of energy. When the cells are fully charged, the cables are detached, and the car is ready to start, and during its journey the second stage of the transmission, viz., that of the energy in the cells into the motor, takes place. By the employment of secondary batteries, we have thus carried the operation beyond the limits of the cables. If the charging station is so situated that cars can enter it, the process of charging can be accelerated by making each set of cells detachable from the car, and charging them whilst the car, furnished with a duplicate set, is on the line. As each car comes in, its set of ex- hausted cells is replaced by a set newly charged, and can 2:0 out aorain within a few minutes. In this case the actual transmission of energy between the dynamo and the cells, which are placed in close proximity, is only over the spac 3 cf a few yards ; yet this energy may, later on be utilized over a very long line. A similar system is in use for the propulsion of small boats by electricity. It can be most conveniently applied in the case of launches belonging to vessels which are fitted with the electric light ; for the same dynamo which works the incandescent lamps at night can be used to charge, or keep charged, the accumulators in the launch during the day-time, so that the latter may at a moment’s notice be lowered into the sea, provided with a sufficient store of energy for some hours’ run. When the launch is stowed away on deck, its accumulators can INTRODUCTORY. 9 also be used for lighting the vessel, if a mishap occurs to the dynamo, or if it be necessary to stop the machinery for some other reason. Examples of this kind might be multiplied to any ex- tent, but sufficient has been said to show that in the pre- sent state of electrical industry the electric transmission of energy is a question of great practical interest. Its use is not only confined to the transmission of power, pure and simple, between two distant points, as commonly understood, but it enters more or less into every application of electricity. CHAPTER I. General Principles — Lines of Force — Relations between Mechanical and Electrical Energy — Absolute ^Measurements — Ideal Motor and Trans- mission of Energy — Practical Units. A PROPER understanding of the principle of the conser- vation of energy^ which exists throughout the whole of nature, must necessarily form the basis of all scientific in- vestigation of mechanical or electrical problems, and of most of the improvements w^e might attempt to introduce in existing machinery and apparatus. In many cases, the fact that the original amount of energy remains un- changed, whilst the form in which it becomes manifest undergoes many alterations, is easily understood. For instance, if a locomotive engine draws a train behind it on a railway, we are at no loss to explain how the energy of fiuid pressure of steam in the boiler is transformed into that of a steady pull overcoming the resistance of the train at a speed of so many miles per hour, and including all the so-called waste caused by deformation, friction, abrasion, and heating of the bodies through which the energy fiows. The means by which, in this case, energy is transformed are, for the most part, purely mechanical, and sufficiently familiar to our imagination to allow us to form a mental picture of the different processes taking place. Even the transformation of heat into energy of fluid pressure, although we are not able to represent it by a LINES OF FORCE. 11 mechanical model, has, through long familiarity with heat engines in one form or another, become comprehensible to us. With electrical energy, and with that of chemical action, this is not so. We can form no kind of mental picture of the process taking place in a voltaic cell where the energy of chemical action is transformed into that of an electric current, nor can we say what are the connect- ing links by the aid of which this current, after passing through hundreds of miles of wire, is made to impart mechanical energy to the armature of an electro-magnet, and thereby produce telegraphic signals. There is no mechanical connection between the sending key and the lever of the Morse instrument by which energy could be transmitted in the form of a pull, as is the case in our example of the coupling between a locomotive and its train, and yet energy is unmistakably transmitted. If we neglect waste, that is energy transformed in a way not immediately useful for the purpose in view, we find that the amount of electrical energy received at the distant station is proportional to the amount of chemical energy used up ; and if we take the waste also into account, we shall find that the energy it represents, added to that received in the form of an electric current at the distant station, is again proportional to the amount of chemical energy developed in the voltaic cell. If we know the nature of the chemical process going on in the cell, we can always calculate, by the aid of electro-chemical equivalents, what total amount of electrical energy can be obtained from a given weight of materials. Similarly there exists a definite and constant propor- tion between electrical and mechanical energy. The re- lation between the two is somewhat complicated by the development of heat, which, indeed, is inseparable from 12 ELECTRIC TRAXSBIISSION OF ENERGY. electric phenomena, but if we make due allowance for the •energy wasted in heat, we shall find that a given amount of electrical energy will always produce the same amount of mechanical energy, irrespective of the time required, or the exact manner of transformation. Although we cannot say what are the connecting links between electric current and mechanical force, experiment shows that cer- tain definite relations exist, and we can, on the basis of experimental facts, conceive a mental picture or model by the aid of which these relations are represented in a fami- liar form. Such a mental picture is the conception of magnetic lines of force, first introduced by Faraday. In adopting this method of rendering electro-mechanical phenomena tangible to our senses, we make no assump- tion whatever about the reality of the lines of force. Whether they actualiy exist is a matter of total indif- ferenice ; but since all the experiments we can make are €ompatible Avith that conception, and since it enables us not only to explain experimental facts, but also to bring them Avithin the region of actual measurement and calcu- lation, it is convenient to make the theory of magnetic lines of force the basis of electro-mechanical investigations. If a sheet of paper be laid over a straight steel magnet having opposite poles at its ends, and sprinkled Avith iron filings, it Avill be found that these arrange themseh^es in -curves, Avhich Ave take to be the magnetic lines of force, ^ Fig. 1. Each of these lines forms a closed curve ^ A very convenient way of fixing these curves is by the aid of a sheet of glass, the surface of which has been coated with a thin layer of paraffin. Tlie glass is laid over the magnet, then sprinkled, and carefully lifted off so ns not to disturb the filings. It is then gently heated, when the paraffin melts, and upon cooling again the iron filings are fixed to the glass by the coating of paraffin. The glass plate may then be handled as if it were a drawing, and the curves can be reproduced by photography. The drawing in text has been obtained in this manner. LINES OF FORCE. 13 issuing from a point at one end of the magnet, and enter- ing at a corresponding point at the other end. Some of the curves extend far out into space, beyond the surface of the paper, and as far as they are visible, they appear as open lines growing fainter the farther we- go from the poles. They must, nevertheless, be con- sidered to be closed lines, only so faint that we cannot trace them throughout their whole length. If the poles, of our magnet were two mathematical points, all the curves would pass through those points, but since wc Fig. 1. have to deal with a physical magnet, the poles of which are surfaces of some extension, the lines issue from all over these surfaces. To investigate the magnetic properties of these lines we can use a long thin magnetic needle (a magnetized knitting-needle answers very well) suspended vertically by a long thread, so that the lower end of the needle is within a short distance of the paper, and free to move all over it. We shall then find that the lower end of the needle will be repelled by one pole of the magnet and attracted by the other, and in following the combined action of these forces, it will move along that particular 14 ELECTRIC TRANSMISSION OF ENERGY. line offeree upon which it was set on to the paper in the first instance, but it will never move across the lines. W e conclude from this experiment that the lines of force are paths along which a free magnet pole is urged under the influence of the magnet. A free magnet pole of oppo- site sign would travel along the same lines, but in opposite direction, and, if of the same strength, it will be urged along with an equal force. If, instead of a long vertical needle, we take a very short one suspended horizontally in its centre close to the surface of the paper, the two Fig. 2. " V?/ 3 . opposite forces will tend to set the needle so as to form a tangent to the line of force passing through its centre, and as then the two forces are equal and opposite, no bodily shifting of the needle can take place. But on whatever point of any of the curves we set the needle, it will always swivel into such a position that its magnetic axis, that is a straight line joining its two poles, becomes a tangent to the curve. (Fig. 2.) It should here be remarked that unless the needle is very short in comparison to the mag- net it will, when placed near one of the poles, be drawn right up to it, because in this case there would be a sen- .sible difference in the distance of either of its poles from CHAIN OF MAGNETIZED MOLECULES, 15 that magnet pole^ and consequently the opposing forces would no longer be in equilibrium. But if the needle is very short, say only the length of a particle of iron filing, this inequality between the attracting and repelling force will at a short distance from the magnet pole become omissible, and then the particle of iron filing will only set itself into the direction of the line of force in that place, but not move bodily along it. We may thus regard each particle of iron filing which has been sprinkled over the paper as a very short magnetic needle, and each line of force as a chain of such needles linked together by their poles of opposite sign — Si — — 71^ ^ 3 — and so on, as shown in Fig. 2 . Imagine now that the particles in one such chain, whilst under the influence of the magnet, could by some process be suddenly hardened into steel, or that we had taken steel filings in the first instance, and then remove the magnet. We would then have a succes- sion of little magnets, whose poles of opposite sign touch, and therefore eliminate each other, with exception of the first and last particle of the chain. Here we would have a free N pole at one end, and a free S pole at the other end, these being a finite distance apart, and therefore able to exert magnetic action on other pieces of iron placed into their neighbourhood. But let each particle be turned round its centre (without however, shifting it, bodily) so as to break contact with its neighbour, and we shall have a disjointed line of very small magnets, (Fig. 3), none of which is able to exert any magnetic attraction or repulsion at a distance, because on account of the proxi- mity of the two opposite poles in each particle, their dis- tances from any external point to be acted on would be sensibly equal, and consequently the opposite forces would be in equilibrium. By turning each particle so as 16 ELECTRIC TRANSMISSION OF ENERGY. to thoroughly break contact with its neighbour, we have completely destroyed the magnetic action of our chain at a distance. If we had turned only a few of the particles^ or if we had turned all through a very small angle, so as not to completely interrupt their magnetic continuity, the magnetism of the chain as a whole would have been weakened but not destroyed completely. We can restore our magnetic chain again by turning each particle back into its original position, and if this process should be too laborious to be performed by hand, we can accomplish it in an instant by replacing our magnet under the paper,. i is. 3. when its line of force corresponding to the chain of particles, will pass through it again and swivel each into a tangential position, whereby poles of opposite sign are again brought into contact, thus eliminating each other^ with the exception of the two free poles at the ends of the chain. According to the modern theory of magnetism^ as de- veloped by Weber, Wiedemann, Hughes, and others, j what has here been described for a chain of iron filino:s ] lying on the sheet of paper, actually takes place within j ^ Proceedings Poyal Society, ^lay 10, 1883 ; also a paper on “The Cause of Evident Magnetism in Iron, Steel, and other Magnetic Metals,*^ read before the Society of Telegraph Engineers and Electricians, and reported in their Journal, vol. xii.. No. 49. PROFESSOR hughes' THEORY. 17 the body of any piece of iron or steel whilst being magnetized. According to this theory, each molecule of iron or steel is a complete magnet ; it is provided at one end with a definite quantity of magnetic matter of one sign, and at the other end wdth precisely the same quantity of magnetic matter of the opposite sign, and these magnetic charges are an inseparable attribute of matter like its gravity or chemical or thermal properties, and they can neither be increased nor diminished. In an unmagnetized bar of steel these molecular magnets are supposed to form either chains closed in themselves or disjointed chains, their magnetic axes pointing in all possible directions, and therefore, as was the case in our chain of iron filings after Ave had rotated them, incapable of magnetic action at a distance. But if, by some means, it were possible to turn all the molecules so as to point one way, without, however, displacing them bodily, we would obtain a number of parallel magnetic chains showing free magnetism at their ends only, and there- fore capable of exerting magnetic attraction and repulsion at a distance ; in other words, our bar of steel would become a magnet. It will be seen that according to this theory the molecules composing a bar of magne- tizable steel must be capable of rotation around their centres, and the more easily and completely they can be rotated, the greater is the degree of magnetization ob- tained. Since we cannot take hold of each molecule and rotate it mechanically, we must adopt the other method, viz., that of sending lines of force through the bar to perform that work, as we did with the chain of iron filings. This can be done either by the aid of another magnet, or by an electric current. The setting of mole- cules into continuous chains will be the more complete, c 18 ELECTRIC TRANSMISSION OF ENERGY. the less resistance or internal friction they offer to rota- tion^ and the more powerful are the lines of feree which are caused to pass through the bar of steel. In very soft steel, or in soft iron, the molecules rotate freely, and can be set almost completely into continuous chains, but the harder the steel the smaller is the angle through which each molecule can be rotated, and the more magnetizing force is required for this purpose. In such cases the magnetic chains are more or less discontinuous, and the magnetism appearing externally is weaker. On the other hand, the molecules once rotated into position of magnetic continuity are not so easily disturbed again, and thus the harder the steel the more permanent is its magnetization. In soft iron the molecules will lose their magnetic con- tinuity as easily as it was acquired, and the slightest mechanical strain or vibration is sufficient to destroy the greater part of the previous magnetization. To illustrate this we may take a glass tube filled loosely wdth iron filings, which can be magnetized by drawing the pole of a magnet along it. We shall then see that the particles of filing which previously were lying in all possible direc- tions, have now become more or less parallel to the tube, and the whole appears more like a solid piece of iron of very fibrous texture. The tube has now become a magnet, and if it be carefully handled so as not to disturb the arrangement of the particles, it can be used as if it were a solid steel magnet, and all the usual phenomena of attraction and repulsion at a distance can be obtained. But on tai:)ping or shaking the tube the particles relapse into their former confused position, and all traces of ex- ternal magnetism of our tube vanish. From this short outline of Professor Hughes’ theory it will be seen that the only way in which we can act upon the molecules in THE MAGNETIC FIELD. 19 the interior of a bar of iron or steel is by sending lines of force through it. The greater the number of lines, or the more powerful the individual lines which we can force through the bar — or, in other words, the greater the magnetizing power — the greater will be the number of molecules which are thereby arranged into more or less complete magnetic chains, and if the metal is hard enough these chains in their turn become the seat and origin of lines of force, and can then be used to magnetize other bars. It will also be clear that after a bar has been magnetized, the space surrounding it becomes filled with lines of force which emanate from it. Strictly speaking, each magnet is surrounded by lines extending into in- finite space, but practically they can only be traced throughout the space immediately surrounding the mag- net, and this space is called the magnetic JlelcV^ Since magnetic lines are not a reality, but only a convenient conception, we can adopt any'simple way of expressing their magnitude, or, to speak more correctly, the inten- sity of the magnetic field at any given point. We can either assume that the lines are of different strength, and that the mechanical force with which a given free magnet pole is urged along any one particular line is dependent on the strength of that line, which may be different from that of any other line belonging to the same field ; or we can assume that all the lines are of the same strength, but that the number of lines passing through unit space of the field is different at different points of it. According to this assumption, the intensity of the field in any given spot, and the mechanical force exerted on a free magnet pole, is proportional to the number of unit lines passing through unit space at that particular spot. This is the more con- venient way of estimating the magnitude of the mechanical 20 ELECTRIC TRANS3IISSI0N OF ENERGY. forces produced by the magnetic field, but it must not be considered to be a representation geometrically true, and if we try to consider it so, the want of reality in our concep- tion of lines of force becomes at once apparent. This will be seen from the following consideration. If, as we assume, a mechanical force can only be exerted by lines actually passing through the magnet pole, it will be evident that in case the pole be a mathematical point, only one line can pass through it and exert mechanical force on it. This force would therefore be quite independent of the density of lines around the pole. If the pole, although of the same strength, had finite dimensions, more lines would actually pass through it, and more mechanical force would be exerted. Experiment, however, shows that this is not the case, and that within reasonable limits the mechanical force is independent of the extent of the pole, and only depends on its free magnetism. From this we conclude that a strictly geometrical repre- sentation of the density of lines in a magnetic field, in the same manner as we might represent the density of trees in a forest, would be incorrect. We cannot pretend to solve the problem of finding a geometrical representation for our conception of the intensity of the magnetic field, and we must be content to use the term in its conventional sense, without having any clear idea of how it could be represented by a mechanical model. Yet this is no reason why we should abandon such an extremely convenient method of representing magnetic action at a distance. | Nobody has as yet succeeded in explaining the action J of gravitation, or has been able to represent it by a If mechanical model. Nevertheless we find no difficulty in I using the conventional terms of acceleration, mass, and | weight of bodies in our calculations. We know that the I FUNDAMENTAL UNITS. 21 weight of a body equals the product of its mass and the acceleration due to gravity. If we put strength of pole for mass and intensity of field for acceleration due to gravity, we find the analogue to weight in the mechanical force with which a free magnet pole is acted on when placed in a magnetic field. From what has been said above, it will be evident that we must define magnetic field of unit intensity as that in w^hich a free magnet pole of unit strength is acted on with unit force. To define a magnet pole of unit strength we must have recourse to the well-known expression for the mechanical attraction or repulsion existing between two poles placed at a certain distance apart. The law has been established experimentally by Coulomb,^ with the aid of his torsion balance, and verified by Gauss,^ who used for the purpose a large fixed magnet, and a smaller suspended magnetic needle. It is as follows. If M and m denote the strength of the two poles, and if they are placed at a distance, r, from each other, the mecha- nical force (attraction or repulsion according to w^hether the poles are of dissimilar or similar sign) acting between J\T Tn thorn is -p- . If both poles are equal and of the strength 772 , we have and if their distance be unity, the force acting between them will equal the square of the free magnetism of one pole. The force will be unity if the free magnetism is unity. We find, therefore, the definition for unit pole to be a pole of such strength that when placed at unit distance from an equal pole^ the tioo will act upon , be the bars receiving the current, and be the bars in Avhich the current is originated by the movement of the slider 6|, and it will be clear that by performing mechanical work on the latter slider, we can cause the slider a h to give out mechanical work by raising a weight, as explained above. We have here the most simple possible case of the electric transmission of energy. The generating system, (7| can be at any distance from the receiving system, A C D, and all that is required are electrical connections (wires to carry the current) between and 2?, and between and D, Let the intensity of the magnetic field be F, at the EFFICIENCY. 4 ^ generator and F at the receiver, and let the pull applied at the generating slider be Pi, whilst that exerted by the receiving slider is P ; let also and v be respectively their velocities, and and e respectively the electro- motive forces, then the following equations evidently obtain : — e e = F d Vy F^ di Vi — F d V P = Pi di V I — F d V F,d,, Fd, Pi P, d, P F flT This equation shows that the pull exerted on the gene- rating slider, and that given out on the receiving slider, bear a fixed proportion to each other which is indepen- dent of the speed, but depends on the intensities of the fields and on the dimensions d^ d of the sliders. The energy expended at the generating system is W, = Pi d, V, Pi c?i — F d V and that given out by the receiving system is iir J Pi C?1 — F dv W — F d V — — * — The ratio between the two, or the efficiency of transmission, is evidently ___ F d V If both systems are identical as regards dimensions and 44 ELECTRIC TRANSMISSION OF ENERGY. Strength of fields n = This would be the case where two identical dynamos are employed, the one as receiver and the other as motor, both machines being series wound, so that the same current circulates around both sets of field magnets. In such cases it has been custo- mary to determine the electrical efficiency of the trans- mission of energy by simply determining the speeds, and taking their ratio. If no losses and no secondary actions would occur in the connecting wires and in the machines, no objection could be raised to this way of determining the efficiency ; but in practice there are some very serious objections to this method. In the first place, the two magnetic fields, although produced by the same magnetizing power, are not of absolutely equal intensity, because the magnetization of the armature produced by the current circulatino; throimh its coils has a certain in- fluence in altering the intensity of the magnetic field, and this alteration is different in a motor from what it is in a dynamo. In the second place — and this is a fatal objec- tion — any leak or loss of current taking place at some in- termediate point in the wires by which the machines are connected, instead of lowering the efficiency, as deter- mined by the speeds, has actually the effect of making it appear higher than it really is. This will become obvious by reference to the equation for the counter-electro-mo- tive force of the receiving machine. Since e = F d v, any reduction in F, consequent upon the loss of some of the magnetizing current through a leak in the line, has naturally the effect of increasing v, the velocity of the receiving machine, and thus it may happen that through the development of a fault in the insulation of the fine the ratio of speeds will increase, thus showing apparently LOST ENERGY. 45 - an increase of efficiency, whereas in reality the system has become less efficient. The variables in the above equations are P, and ; the dimensions of the machines (or sliders) d and and the intensities of the fields being constant. Since the ratio between the static efforts, P and is also a constant, the number of' variables is reduced to three, and, if two of these are given, the third can be found. As an example, we will take the case that the load P to be put on the receiving machine shall be given (say, for instance, the pull re- quired to haul up a train on a steep gradient, but neglect- ing for the moment the difference in pull caused by varia- tions of speed) and the speed of the generating machine* shall also be given. We require to know the power neces- sary to drive the generating machine, and the speed and energy developed by the receiving machine. From the equation for P, we find immediately the speed of the receiving machine, F^d, rP F d d^' As will be seen, this speed is not directly proportional to- the speed of the generator, and if the latter be increased the speed of the receiver will increase in a somewhat faster ratio. Since the ratio of speeds enters into the formula for the efficiency, it will be evidently advanta- geous to work the machines at the highest possible speed consistent with mechanical safety. On the other hand, if we lower the speed of the generator beyond a certain point the receiver will not be set in motion at all. This will happen if ^ ^ ^ F d P^ r P FdF, d; d- V 46 ELECTRIC TRANSMISSION OF ENERGY. In this case the efficiency is zero. The minimum speed of the generator is therefore dependent on the dimensions of the two machines and on the strength of the two fields, and is inversely proportional to the product of these four ■quantities. The mechanical energy which has to be applied to the . dt generator is yVi = Fv^ ^ ^ ? iind that obtained from the receiver is F, d, F d r F^ d:' the difference between the two being lost, which is represented by the expression r This loss. we must regard as energy transformed in a way not suit- able for the purpose in view. Since it does not appear in the shape of mechanical energy we must expect to find it appearing in the shape of heat, and this is indeed the •case, as can easily be proved. It has been pointed out above that the static pull is the product of current, field- intensity, and the dimension of the machine. The P quotient represents, therefore, nothing else but the current flowing through the circuit, and the above term for the energy lost can also be written in the form r which, as is well known, represents the heat developed by the passage of the current c through a circuit, the electrical resistance of which is r. Thus the whole of the energy applied at the generator is accounted for, partly by that given out by the receiver, and partly by that used up in heating the circuit. It need hardly be men- tioned that the formulas given here for the transmission PRACTICAL UNITS, 47 of energy refer to ideal machines which are free from all other losses, both mechanical and electrical, but that in actual practice these other losses cannot be neglected, and considerably complicate the problems to be solved. The author, nevertheless, has thought it advisable to enter at some detail into the case of transmission of energy by means of ideal machines, not because the formulas obtained are immediately applicable to practical cases, but because they form the basis of other formulas suitably altered for practical purposes, and which will be given in a subsequent chapter. The example cited is also intended to show how easily and simply the system of absolute electro-magnetic measurement can be applied to apparently complicated problems. Before leaving this subject we must refer to the relation between electrical units in absolute measure and those units commonly used in practice. The units as given in the centimeter- gram-second system are of inconvenient magnitude for practical purposes ; some of them are so small that millions and even larger figures are required to express quantities commonly dealt with in practical work, and others are, again, so large as to necessitate the use of fractions. We had already occasion to refer to the three units most often occurring in electro-mechanical pro- blems, viz., current, electro-motive force, and resistance. The unit of quantity of electricity has also incidentally been mentioned as represented by that amount of elec- trical matter which a given current conveys in one second. For the sake of completing the list we must also mention a property of conductors called their capacity^ by which term we mean their capacity or power to hold an elec- trical charge. The capacity is measured by the quantity of electricity with which a body can be charged under an 48 ELECTRIC TRANSMISSION OF ENERGY. electro-motive force equal to unity. The relation between the so-called practical units and their equivalents in the ■^entimeter-gram-second system is as follows : — Name of Electrical Quant Practical Unit. Equivalent c. g. s. Unit. Current strength . . Ampere . . 10" Electro-motive force . Volt. . 10® Resistance . Ohm . 10® Quantity of electricity . Coulomb . . 10" Capacity J Farad \ Microfartid . 10® . 10” Rate of doing work . W att . 10" CHAPTER II. First Electro-motor — Professor Forbes’ Dynamo — Ideal Alternating Cur- rent Dynamo — Ideal Continuous Current Dynamo — Siemens’ Shuttle- Wound Armature — Effect of Self-Induction — Experiments with Electro- motors — Hefner-Alteneck Armature — Gramme Armature — Pacinotti Armature — Electro-motive Force created in any armature. In the preceding chapter it was shown how mechanical energy can be converted into that of an electric cur- rent, and how the electric energy represented by a cur- rent flowing under a given difference of potential can be reconverted again into mechanical energy and do useful work. The apparatus employed for this double conver- sion was assumed to be of extremely simple form, in order to limit our investigation to the fundamental laws with- out obscuring these laws by the introduction of secondary actions and losses. It will now be necessary to confront the subject from a somewhat more practical standpoint, and to show how the conversion between mechanical and electrical energy can be obtained with machinery of a practical form. As a first step towards a practical solution of the problem to produce motive power by an electric current, we must consider Barlow’s wheel,^ in- vented by Sturgeon and Barlow about seventy years ago. A star-shaped wheel was mounted on a horizontal axis and set over a trough containing mercury in such ^ Barlow, On Magnetic Attraction.” London, 1823. E 50 ELECTRIC TRANSMISSION OF ENERGY. way that during rotation of the wheel one or two spokes were always dipping into the mercury. Fig. 10. A per- manent steel magnet N S was placed in such position that the lines of force joining its two poles passed trans- versely across the plane of rotation of the wheel, and upon sending a current through the wheel in the direction in- dicated by the arrows, rotation was produced in the oppo- site sense to the hands of a watch as seen from the side on which was placed the N pole of the magnet. It will be seen at a glance that this apparatus is nothing else but our arrangement of a slider in rotary form, the lines Fig. 10. of the magnetic field being in this case horizontal where they cut through the wheel. Each spoke is a slider coming successively into action as its extremity touches the mercury in the trough and is thus put in electrical connection with the rest of the circuit. It was also found that the experiment succeeded if, instead of a star wheel, a plain metallic disc was employed, the lowest point of the circumference just touching the mercury. In 1831 F araday reversed the experiment and obtained an electric current from a disc rotating between the poles of a magnet. Fig. 11. The magnet was so placed that the induction between the poles, that is, the lines of force passing from one pole to the other, should pierce the FORBES' NON-POLAR DYNAMO. 51 surface of the disc, and the current was taken off by contact springs on the axis and on the circumference ; the latter being placed on the radius of greatest induc- tion. Lately Professor George Forbes has constructed dynamos on the same principle, the only difference being that, instead of using a permanent steel magnet, he uses an electro-magnet which becomes excited by the current produced. Professor Forbes’ machine^ is remarkable for the very powerful current it produces as compared to its small size. He has devised several modifications, but Fig. 11. for our purpose it will be suflScient to describe one of his arrangements. The armature of this dynamo, which is illustrated in Fig. 12 in longitudinal section, consists of a wrought iron cylinder without any wire on it. The field magnet is a closed iron casing surrounding the armature on all sides, and containing two circular grooves of taper- ing section into which are laid the exciting coils jF, formed of insulated copper wire. If a current passes through these coils, it produces lines of force which com- pletely surround each coil, and which pass partly through the iron shell C D forming the field magnet, and partly through the armature A, The general character of these lines is shown by the dotted curves. It will be seen that as the armature cylinder revolves it becomes the See The Engineer” of July 17, 1885. The author is indebted to the editor of that paper fur the use of the engraving. 52 ELECTRIC TRANSMISSION OF ENERGY. held in two cojDi^er rings, which are connected to the two> terminal plates G G, The current is thus taken off all around the armature, and the latter contains absolutely no idle portion. This is one of the reasons why the machines are so j^owerful as compared to their size. The other reason is that the intensity of the magnetic field is very great. As will be shown in a subsequent chapter, when the theory of continuous current motors will be seat of electro-motive forces acting at right angles to the lines, as indicated by the arrows, and if we provide rub- bing contacts at the ends of the cylinder we can obtain the current due to these electro-motive forces. The contacts are arranged at the inner periphery of the exciting coils, and consist of a series of carbon blocks Fig. 12. FORBES’ KON-POLAR DYNAMO. FORBES' NON-POLAR DYNAMO. 53 given, the intensity of the magnetic field is the greater the smaller the clearance between the polar surface of the magnet and the core of the armature. In motors or dynamos, which contain copper wire coiled over the armature core, this clearance is necessarily greater than in Professor Forbes’ dynamo, where the space between armature and magnet is just sufficient to allow of free rotation. The following figures will serve to give an idea of the relation between the size of these machines and their output of electrical energy. A dynamo having an armature six inches in diameter and nine inches in length, will, when driven at a speed of 2,000 revolutions a minute, give a current of 5,000 Amperes at a difference of poten- tial of two Volts. According to the inventor, an arma- ture four feet in diameter by four feet in length would produce an electro-motive force of sixty Volts when driven at a speed of 1,000 revolutions a minute. If we were to allow the current to increase in the same propor- tion as the area of the armature cylinder, this machine €ould produce 320,000 Amperes, and would require about 30.000 h.-p. to drive it. This heavy current would, how- ever, generate more heat in the metal of the armature than could be dissipated at a moderate temperature, and the em- jfioyment of such an enormous power at the high speed of 1.000 revolutions is of course out of the question, but on j)urely theoretical grounds it is interesting to notice how easily our simple experiment of the slider when suitably arranged in rotary form will lead to results which on ac- count of their magnitude are quite beyond the capability of modern engineering. Dynamos similar to that just de- scribed are generally called Uni-polar Dynamos. Pro- fessor Forbes prefers the title Non-polar Dynamos^ and with good reason, for, as was pointed out already in the 54 ELECTRIC TRANSMISSION OF ENERGY. first chapter, a magnet with only one pole is a physical impossibility. All the dynamos of this class have the disadvantage of requiring to be driven at a very high speed in comparison with the electro-motive force they can produce. The reason lies in this, that the length of con- ductor cutting through the field is limited by the size of the armature. These machines are practically nothiog else but dynamos having only one turn of wire wound on their armature core. An ideal machine of this kind is shown in Fig. 13. The field magnets N S are placed Fig. 13. IDEAL ALTERNATING CURRENT DYNAMO. horizontally opposite each other, and their polar surfiices are bored out to form a cylindrical cavity within which one single turn of wire can be revolved by means of a crank. One end of the wire is joined to the axis A A, and the other to a metal sleeve A/, rubbing contact springs B.^ being arranged in order to take the cur- rent off the sleeve and axis respectively. The lines of force pass horizontally across the cylindrical cavity from N to aS', and those which are contained within the space swept by the wire are cut twice during each revolution. The effect is the same as if we had attached our slider to IDEAL ALTERNATING CURRENT DYNAMO. 55 a crank and by turning the latter had caused the slider to assume a reciprocating motion across the lines of the field. In that case^ when the crank is vertical, that is, parallel to the lines of the field, the speed of the slider is a maximum, and therefore its electro-motive force is also a maximum. As the crank approaches either of its dead points, where it is horizontal, the speed of the slider and its electro-motive force diminishes and becomes zero at the moment the motion is reversed. From what was said in the preceding chapter, it will also be seen that the direction in which the electro-motive force acts depends on the direction of motion, and the current produced must therefore be alternating. If we plot the angles of the crank on the horizontal, starting from any given position, say, for instance, from its position at the end of the stroke, and the electro-motive forces on the vertical, we obtain a graphic representation of the relation between these tAvo quantities. In a' uniform field, where the electro-motive force depends only on the speed of the slider, but not on its position in the field, the electro- motive force is evidently proportional to the sine of the angle of the crank, and is given by the expression E = F d 0 ) sin a, Avhere w is the circumferential speed of the crank, and oc its angular position, the other symbols being the same as before. It Avill be seen that E = for a = o and a = tt, whilst for oc = ^^ 0 ^ oL ~ — attains its greatest numerical value, Hi a being positive or negative according to the sign of the angle. The same relations obtain in the ideal alternating current dynamo, Fig. 13. If the crank is in the position shown, the Avire is in the middle of the S pole piece and cuts the lines of force at maximum speed ; if the crank is 56 ELECTRIC TRANS3IISSI0N OF EXERGY, vertical, the wire moves parallel to the lines, and its rate of cutting lines is zero. This position corresponds to the end of the stroke with an oscillating slider.x When the crank is again horizontal, but pointing to the left, the wire is in the middle of the N pole piece, and again its speed across the lines, or its rate of cutting lines, and the electro-motive force are maxima, but the current will be in an opposite direction to what it was at first. If the crank be turned in the direction indicated by the arrow, the current will leave the machine at the contact spring during the time the crank is on the iwht-hand side of the vertical diameter, and it will fiow from through the external circuit, and enter the machine at B^ during the time the crank is on the left-hand side of the vertical diameter. Let n be the number of revolutions per , ix minute, then — 2 97- r = w, the circumferential speed of oO the wire, and the maximum of electro-motive force, irre- spective of sign is evidently E=Fd^^2.r. Xow 2 r cZ is the total space swept by the wire, and jp 2 r 6? is the total number of lines passing through that space ; let 2: be that number, and we find for the maxi- mum of electro-motive force the expression, n E = z 60 1 During one half revolution the electro-motive force in- creases from zero to this maximum, and then decreases again to zero. As far as practical applications of the dynamo are concerned, it is not the maximum electro-mo- tive force which we require to know, but the mean electro- motive force, which acting during the same time as the MAXIMUM ELECTRO-MOTIVE FORCE, 57 variable electro-motive force, would cause the same quantity of electricity to flow through the circuit. Let, at any given moment, the wire occui3y a position defined by the angle a from the vertical, and let it advance through an angle ? a during the time c f, then the quantity of electricity flowing through the whole circuit of resistance B is evidently oq = 6 a F d ct) sin t R F d r R sin a 0 a. and since = r 5 = J t ^ During one half revolution a increases from zero to tt, and the integral of the above exj)ression taken between these limits gives F d ‘. ~~R The time occupied in this transfer of q units of electricity TT r is ^ = — , and if, during that time, a constant electro- motive force were acting, the quantity transferred FJ Tur would be R If this quantity is equal to q, we con- sider the average electro-motive force, and its value is given by the equation E^ = - F do^. 7T Since F do)i'& the maximum electro-motive force generated at the moment when the wire is cutting the lines of the field at right angles, we have also = ^-E. 7T It should be noted that the mean value oi E MF as here defined refers to the total quantity of electricity which 58 ELECTRIC TRANSMISSION OF ENERGY, the apparatus is capable of forcing through a given re- sistance^ but not to the amount of work produced and transformed into heat in the resistance. Inserting the value for E from equation we find the averasre electro- motive force = " z beings as before, the total number of lines contained in 71 the space swept by the wire, whilst — is the number of revolutions per second. In the ideal alternating current dynamo represented in Fig. 13, the wire in which the currents are generated is arranged to one side of the spindle only. We could easily improve the design by carrying the wire symmetri- cally to the other side of the spindle, but insulated from it, and attach its end to a second metal sleeve insulated from M, The contact spring or brush ^2 would then have to be set so as to touch this second sleeve, and since the electro-motive forces created in the two wires are at any moment in the same direction as regards the circuit — although opposite as regards a fixed point in space — this improved dynamo with two wires will give double the electro-motive force of the original arrangement. We could still further increase the electro-motive force by coilins: the wire several times round the axis, formino^ a rectangular coil, each convolution being insulated from its neighbours, and if the number of turns counted on both sides of the spindle is Nt, the average electro-motive force will be For most practical purposes, and especially for the trans- IDEAL CONTINUOUS CURRENT DYNAMO. 59 mission of energy, alternating currents are, however^ not so convenient as continuous currents, and to produce the latter it will be necessary to add to our dynamo a device by which the currents are all directed to flow in the same sense as far as the external circuit is concerned. Such a device is the commutator^ and its action can be explained by reference to Fig. 14. In the position shown, the electro-motive force created in the wire a b will be directed towards the observer, and that created in the wire c d will be directed from the observer. The ends Fig. 14. IDEAL CONTINUOUS CURRENT DYNAMO. of these wires are joined at the back by a cross connec- tion a c, and at the front by two wires df and b to the two halves of a metal cylinder, which for the purpose of insulation are secured on a Avooden hub. The electro- motive forces created in d c and a b tend to draAv a cur- rent from the line in the direction of the arrow to the brush thence through f d^ c b g, to the brush and out again into the external circuit. This process Avill go on until the crank reaches the lower vertical position, the strength of the current meanwhile decreasing to zero. When the crank is vertical, each brush touches simul- taneously both halves of the metal cylinder or com- 60 ELECTRIC TRANSMISSION OF ENERGY. mutator, as it is technically termed, and a moment later the connections become reversed, the brush B.^ now touching the half cylinder to which the wire f is at- tached, and the brush touching the half cylinder to which the wire g is attached. But, at the same time, the direction of electro-motive force in the two wires has been reversed, the wire c d entering the right-hand side of the field, and a b entering the left-hand side. Consequently the external current flows in the same direction as before, growing from zero to a maximum when the crank stands horizontally on the left, and again diminishing to zero when it is vertical. Graphically represented, the current Fig. 15. is of the character shown by the curve. Fig. 15, the abscissae being consecutive angles of the crank, and the ordinates being proportional to the sines of these angles. It should be noted that the reversal of current always takes place when the electro-motive force is zero, and con- sequently the change in the contact with the brushes from one commutator plate to the other takes place without sparking. To increase the power of the machine, we can replace the single rectangle of wire by a coil of many turns. Fig. 16. Hitherto we have tacitly assumed that the space contained within the wire coils forming the armature contains air or other non-magnetic substance. The lines of force passing between the polar surfaces S iVhave to leap across a considerable air space, and if SHUTTLE- WO UND ARM A TURE, 61 by some means we could shorten that portion of their path which lies entirely in aii% we would facilitate the flow of lines and increase the strength of the magnetic field. Roughly speaking, we may take it that air offers to the lines of force about 800 times the resistance of iron, and if Ave can contrive to fill part of the space between the polar surfaces with iron, a considerable in- crease of electro-motive force, and consequently of cur- rent, Avill be the result. The space available for this purpose is that contained within the armature coil ; in other Avords, to increase the poAver of the machine avc must wind the armature coils over an iron core. An early FUr. 16 . dynamo constructed on this principle is that of Siemens, invented in 1855, and provided with the so-called shuttle- wound armature. The core consists of an iron cylinder provided with tAvo deep longitudinal grooves placed oppo- site so that the cross-section resembles a double T Avith rounded heads. The wire is Avound into these grooves, and the two ends of it are joined to the plates of a two- part commutator. Fig. 17 shows a cross-section of this armature. In the first machines the core AA^as in one solid piece, but it was found to heat considerably on account of internal currents. It is well knoAvn that if a solid body of metal be rapidly rotated betAveen tAA^n poAverful magnet poles it becomes hot. The reason for this phenomenon is that the outer portions of the metal .62 ELECTRIC TRANSMISSION OF ENERGY. in cutting througli the lines of force become themselves the seat of electro-motive forces acting at right angles to the direction of motion and to the lines, and powerful currents are started parallel to the axis which run in opposite directions, up on one side and down on the other side of the axis. In a solid armature core there is nothing to check the flow of these currents but the re- sistance of the metal, which, on account of the large cross-sectional area, is extremely low. These wasteful currents are consequently very strong, and not only absorb much power, but they also weaken the current generated in the copper wire by induction. To avoid Fig. 17. SIEMENS SHUTTLE-WOUND ARMATURE. their creation, it is necessary to subdivide the mass of the core by planes at right angles to the axis, and to insulate as much as possible the subdivided portions from each other. This can be done either by cutting deep narrow circular grooves in the core, or by building it up of thin discs insulated from each other either by paper discs or by being coated with some insulating paint. These arma- tures are not much used for dynamos at the present day, having been replaced by more perfect forms to be de- scribed presently ; but they are still extensively employed for small electro-motors. By referring to Fig. 15 it will be seen that the counter-electro-motive force of these motors is a variable quantity depending on the angular position of the armature. If the heads of the double « SHUTTLE- WO UND ARM A TUBE. 63 T core are opposite the field magnet poles, the coil is at right angles to the lines of force and the counter-electro- motive force is zero. This happens precisely at the moment when the brushes touch simultaneously both plates of the commutator, and are therefore short circuited. A current sent through the motor while at rest in this position cannot start it, and this condition is expressed by saying that the armature has two dead points. When at work the momentum of the armature is sufficient to carry it over the dead points, and, apart from the inconvenience to have to start the motor occasionally by hand, these dead points present no mechanical imperfection. But it might be thought that they present a serious electrical imperfection for the following reason : The strength of the current which is allowed to pass through the motor at any given moment depends partly on the electrical resistance of the motor, and partly on its counter-electro- motive force at that particular moment. But since at the dead points there is no counter-electro-motive force, the strength of the current will be a maximum, whilst at those moments the mechanical energy produced is nil. We assume here that the motor is fed by a current flowing under a constant electro-motive force, which is the case most commonly met with in practice. We have now to distinguish between two cases : the motor may be either series wound or shunt wound. If the former, the current is passing through the motor whilst the armature is at a dead point has only to overcome the resistance of the field magnet coils. If the armature is in the position of greatest counter-electro-motive force the current has to overcome not only that, but also the combined resistance of field magnet and armature coils. In that position the mechanical energy of the armature is at its greatest 64 ELECTRIC TRANSMISSION OF ENERGY. value, but the strength of the current is a minimum. We find^ therefore, on the one hand, that the strength of the field magnets (which depends on the current) is least at the very moments when the armature is in a position to exert most power, and on the other hand, that it is greatest when the armature is at its dead points and can- not exert any power. From the foregoing we should expect that twice during each revolution a great w^aste of current must take place when momentarily the brushes are short-circuited by the commutator. Although the time during which such short circuits lasts may appear to our senses very brief, it would in comparison with the speed of electric phenomena be still considerable, and have an appreciable effect on the economy of the motor. But there is one circumstance which greatly tends to mitigate the evil effect of the dead points just described, and this is the property of electric currents called self-induction. It can best be described as a kind of inertia opposing any sudden change in the strength of the current. If a circuit contains a coil of wire surrounding iron (as in the present case the field magnets) the self-induction is so great that it requires an appreciable tiriie to change the strength of the current. The increase of current at the dead points is, therefore, checked by this property of self- induction, and the current, instead of being subjected to abrupt and violent changes, becomes simply undulatory. The case is different if the motor be shunt-wound and fed from a source of constant electro-motive force. Since the field magnet coils are excited independently from the current which passes through the armature, their self- induction cannot in any w^ay steady that current, and abrupt changes in its strength and great w^aste of electrical energy must occur at the dead points. This is EXPERIMENTS ON SELF-INDUCTION. 65 a matter of considerable practical importance, and shows that motors with shuttle-wound armatures should never be used coupled up otherwise than armature and field magnets in series. If it be absolutely necessary to use a motor of that class, the field magnets of which are either permanent steel magnets or are electro-magnets excited independently, the waste can to a certain extent be pre- vented by inserting into the armature circuit an electro- magnet which will by its self-induction steady the current. Since this point is of importance, the author has thought it necessary to verify the above theory by experiments. These were undertaken with a twofold object. First, to prove that in a series-wound motor there is no appreciable waste of current at the dead points, and, secondly, to prove that in a motor the field magnets of which are separately excited, such waste occurs. The experiments were carried out as follows. Two small Griscom motors were placed in line behind each other, and their spindles were coupled, so that the armatures stood at right angles to each other, that is to say, when one armature w^as at its dead point the other was in the position of best action, and its counter-electro-motive force was a maximum. This disposition is represented in Fig. 15 by the dotted curve overlapping that shown in full lines by 90"^. The resultant counter-electro-motive force is at any point the sum of the ordinates of the two curves, and is shown by the undulating line a b. It will be seen that this curve nowhere touches the horizontal and, therefore, the total counter-electro-motive force of the two motors coupled in series never is zero. An abnormal rush of current at the dead points of any of the armatures can, therefore, not take place. The motors were supplied with a current, the electro-motive force of which was kept as nearly as 66 ELECTRIC TRANSMISSION OF ENERGY. possible constant during each experiment, whilst the mechanical energy developed was measured on one of the author’s absorption dynamometers. The commercial effi- ciency of the two motors combined was thus ascertained, as shown in Table I. The motors were then coupled parallel, and their efficiency was determined under the same conditions. In this case there were, during each revolution, four dead points, at which the counter- electro-motive force was zero, and Avhen an abnormal rush of current could take place if not checked by the self-induction of the magnet coils. As was to be ex- pected, the current passing through both motors was about double, and its electro-motive force was about half of the former values. But the commercial efficiency was about the same. Table II. One motor alone was then tried, and its commercial efficiency was found to be about the same as that of the two motors combined. Table III. The field magnets of both motors were then excited separately, and the armatures coupled at right angles and connected in series, as per Fig. 15, when the com- mercial efficiency was found to be rather higher than in the former experiments. Table IV. This is but natural, because the energy necessary to excite the field magnets was not taken into account when calculating the efficiency. The two armatures were then coupled parallel — field magnets still independently excited — and thus during each revolution there were four points where the counter- electro-motive force w^as zero and waste of current did take place, as is clearly shown by the low efficiency in Table V. One motor alone was then tried under the same conditions and the same result was found, Table VI. These experiments prove conclusively that our above reasoning about the effects of the dead points is correct. EXPERIMENTS OF SELF-INDUCTION. 67 Test of Two Griscom Motors, Numbers 1017 and 1027. Resistance of . Armature Magnets Total • 1017 . •328 . •596 . •924 . . 1027 •352 •522 •874 Table L Armatures Coupled at Right Angles, both Field Magnets and Armatures connected in Series, Revolutions per minute. Current. E. M. F. Foot Pounds on Brake. Commercial Efficiency 7<>* 2,440 1-31 6-90 0 0 2,368 3-85 18-20 588 19-0 2,440 3-50 16-00 535 21-7 Table II, Armatures Coupled at Right Angles, Each Armature in Series with Us Field, Both Motors connected Parallel, Revolutions per minute. Current. E. M. F. Foot Pounds on Brake. Commercial Efficiency ®/o. 2,120 2-35 2-94 0 0 2,480 5-25 6-05 206 14-7 2,775 6-60 7-57 432 19-5 2,340 6-80 7-52 366 16-3 2,060 7-50 7-63 450 18-0 2,884 7-90 9-27 748 23-0 2,328 7-60 8-50 578 21-0 68 ELECTRIC TRANS3IISSI0N OF ENERGY. Table III. One Motor only. Armature and Field Mag- nets connected in Series. Revolutions per minute. Current. E. M. E. Foot Pounds on Brake. Commercial Efficiency ‘^/o. 1,980 1-02 4-00 0 0 2,024 4-15 8-20 303 28-0 1,772 4-15 8-40 265 17-0 2,334 4-22 9-25 381 22-3 1,954 3-82 8-10 246 18-0 2,241 3-70 8-25 283 20-9 2,118 3-50 7-60 240 20-5 2,070 5-37 12-00 532 18-6 Table IV Armatures coupled at Right Angles and Con- nected in Series. Field Magnets excited separately. Revolutions per minute. Current. E. M. F. Foot Pounds on Brake. Commercial Efficiency 1,536 1-42 7-20 0 0 2,030 3-30 11-10 370 22-8 1,632 3-10 9-50 300 23-2 2,190 3-70 12-90 483 22-7 2,264 3-93 13-40 500 21-4 Table V. Armatures Coupled at Right Angles and Con- nected in Parallel. Field Magnets excited separately . i Revolutions per minute. Current. E. M. F. Foot Pounds on Brake. Commercial Efficiency 2,000 3-90 4-40 0 0 3,040 4-50 5-20 0 0 1,094 7-50 5-50 242 13-3 1,746 8-50 6-60 385 15-6 1,680 9-10 7-50 396 13-1 HEFNER- ALTENECK AR3IATURE. 69 Table VL One Motor only. Field Magnets excited separately. Eevolutions per minute. Current. E. M. F. Foot Pounds on Brake, j Commercial Efficiency ®/o. 1,778 1-65 • 3-80 0 0 2,330 4*80 5-60 87 7-4 2,422 4-75 5-80 126 10-3 As already mentioned, motors with ordinary shuttle- wound armatures have the disadvantage of requiring to be started by hand if they happen to have stopped on a dead point. They are, consequently, only made of small size, and for larger motors armatures without dead points are used. Such an armature can be evolved out of the simple shuttle-wound pattern by employing two sets of coils placed at right angles to each other. This arrangement is shown in Fig. 18, which represents the Hefner- Alteneck winding invented in 1872. In order to avoid complication the shaft is omitted and the core is indicated by two dotted circles. From what has already been explained it will be seen that in all those wires which at a given moment lie on the right hand side of the vertical centre line, the electro-motive force is directed towards the observer, and in all the wdres lying to the left of that line it is directed from the observer. The diameter of commutation joining the points of con- tact of the brushes with the commutator cylinder will, therefore, be horizontal. In the position shown the nega- tive, or left brush, will touch segment Z>, and the right or positive brush will touch segment B, The current enters the armature at the negative brush and splits into two circuits as follows : — One portion goes through VII., 70 ELECTRIC TRANS3IISSI0N OF ENERGY. 7, 8, yill., I., 2, II., and out by the positive segment B ; the other goes through VI., 6, 5, V., IV., 4, 3, III., and out by the same segment B. The two currents are, therefore, in parallel connection. When the armature has turned so far as to bring the segment C into contact with the negative brush it will touch for a short time both segments D and C, whilst the positive brush will Fig. 18 . simultaneously touch A and B. In this position the wires I., VI., V., II. will be in the strongest part of the field, and the wires VII., IV., III., VIII. will stand on the vertical diameter and contribute nothing towards the total electro-motive force. The current now splits into the following t'wo circuits : From D to VI., 6, 5, V., to and from C to I., 1, 2, II., to B. In this case the total HEFNER- ALTENECK ARMATURE. 71 electro-motive force is that due to two wires in the position of best action, whereas in all the other positions it is due to four wires. It has been shown above that the average electro-motive force of a loop such as I., 1, 2, II., con- sisting of two external wires (iW = 2) is Since two such loops are placed in series, we find the average electro-motive force of the whole armature But 8 is the number of wires counted all round the armature ; and if, instead of a four-part commutator, we had employed a six-part commutator, and had wound the core with three sets of double coils, we would have three coils in series and the expression for Ea would have been there being twelve external wires on the armature if counted all around. We might thus construct armatures with any even number of external wires. Let Nt be that number, and we have the general expression for the electro-motive force created in the armature of a dynamo, or the counter-electro-motive force created in the armature of a motor : 11 Ea^NtZ ~ 3 60 For the sake of simplicity we have, in Fig. 18, only shown one wire to each coil. It is, however, obvious that by multiplying the turns or wires in each coil the electro- motive force can be proportionately increased. This case is provided for in formula 3, where N signifies the number 72 ELECTRIC TRANSMISSION OF ENERGY, of coils, and t the number of turns in each coil, the pro- duct of the two being equal to the total number of single wires if counted all around the armature. An armature of the Hefner- Alteneck pattern with eight-part commu- tator, is shown in Fig. 19. Denoting by Homan figures the ends of the wires on the front end of the armature. Fig. 19. IlEFNER-ALTENECK ARMATURE. and by Arab figures those on the rear end, the winding is as follows : From the negative brush to I., 1, 2, II., III., 3, 4, IV., V., 5, 6, VL, VII., 7, 8, VIII. XVI., 16, 15, XV., XIV., 14, 13, XIII., XII., 12, 11, XI., X., 10, 9, IX. To the positive brush The greater the number of parts in the commutator the more nearly constant will be the electro-motive force and current. This system of winding armatures has the great advantage of utilizing nearly the whole length of the wire, since, with the exception of the cross connections at the ends, all the wire is active. But it has the practical disadvantage that repairs are troublesome to execute. If a fault of insulation should develop in any of the coils. GRAMME ARMATURE. 73 in order to reach it^ a large portion of the wire must be taken off^ because the coils — especially at the ends — overlap each other in many layers. In this respect the style of armature known as the Gramme, or Pacinotti type, is preferable. A circular iron ring, Fig. 20, is Fig. 20 . wound with a continuous helix of insulated copper wire, and certain points of the helix are joined by connecting wires, which in our illustration are shown radial, to the commutator plates. Two brushes, and B. 2 , serve as connections between the external circuit and the armature wire. The action of the Gramme armature will best be explained by reference to Fig. 21, which shows the lines of force. It has already been pointed out that iron offers very much less resistance to the passage of magnetic lines of force than air. If there be no armature between the field magnet poles, we assume that the majority of the lines will go straight from pole to pole. Fig. 22. If now a circular core is inserted, their course will be so altered that each line takes the path of least resistance — that is, runs as long as possible in iron, and only leaps across the 74 ELECTRIC TRANSMISSION OF ENERGY. air at the external circumference of the core, because this is the only way in which it can enter the pole piece. Fig. 23. At the internal circumference of the armature Fis:. 21. there is no necessity for the lines to leave the core, and the central space is therefore almost free of lines. We say almost, because parallel lines exert a repelling action upon Fig. 22. FIELD OF DYNAMO WITH ARMATURE REMOVED. each other, and it may happen that in case the core is thin, and a large number of lines have to be accommodated, some of them may be elbowed out into the central space. FIELD OF DYNA3I0, 75 In well-designed machines the number of lines thus forced across the central sj)ace is always so small as to be omissible. The fact of the central space being free from lines ; oi% as we may also put it, being shielded by the iron of the core from the influence of the magnet poles is of great importance, since in consequence of it the inner wires of the helix are removed from all induc- tive action. If this were not the case electro-motive Fig. 23. FIELD OF DYNAMO WITH ARMATURE INSERTED. forces would be created in these wires, which, belne* opposed to the electro-motive forces developed in the external wires, would weaken the power of the machine. After what has been explained at length with reference to the ideal continuous current dynamo. Fig. 14, it will be easy to trace the direction of electro-motive forces in the external wires of the Gramme armature, Fier. 20. If rotated clock-wise, the electro-motive force will be directed towards the observer in all the wires lying to the right of the vertical centre line, and from the observer in the wires 76 ELECTRIC TRANSMISSION OF ENERGY. on the opposite side. The two currents resulting from these forces are indicated by the arrows. In the wires 1 and 7, which for the time being move parallel to the direc- tion of the lines of force, there is no electro-motive force generated, whilst in 4 and 10, which move at right angles to the lines, the electro-motive force is a maximum. By virtue of the continuity of the helix the electro-motive forces in the wires 2, 3, 4, 5, 6 are added, and those in 12, 11, 10, 9, 8 are also added, the two circuits being at all times in parallel connection. The current enters the armature at the brush which is called negative, then splits into the two circuits mentioned, and uniting again at the brush which is called positive, leaves the armature, and enters the external circuit. It will be seen from the figure that either brush, when touching two consecutive plates of the commutator, establishes a metallic connection between the beginning and end of the corresponding coil, or, in technical language, short circuits that coil. If the brushes are in the position shown — the neutral diameter on the commutator — the short circuit is perfectly harmless, because there is no electro-motive force in the coil ; but if we were to shift the brushes into an active part of the field either to the right or left of the neutral line, each coil, as its extremities pass under the brush, would be traversed by an excessive current, causing heavy sparking at the brush, and probably the ultimate destruction of the armature. The best position at which to place the brushes is always found experimen- tally ; it does not accurately coincide with the geometrical neutral line, but is found to be in dynamos slightly in advance of it, and in motors slightly behind it. Opinions are divided as to the reason of this phenomenon. At one time it was ascribed to a certain sluggishness in the iron GRAMME ARMATURE. 77 of the core in taking up and losing magnetism, but this theory has long since been discarded by most practical electricians. Some hold that the shifting of the neutral line is due to the magnetizing influence of the armature current upon the iron core by which the latter is trans- formed into a double horseshoe magnet with like poles joined, and the magnetic axis of which stands nearly at right angles to that of the field magnets. Others again maintain that the brushes must be set forward in a dynamo and backward in a motor, on account of the in- fluence of self-induction in the armature coils. In reality both the last-mentioned causes have something to do with the position of the brushes, as will be more particularly explained in Chapter IV. The first electro-motor having an armature Avound on the principle above explained, was constructed by Pro- fessor Pacinotti, of Pisa, and the design was published in the journal II Nuovo Cimento,” in 1864. This machine is illustrated in Fig. 24, and the core of the armature differed only in so far from that employed by Gramme seven years later, as it had external projections between the wire coils, which considerably increased the magnetic attraction between the armature and the pole pieces, thus rendering the machine more powerful. Fig. 25 shows part of the core and winding. The core of the Gramme machine consists of iron wire coiled into a ring of oblong cross-section. After being lapped round with tape for the purpose of insulation, it is wound transversely with cotton-covered copper wire. The winding consists of a number of coils which cover the core completely inside and out, and the beginning of each coil is joined with the end of its neighbour to the same commutator plate. When the winding is completed the armature is driven tight 78 ELECTRIC TRANSMISSION OF ENERGY, over a wooden centre by which it is fastened to the spindle. By means of the fundamental formulas established in the previous chapter, we can now determine the electro- Fig. 24 . motive force of a Gramme armature. Let D be its dia- meter, h its length, and a the radial depth of the core. PACINOTTI ARMATURE. Let Nt represent the total number of external wires, counted all around the circumference, t representing the number of wires corresponding to one jdate in the com- mutator, and N the number of plates. If n denotes the FACINOTTI ARMATURE, 79 speed in revolutions per minute^ and z tlie total number of lines emanating from one pole and entering the halt circumference of the armature, then the average electro- motive force created in each wire is by equation 2, Since Nt wires are for the time being connected in series, the average total electro-motive force in the arma- ture is = ^ It might be objected that this expression, which is based on equation 2, will only be correct if the condition under which this equation was obtained is fulfilled in the dynamo. This condition was that the field should be perfectly uniform throughout the space occupied by the armature. In reality this is never the case, and the exact distribution is not accurately known. A doubt might therefore be entertained whether equation 4 be rigorously true in the case where the intensity of the field is not uniform, but varies in different parts of the field. It will consequently be desirable to deduce the formula for the electro-motive force under the supposition that the intensity of the field in any point on the circumference of the armature, is a function of the angle, a, which the radius to that point forms with the neutral line. What that function is we cannot say, nor is it necessary that we should be able to define it. We only make this assumption : that there shall not be any abrupt changes in the strength of the field. AA^e assume that the density of lines varies gradually from place to place. Assume also the number of wires on the armature so large, that 80 ELECTRIC TRANSMISSION OF ENERGY, their angular distance A a is very small^ in fact so small that the intensity of the field can be considered as con- stant within that angular distance. Since the electro- motive force created in the Avires is proportional to their speed, Ave can determine it for any convenient speed, and if it be required for a different speed, Ave can obtain it by multiplying the result first obtained with the ratio of the tAvo speeds. In the present instance we fix as a conve- nient speed that which Avill bring each Avire at the end of one second into the position occupied by its immediate neighbour at the beginning of the second, or ^ D V IX a, — . This is a A^ery sIoav speed, and if Ave wish to knoAv Avhat Avill be the electro-motive force at the faster speed of n revolutions a minute, Ave shall have to multiply the electro-motive force at the Ioav speed Avith the ratio of — tt . Z) and v. Since A a Nt = 2 we have also 60 V = f^he ratio of the tAvo speeds is n 97 D 7T D Wt Let iq, .. F ^ he the intensity of the field at the first, second, . . . Avire, counted from the neutral line, on one-half of the circumference of the armature, then ELECTRO-MOTIVE FORCE IN ARMATURE, 81 the electro-motive force in these wires will be given by the expressions, E^ = F ^ b V En t — F Ntb V, ~2~ The sum of all these forces gives the total electro-motive force created within the armature, which we denote in future by Fa. E a = 'Z F h V. But the expression F^hv represents the number of lines contained between the first and second wire on the arma- ture, since is the density, and b v the area of the space swept by the first wire in one second. Similarly F.^^b v represents the number of lines between the second and third Avire, and so on, the sum of all these expressions re- presenting the total number of lines entering behveen the first and last wire on one-half circumference of the arma- ture. Let 2 : be that total number, and we find for the electro-motive force at the low speed. Fa = 2 ’. At the high speed we have, therefore. 4 ) precisely the same expression as already obtained above. If 2 be inserted in absolute measure, Ea will also be obtained in absolute measure, and to obtain it in volts the right side of the equation must be multiplied with 10-*. We can also write 82 ELECTRIC TRANSMISSION OF ENERGY. and if we measure the field intensity by means of a unit 6^000 times as great as the absolute unit, we can further simplify the equation to E, = Z Ntn 10-^ 5) Z being the total number of lines in the new system, which is related to the absolute system by the equation Z = -^ 6000* The cross-sectional area of the armature core is 2 a i, and if we denote by m the average density of lines per square inch of armature core, we have, Z = 2 a h and inserting this value in 5), we find for the electro- motive force also the expression, E^~ 2 a b m Nt n lO'"^ 6). This expression is sometimes more convenient than the former, because it enables us at once to see how the dimensions of the armature affect the electro-motive force. Experience has shown that the density of lines, ruy in the core cannot exceed a certain limit, which is reached when the core is saturated with magnetism. This limit is m = 30, but in practical work a lower density is gene- rally adopted, for reasons which will be explained in the following chapter. A fair average value in good modern dynamos and -motors is m = 20, and the area, a i, must be taken as that actually filled by iron, and not the gross area of the core. To avoid waste of power and heating, the armature core of dynamos and motors must be sub- divided into portions insulated from each other, the planes of division being j^arallel to the direction of the lines of force, and to the direction of motion. The space wasted by such insulation must be deducted from the gross area of ELECTRO-MOTIVE FORCE IN ARMATURE. 83 the core, and the remainder — from 70 to 90 per cent, of it — is the portion actually carrying lines of force. The electrical energy developed in the armature, if a current c be flowing through its coils, is c, and the horse-power represented by this energy is H-P = c 2 a b m Nt n 10*\ 746 The power to be applied must naturally be somewhat in excess of this in order to overcome mechanical resistances, as friction in the bearings and air resistance, and also the magnetic resistance due to imperfect subdivision and heating of the core, and reaction of the armature on the magnets. In good dynamos these losses do not exceed about 10 per cent, and may even be less. / CHAPTER III. Reversibility of Dynamo Machines — Different conditions in Dynamos and Motors — Theory of Motors — Horse-power of Motors — ;Losses due to Me- chanical and Magnetic Friction — Efficiency of Conversion — Electrical Efficiency — Formulas for Dynamos and Motors. After what has been explained in the previous chapters it will be evident that dynamo machine and electro-motor are convertible terms. Any dynamo can be used prac- tically as a motor, and in most cases any motor can be used to generate a current. On purely theoretical grounds this should be possible in all cases, but in practice it is found that the speed which is required to make some small motors act as self-exciting dynamos is so high as to render that application mechanically impossible. The reason for this is, that in small motors the polar surfaces are of very limited extent, and consequently the magnetic resistance of the path traversed by the lines of force is excessively high, requiring more electrical energy to excite the field magnets than the armature is capable of developing at a moderate and practical speed. This point will be more fully explained further on. For our present purpose it suffices to note that on purely theoretical grounds the same machine can act as a motor or as a dynamo. A separate investigation as to the theory of motors might, therefore, almost seem superfluous. But, on the other hand, experience has shown that although THEORY OF MOTORS. 85 this reversibility of the dynamo machine exists, it is not always the best dynamo which makes the best motor, and that certain details have to be altered according to the use for which the machine is intended, if we wish to pro- duce the best possible machine for each purpose. The conditions which have to be fulfilled in the case of dynamos are also generally different from those required in motors. The dynamo must have a high efficiency, it must be able to work continuously without undue heat- ing in any of its parts, must not be injured by an occa- sional excess of current, and must work equally well at extreme variations of electrical output. Its weight is, as a rule, of secondary importance, and in many cases there is no objection to large weights. The motors, on the other hand, are generally required to be of the smallest possible weight, they work intermittently, and high effi- ciency, although desirable, is not of so much importance, especially not in small motors. In the early days of electric transmission of energy the difference between the conditions in dynamos and motors was overlooked, and the usual arrangement was to employ two identical machines, one acting as generator, the other as receiver, but at the present time this rough-and-ready method does not satisfy all the requirements which can justly be made, and special motors must be provided. It has thus be- come necessary to study the theory of motors apart from that of dynamos. Let in Fig. 26, iV iS' be the pole pieces and D the mean diameter of the annular space filled by the external wires on a cylindrical armature of the Gramme or Hefner- Alteneck pattern. Let, as before, b represent the length of the wire and F the intensity of the field at a given point Ry the radius to which forms with the neutral line 86 ELECTRIC TRANSMISSION OF ENERGY, the angle a. All the wires on the upper half of the arma- ture will be traversed by currents flowing in the same direction^ say from the observer, and all the wires on the lower half will be traversed by currents flowing towards the observer. Let c be the current in each single wire and let there be Nt external wires counted all around the circumference. If these wires lie close together with only as much space between them as is necessary for mutual insulation, the effect of the current c traversing Nt successively the — wires on one half of the circumference will evidently be the same as that of a semicircular sheet Nt of current of total strength — c, the width of this sheet measured transversely to the direction of flow being . The density of current in the sheet, that is, the strength n . V • Nt 7T D Nt c, ot current per unit ot width, is — c : — — - = 2 TT U and the current flowing down an elementary section at THEORY OF 3I0T0RS. 87 R, the angular width of which (A a) we take to be very small, is, . NtcD ^ ~ 7T D 2 ^ The mechanical force tending to rotate the elementary strip of our sheet of current in the direction of the arrow is A P = F b A c AP-Fb-Aa^jy Now the intensity of the field, multiplied with h ^ A oL^ the total area of the elementary strip, gives the number of lines of force which enter the core through that area. Let A Z represent that number, and w^e can also write .Ntc AP = A Z D' Now consider a second elementary strip of the sheet of current contiguous to the first. The force exerted by this strip will be represented by a similar expression, but in it the value of A ^ may be different. This will be the case if the field intensity is not uniform, but varies in any way with the angle a. For our purpose it is not necessary to know in what manner the intensity of field F may vary in different points ; whatever the law of variation may be, the sum of all the values oi A Z must always be the same, and equal to the total number of lines passing into the armature core. The mechanical force exerted by the upper semicircular sheet of current, or, which comes to the same thing, by the upper half of Nt the armature winding, -- , is therefore 88 ELECTRIC TRAESMISSION OF ENERGY. Z being the total number of lines. Simultaneously the lower half of the armature exerts the same force, and we have the total force tending to rotate the armature, and acting at a radius equal to that of the winding, D 2 ’ 2 Z Nt c •n D The turning moment, or torque. or 7). 7T If we express the total number of lines by the product of their density within the armature core and the dimensions of the latter, we can also write for the torque T = 2 ah m Kt c 7T 8 ). It has already been mentioned that there exists a limit beyond which m cannot be increased, however powerful the field magnets may be. Assume that in two motors of different size the field magnets are excited so as to produce equal and maximum density of lines in both armature cores, and assume also that both armatures are wound with wire of the same gauge, then the number of turns wdll in the larger machine be greater than in the smaller, the proportion being evidently as the squares of their linear dimensions. Since the areas of the cores are also in the same proportion, it follows that the torques or turning moments are in the proportion of the fourth power of the linear dimensions. Thus, if the larger motor be double the linear dimensions of the smaller, its TORQUE EXERTED BY ARMATURE, 89 torque will be sixteen times as great. It will be seen from formula 7, that the torque of a motor depends only on the strength of the field and on the current, but does not depend on the speed. This can be shown experimen- tally in the following manner. Let two series-wound dynamos be connected by a pair of cables, and let one of these act as generator, whilst the other, which is the motor, is provided with a friction brake, on which the energy given out can be measured. AVhatever the speed of the motor may be, the brake, if its lever be floating free, indicates the turning moment in the shaft of the motor. This turning moment is equal to the product of the length of the lever and the load suspended. If now the speed of the generator be varied so as to vary the electro-motive force, the speed of the motor will accord- ingly vary, but the current and the load on the brake will remain unaltered. In dealing with this matter, M. Marcel Deprez, in La Lumiere Electrique ” of the 3rd of October, 1885, says: — If a current traverses a motor having an armature of the Pacinotti type, the turning effort of the latter is independent of its state of movement or rest, and in motion it is independent of the speed, provided the strength of the current is maintained constant. Inversely, if the static moment tending to resist the motion of the armature is maintained constant, the current will thereby automatically be kept constant, whatever means we may employ to vary it. The experi- ment must be made in the following way. Mount upon the spindle of the motor a self-adjusting dynamometric brake, the load on which is automatically kept constant whatever variation may take place in the friction of the brake or in the speed of the motor, so that the tangential resistance which tends to oppose rotation shall be kept 90 ELECTRIC TRANSMISSION OF ENERGY', constant. Supply the motor Avith current from any given source of electricity (a battery or a dynamo machine), and note the strength of the current and its electro- motive force. If the latter be gradually increased from zero we observe that as long as the motor remains at rest the current groAvs in the same proportion, but as soon as it has reached a certain value and the motor has begun to turn, the current does not further increase, although the rise in the electro-motive force may con- tinue, and with it the rise in the speed of the motor. In an experiment made three years ago the source of elec- tricity Avas a Gramme dynamo and the motor a Hefner- Alteneck machine, the brake being loaded Avith 5^ lbs. at a radius of 6|- inches. When the motor began to turn, the needle of the ampere-meter indicated tAventy-six divisions. I then augmented the speed of the dynamo until the motor made thirty -tAvo revolutions per second, and yet the ampere-meter only indicated twenty-seven divisions instead of tAventy-six.” Since Avith a constant load on the brake, the energy given out is proportional to the speed, and since the electrical energy supplied to the motor is the product of current and electro-motive force, it folloAvs that if the current is constant the speed must be proportional to the electro-motive force. The folloAving table taken from M. Marcel Deprez’s article shows that this is indeed the case. It Avill be seen that in all the four motors tested the ratio of electro-motive force to speed remained nearly constant throughout a very Avide range of speed, and that the current also remained practically constant TORQUE INDEPENDENT OF SPEED, 91 Electro-motive Type of motor. Uevolntions Current. force. per mifiuie. — Speed. 425 13-53 -0267 Hefner-Alteneck . - 783 1165 12-68 13-65 -0262 -0278 1660 13-00 -0250 270 8-16 ' -06496 ~ 526 8-16 -06437 608 8-23 -06768 A Gramme . • . ^ 742 8-40 -06792 944 8-23 -06713 1004 8-23 -06803 1160 8-23 -06704 1460 8-23 -06736 / 356 5-60 ’ -0132 1 618 5-78 -0139 1016 5-42 -0127 Hefner-Alteneck . ^ 1 1236 5-60 -0130 1470 5-95 -0129 1636 5-60 -0127 [ 1662 5-42 -0127 r 200 ^ 5-60^ ”1-659 High tension ma- 1 chine .... 384 6-30 1-692 470 606 6-12 5-95 1-775 1-633 710 5-95 , 1-662 ' Going now back to equation 7), the mechanical energy represented by one revolution of the motor shaft is evidently 2 tt and if the motor runs at a speed of n 71 revolutions a minute, or — - revolutions a second, the dO energy developed during that time is W = Z Nt 2c ~ 60 9 ). 92 ELECTRIC TRANSMISSION OF ENERGY. It will be remembered that each half of the armature carries the current c ; 2c is consequently the total cur- rent passing into the armature at one brush and out at the other. Write Ca (armature current) for 2c and we have W= Z Nt~Ca 10). b(J But from equation 4) we found that the counter electro- motive force of the armature is Ea ryr n 4 ). and combining the two equations we find Ea Ca 11). The mechanical energy equals the product of current and electro-motive force, that is, equals the electrical energy. This, indeed, is self-evident from the principle of the conservation of energy ; and starting with the equations 4) and 11), we could have deduced the expres- sions for IV and T from these. But on the other hand it is more satisfactory to have determined these values inde- jDcndently, and to find that our conclusions are verified by the principle of the conservation of energy. All the equations above are based on the absolute system of measurement. For practical purposes, how- ever, the employment of these units is not convenient, and instead of using dynes or ergs we prefer to make our calculation in pounds and horse-powers. It will therefore be necessary to determine the relation between the abso- lute and practical units. According to the definition of the dyne given in the first chapter, it is that force which accelerates the mass of one gram by one centimeter in one second. It would ENERGY GIVEN OUT, 93 not be strictly correct to represent the dyne as equal to a certain fraction of a kilogram or of a pound, because the weight of unit mass (that of one gram) changes according to the position on the surface of the earth where we may happen to measure it. But in all places the following equations hold good : — P = m G = m g, P being the force to which corresponds the acceleration p, G being the weight of the body measured by the acceleration of gravity g, and m being the mass of the body P = G^. 9 If g be given in meters per second and the weight in kilograms, the force of one dyne is, Dyne = 10 -^ 10 - ~9 kiloofr ams. 10 "" Dyne = 9 kilograms. The energy represented by one dyne acting through the distance of one centimeter, the erg, is therefore . . . kilogram-centimeters, or, Ji,rg = — — . . . kilogram-meters. According to equation 11) the number of ergs developed by the armature of the motor is numerically equal to the product of current and electro-motive force in absolute measure. If we wish to insert these values expressed in practical units of amperes and volts we have II^=10®xl0 + ^... volt amperes, W — 10“'^ watts. 94 ELECTRIC TRANSMISSION OF ENERGY. To obtain the number of watts represented by a certain number of ergs^ we have therefore to multiply the latter by 10“h Similarly to obtain the number of kilogram- meters represented by a certain number of ergs, Ave have to multiply the latter by 10 - Watts = 10 ^ X ergs, Kilogram-meters = x ergs. From these two equations we find that Watts Ivilogram-meters = . The energy required to lift 75 kilograms one meter high in one second is a standard horse-power in the metric system. The acceleration of gravity may be taken as 9*81 meters per second, hence one horse-power is repre- sented by 75 X 9*81 . . . watts, or in round numbers : 736 watts correspond to one standard horse-power. In English measure the standard horse-power is equal to 32,500 foot pounds work done per minute. The usual English horse-power is equal to 33,000 foot pounds. Hence, to obtain the number of watts representing an English horse-powTr, Ave must multiply 736 with the ratio of 33,000 to 32,500. This gives the figure 746. Let Ea represent the counter-electro-motive force of the arma- ture in volts, and Ca the current in amperes, then the number of English horse-powers which could be obtained from it, if there were no losses, is H-P = -f- 12). /46 Retaining the notation of equations 5) and 6), we have also HORSE-POWER GIVEN OUT. 95 H-P = „ Z Nt n 10-® Ca 746 13 ), H-P = J^-2abmNtn IQ-'* Ca 746 14 ). The power which is actually obtainable is somewhat smaller, as certain losses occur. These might be classi- fied under two headings, mechanical friction and mag- netic friction. The former consists of the friction in the journals, of that between the commutator and the brushes, and of the resistance which the air offers to the rapid rotation of the armature, or the ^Svindage,” as it is techni- cally termed. The magnetic friction is of a somewhat complicated nature, and may manifest itself in various ways, but more especially in the heating of the armature core and of the pole pieces. If the armature core is not sufficiently subdivided, a fault very common in small motors, currents will be generated in it, which will be the stronger the more intense the field and the quicker the speed. It is as though the motor contained within itself a dynamo working on short circuit, and the power neces- sary for producing these currents must be supplied by the current flowing through the coils of the armature, and re- presents therefore so much power withdrawn from external use. Another source of loss is the limited number of the sections in the commutator. In establishing our formulas we have assumed that the aggregate of the currents in the different wires can be represented by a continuous semicircular sheet of current. This assumption is, strictly speaking, only correct if the number of wires and the corresponding number of sections is infinite. But when these numbers are limited, and especially when one sec- tion of the commutator corresponds to a wide coil, con- 96 ELECTRIC TRANSMISSION OF ENERGY, sisting of a great many turns of wire on the armature, then the change of contact between the brushes and suc- cessive commutator strips produces abrupt changes in the magnetizing effect of the current on the core of the armature, and our sheet of current, instead of being fixed in space as first assumed, undergoes violent oscillations, the amplitude of which is equal to the angular distance between two neighbouring coils. It is as though a magnet placed at right angles to the centre line through the pole pieces were kept in rapid oscillation, and since any m.agnet, if moved in the neighbourhood of metallic masses will heat the latter and absorb power, it follows that the pole pieces will become hot, and part of the energy produced by the motor will be wasted in this way. From what has just been explained, it will be evident that this loss can be reduced by increasing the number of sections in the commutator, and by subdividing the metal of -the pole pieces by planes at right angles to the axis of the armature. Another source of loss in some motors is the discon- tinuity of the armature core. This loss does not occur in Gramme armatures with smooth cylindrical cores ; but in armatures of the Pacinotti type, the projecting teeth, in sweeping closely by the polar surfaces, react on the latter, and produce eddy currents therein, which in their turn exert a retarding force upon the teeth. That this is really the case is shown in a striking manner in many dynamos having Pacinotti projections, notably in the Brush and Weston machines. Everyone who has examined these machines after some hours’ work, must have noticed that the pole pieces, especially where the coils and projections leave them, grow hot. At the entering side the heating is not so great, because there INTERNAL LOSSES. 97 the magnetizing effect of the armature current is to repel and weaken the lines, whereas at the leaving side it is to attract and strengthen them. If the machines be used as motors an opposite effect is produced, the pole pieces becoming hottest at the entering side. Cores with Paci- notti projections are very much in favour with the de- signers of motors, because it is thought that they increase the magnetic attraction which determines the force of the motor. On purely theoretical grounds this is so. It will be shown presently that the number of lines Z, pass- ing from the pole piece to the armature is the greater, the smaller the distance they have to leap through air, and by allowing the teeth to project so far as to almost touch the polar surfaces, the magnetic resistance of the air space can be very considerably reduced. But in prac- tice such perfection is unattainable on account of the heating and waste of power just explained. It is found necessary to make the clearance between the outer sur- face of the teeth and the inner surface of the pole pieces much greater than would suffice for free rotation, and it may be doubted whether the Pacinotti core is, after all, so great an improvement over the Gramme core as on purely theoretical grounds it seems to be. There is also another source of loss occurring even in armature cores which are perfectly subdivided and smooth on the outside. This is due to a molecular effect in the iron which has been termed hysteresis by Professor Ewing. In ordinary motors, having two or four field magnet poles, this loss is, however, very small and is generally neglected. In good motors the sum total of all the losses here enumerated at length amounts to only a small fraction of the total power. The ratio between that and the power actually obtainable on the shaft is called the efficiency of H 98 ELECTRIC TRANSMISSION OF ENERGY. conversion, and it should never be less than 90 per cent, in medium-sized and large motors. The electrical efficiency of the motor is the ratio of total internal electrical horse-power, as given by our for- mulas 13) and 14), to the external electrical horse-power applied at the terminals of the motor. Let represent the electro-motive force created in the armature coils. Ej, represent the electro-motive force appearing at the brushes. Ej represent the electro-motive force appearing at the terminals. r^ represent the total resistance of the armature. represent the total resistance of main coils on field magnets. r^ represent the total resistance of shunt coils on field magnets. (7, (7^, (7,, represent the external current, the current through the armature, through the shunt coils and main coils on field magnets respectively. Then for a compound- wound dynamo, in which the shunt coils are coupled across the brushes, the following equations evidently obtain: c = C^, C, = rs c.~ c^+ c .... 15), ^6 ~ '^a 16), E, = Ei — r„ C„, . . . . 17). The electrical efficiency is E, C ” E,Ca .... 18). F0R3IULAS FOR DYNAMOS AND MOTORS. 99 For an electro-motor, also compound-wound, the equa- tions are c ^ c C — ^ s C^=C„-C, 19), Ei = E, — r„ C„, 20 ), Ea = Ei — 21 ), E C ” = If the shunt coils are coupled to the terminals the for- mulas are for the dynamo," c= c„- a, c, = ^,c^= c^, 'f's 16), 17), and 18) remaining unaltered. For the motor we have c„= c - c,,c,= ^,c^= c^. 20), 21), and 22) remaining unaltered. The same formulas are applicable to the case of plain series or shunt machines, whether dynamos or motors, but in the case of series machines we insert = oo , and in the case of shunt machines we insert — o. CHAPTER IV. Types of Field Magnets — Types of Armatures — Exciting Power — Magnetic Circuit — Magnetic Resistance — Formulas for strength of Field — Single and Double Magnets — Difficulty in Small Dynamos— Characteristic Curves — Pre-Determination of Characteristics — Armature Reaction-^ Horse-pow'er Curves — Speed Characteristics — Application to Electric Tramcars. In the preceding chapter it has been shown how the electro-motive force of an armature can be found if the total number of lines passing through its core be known. It will now be necessary to determine the number ot lines, that is the strength of the magnetic field, from the constructive data of the machine. Before entering into a scientific investigation of the subject a cursory glance at the different types of field magnets adopted by the various makers of dynamos and motors, will be of interest. These are shown in Figs. 27 to 51. To make the classification comprehensive the type of armature is written beneath each field and the maker’s or designer’s name is written above it. We distinguish three types of armature. 1. The Drurriy wDund on the Hefner- Alteneck principle, as ex- lolained in Chapter II., and shown in Figs. 18 and 19 ; 2. The Cylinder^ w’ound on the Pacinotti or Gramme principle, also explained in Chapter II., and shown in Figs. 25 and 20 ; and 3. The Disc, wound on the Paci- notti or Gramme principle and only differing from the cylinder by the shape of the core*. It is a cylinder of TYPES OF FIELDS. 101 considerable diameter and small length, in fact a flat ring or disc. All the magnets employed in dynamos or motors are horse-shoes ; straight-bar magnets with poles at the ends being never used. The reason is obvious. We must in all cases bring opposite poles to the same armature, and that necessitates the employment of a bent magnet. It is necessary to distinguish between single, double, and multiple magnets. In the single horse-shoe magnet all the lines passing across the armature go through the magnet in the same direction. As an example we may take the Edison-Hopkinson dynamo. Fig. 27. The lines passing across the armature from N to continue all in the same direction, viz., vertically upwards from S to B, thence across the yoke from B to A, and finally vertically, downwards from A to N. A free unit pole would be urged along the closed magnetic circuit N S B A N, and there is no other way along Avhich it could travel. Now in a double horse-shoe, as represented for instance by the Weston machine. Fig. 41, there are two ways along which a unit pole might travel. One of these is JV S B A N, and the other N S B> C N, ov in other words, of the total number of lines passing across the armature, one half will go through the horse-shoe NABS, and the other half will go through the horse-shoe N C D S. We may con- sider the field magnets to consist of these two horse-shoes placed with like poles in contact to the left and right of the vertical center line. The arrangement of the Man- chester” dynamo is similar, but in this case the portions A B and C D, which in the Weston dynamo constitute the yokes, form the excited or active parts of the magnets and are surrounded by the magnetizing coils. The field magnets of the original Gramme dynamo (or motor) also 102 ELECTRIC TRANS3IISSI0N OF ENERGY. belong to the double horse-shoe pattern. But in this case a plane laid through the center lines of the cores of the magnets is parallel to and contains the center line of armature shaft, whereas in the AYeston type it is at right angles to it. Here, again, the lines are split up to the right and left of the vertical center line into two distinct circuits. Fig. 37 shows a similar arrangement, but with a single magnet. Figs. 39, 40, and 50 show single magnets, the plane of the horse-shoe being at right angles to the armature. Fig. 48 shows a quadruple horse-shoe magnet. Here the lines of force passing across the armature belong to four distinct circuits : S D A N, S I) C N, S B A and S B C N. The field magnets of the Mordey-Victoria machine shown in Figs. 46 and 47 consist of 8 complete horse-shoes, four on each side of the disc, and in some multipolar machines even a larger num- ber of magnetic circuits is sometimes employed. The machines which M. Marcel Deprez employed in his ex- periments (Fig. 51) had two cylinder aimiatures mounted on the same sj)indle a 5, and around them were placed eight horse-shoes, of which two, S B A N and S C D N, are shown in the illustration. It is not necessary to enter into a detailed description of all the types shown, as the diagrams are sufliciently clear. After what has been said above it will be evident that the proper function of the field magnets in a dynamo or motor is to produce lines of force which pass across the armature core. All other lines which miss the armature are useless and may even be detrimental to the working of the machine. The greater the number of useful lines the greater will be the electro-motive force generated at a given speed and with a given armature. Our aim should therefore be to produce a maximum number of SHORT CYLINDER. Fig. 43. Thomson-Houston. SPHERE. Fig. 51. Marcel-Deprez. A B TWO CYLINDERS. To face page 102. CVLINDEIl. Du Mkritkxs. CYLINDER. CYLINDER. CYI.INDKU. To focc jxtge 102. MAGNETIC RESISTANCE, 103 lines, and as a first step towards the realization of this object we must determine the relation between the number of lines and the constructive data of the machine. One of these data is the exciting power, that is the product of the number of turns of wire wound on the magnet, and the magnetizing current sent through the wire. It is customary to reckon the exciting power in Ampere- Turns ^ and it is shown by experiment and theory that the manner in which the product is made up is quite immaterial. We may have a large number of turns of fine wire and a small current, or we may have few turns of stout wire and a large current. The effect will always be the same if the product of amperes and turns be the same. Ex- periment also shows that for low degrees of magnetization, the electro-motive force produced in the armature is pro- portional, or nearly so, to the. exciting power X applied to the field magnets ; and since electro-motive force and strength of field Z are always proportional, we find that in these cases Z is proportional to X, We can represent this relation mathematically by introducing the concep- tion of magnetic resistance. According to this there is in every magnetic circuit a passive force opposing the crea- tion of lines, and the number of lines which are created is the quotient of the magnetizing force and this resistance. Calling the latter i?, we have ^ = f “)■ This formula is rigorously correct, provided we succeed in determining the magnetic resistance for every condition of magnetization. For low degrees of magnetization the resistance is nearly constant, and in these cases there exists simple proportionality between Z and X ; for higher 104 ELECTRIC TRANS3IISSI0N OF ENERGY. degrees of magnetization the resistance increases and the relation between Z and X becomes more complicated. A limit is ultimately approached beyond w^hich we cannot increase the strength of the field although we may increase the exciting power indefinitely. In this case the magnetic resistance has become infinite, and this condition is generally expressed by saying the magnet is saturated. The relations existing between magnetizing power and the magnetic moment have in the case of straight-bar magnets, spheres, and ellipsoids been investigated by Jacobi, Dub, Muller, and others, and a variety of formulas have been proposed to express these relations mathematically. Apart from the fact that these formulas in themselves are only rough approximations but imper- fectly fitting the results of experiments, they are for practical purposes almost useless, since the field magnets of dynamos and motors are not straight-bar magnets, but horse-shoes of every possible form and variety. In some cases these formulas are even misleading, and as an ex- ample we may cite the original Edison machines. According to the orthodox theory the magnetic moment of a cylindrical bar is proportional to some function of the exciting power, to the square root of the diameter of the bar and to the square root of the cube of its length. Hence to obtain a maximum of magnetic moment with a given weight of iron we must shape it into a long cylinder, and the original Edison machines were constructed on these lines. Experience has since then taught us that this was the worst possible form which could have been adopted, and the Edison machines built subsequently have stout and short magnets. The explanation for this apparent discrepancy between theory and practice is this, that in a dynamo or motor the magnetic moment of each MAGNETIC RESISTANCE. 105 bar composing the field magnet is of no account whatever, the electro-motive force depending only on the total number of lines produced, which is governed by laws totally different from those relating to the magnetic moment. It is very desirable that the relations between strength of field and exciting power should be mathe- matically established for those forms of magnets which are actually used in the construction of dynamos and motors. As yet no formula rigorously true for all degrees of magnetization has been found, and the difficulty is Fig. 52. principally due to the fact that the chemical composition and molecular properties of the iron play an important part which is not easily determinable beforehand. This is especially the case if the magnetization is pushed towards the saturation limit. For lower degrees of mag- netization the difficulties are still present, but they are of relatively less importance, and it is possible to establish formulas for the strength of the field which are sufficiently approximate for practical purposes. Let in Fig. 52 a series of wedge-shaped and very short magnets, ... be placed with polar faces of oppo- site sign in contact, so as to form a continuous ring inter- 106 ELECTRIC TRANSMISSION OF ENERGY. rupted only by the air space, A Lines of force will then pass across this air space, and an electro-motive force could be created by moving a conductor or series of conductors, so as to cut these lines. Let the polar sur- face of each elementary magnet be and let the density of magnetic matter, which we imagine to be distributed over the polar surfaces, be (t, then a- S is the strength of each polar surface. According to Ampere’s theory each elementary magnet can be replaced by an equivalent magnetic shell (page 28), consisting of a closed con- ductor in which a current flows, the product of current and area enclosed being numerically equal to the mag- netic moment of the elementary magnet. Imagine now the magnets replaced by a spiral of wire or soleyioid, then we can without appreciable error consider each turn of wire in the spiral as a current closed in itself, and if there be n such turns, and if the current be (7, the total magnetic moment will be in absolute measure n C S. Since with the exception of the two end faces A B^ A^ B^y the polar surfaces are in contact and cannot exert any action at a distance, the total magnetic moment of the series of elementary magnets is represented by the product of the magnetism on the end faces, and their distance, d. We have therefore the equation, a Sd=n C S. It has been shown (page 24), that the total number of lines emanating from unit pole is 4 tt. From a pole of the strength a- S there must emanate 4 tt a- S lines. Let Z be the total number of lines, or strength of field within the air space, then we find Z = 4 TT a S, and by inserting the value of a S from above equation. MAGNETIC RESISTANCE. 107 z = 4: 7T n C S d ^ which can also be written in the form ^ . IX C Z — 4 7T ^ d S The product n C h exciting power in absolute measure^ or ampere turns x 10~h S is the polar surface, and d the distance between the two poles. In deducing the formula for Z we have assumed the polar surfaces to be two parallel planes, but it can be proved that the same law holds good for surfaces of any shape, provided that their distance is very small as compared to their area. We can therefore apply the formula to the case of a cylindrical polar cavity partly filled by a cylindrical armature. Here we have two air spaces, and the polar surface S is the product of length of armature, b and the arc spanned by either pole, K Let J be the distance between the polar surface of the magnets and the ex- ternal surface of the armature core, and let X represent the exciting power joroducing Z lines, then the above for- mula becomes b -A- 4 7T b The strength of the field is represented by the quotient of exciting power, and an expression which is of the character length divided by area. The analogy with 108 ELECTRIC TRANSMISSION OF ENERGY. Ohm’s law will be obvious. The electrical resistance of a conductor is found by multiplying its specific elec- trical resistance with the length, and dividing by the area of the wire. In the same manner the magnetic resistance of the air space is found by multiplying ^ with the length (2 S'), and dividing by the area (a b) of the air space. We can therefore regard as the 4 TT specific magnetic resistance of air. The expression 24) gives the field in absolute lines ; to obtain it in such measure as to be directly applicable for the determina- tion of electro-motive force by equation 5) we must divide by 6,000. If for convenience we also use inches instead of centimeters in the dimensions S', and b and Ampere turns, instead of exciting power in absolute measure we find Z = A" 1880 2_J 24 « This formula is only correct under the supposition that there be no other resistance in the magnetic circuit but that of the interpolar air space, and that this space be really filled with air, and not with some other material. Mate- rials differ in regard to the resistance they offer to the passage of lines of force, or as it may also be expressed in regard to the degree to which they are permeable to mag- netic lines of force. Iron is more permeable than nickel or cobalt, and these metals are more permeable than copper, whilst copper again is more permeable than air. The magnetic permeabilitij of any substance can there- fore, be expressed by a coefficient /x, which denotes its STRENGTH OF FIELD, 101 > relation to the permeability of air, which is taken as 1. The equation 24) is valid for an interpolar space filled with air, but if this space be filled with a substance of permeability /tz, w^e have 2 = Vf A 7T h and, as in a dynamo machine, a portion of the interpolar space is filled by the copper forming the armature con- ductors, and as the permeability of copper is greater than 1, the magnetic resistance of this space will thereby become diminished. We must therefore expect that the figure in the denominator of formula 2Aa) will in reality be somewhat less than 1*880. From a large number of experiments made by the author with dynamos of various sizes and types, it was found that the resistance of the interpolar space filled by air and copper is more accu- rately represented by 1440 than bv 1880 The resistance of the interpolar space is, however, not the only passive force which opposes the creation of lines. There is also the resistance of the field magnet itself and that of the armature. After what has just been said with reference to the magnetic permeability of different mate- rials it is obvious that these resistances may be given by expressions of the character : length divided by area, the fraction being multiplied with a coefficient depending on the quality of the iron. Let in Fig. 53, L represent the average length of the magnetic circuit within the field magnet, and A B the cross-sectional area of the magnet core. Let, in the same manner, I represent the average 110 ELECTRIC TRANS31ISSI0N OF ENERGY. length of the magnetic circuit within the core of the armature, w’hich we suppose to be of the cylindrical Gramme type, and a h the area of the core. The magnetic resistance of the single horse-shoe magnet will then be proportional to ^iid that of the armature to each of these fractions being multiplied by a co- .efficient depending on the quality of the iron. From numerous experiments the author has found that for dynamos and motors of this type made of well annealed wrought iron, the strength of the field for low degrees of magnetization is represented by the expression. Z = U40 ^ +-■ - hb a b + 2 L AB 25 ). This is a purely empirical formula as far as the constants STRENGTH OF FIELD. Ill are concerned, but it is interesting to note that the same formula can be obtained by a theoretical investigation provided we assume a permeability o^ fx = 940 for the iron in field magnets and armature ; and that this is a fair average value when the magnetization is not very great. For cast-iron magnets the following empirical formula may be used : Z = 0*8 X 2 ^ I 1800 ^ + ^ + 7\b ah 3 Z A B 26). In the case of double horse-shoe magnets, as shown in Fig. 54, each horse-shoe contributes half the total number of lines, and we have for wrought iron. Z 2 X 2J , 2 1 , 2 Z 1440 — d 7 + X b a b A B . . . 27), and for cast-iron Z T 0*8 A" ;; 2^~~~2T 3 z ‘ 1800 -7 + -7 -h . „ h b ab A B . . . 28). The exciting power required to produce a certain strength of field can be found by multiplying the number of lines with the magnetic resistance, as represented by the de- nominator in the above four equations. It should be remembered that these only apply to cases where the intensity of magnetization is not too great, say up to ten lines per square inch. For more intense fields the mag- netic resistance of the iron part of the circuit increases considerably, and in practice it is found that to bring the field magnets of dynamos or motors near to saturation from 40 to 100 per cent, more exciting power must be 112 ELECTRIC TRANSMISSION OF ENERGY. applied than given by the above formulas. A method for the more exact determination of the exciting power for any degree of magnetization will be found on a sub- sequent page. A question of great practical importance is the rela- tive advantage of the single and double magnet. Since in the latter the area of air space is only half that of the former, the resistance of air space is doubled. On com- paring equations 25) and 27), it will be seen that the magnetic resistance of the magnet is equal in both Fig. 54. cases, and that the magnetic resistance of armature and air in the double horse-shoe is about double that of the single horse-shoe. On the other hand, only half the total number of lines have to pass through one magnet of the double horse-shoe type, and therefore the exciting power in the single and double magnet is about the same. But in the latter case this exciting power has to be applied on each of the two horse-shoes, whereas in the single-magnet machine it has to be applied on one horse- shoe only, but of double the sectional area. The length of wire required will therefore be as 2 : ^ 2, or by the SINGLE AND DOUBLE MAGNETS. 113 single magnet system a saving of about 25 per cent, of wire can be effected. On the other hand, the iron portion of the single magnet is somewhat heavier than that of an equivalent double magnet, and in cases where smallness of total weight is a consideration, as for instance in motors used for locomotive purposes, the double magnet, notwithstanding that it requires more wire, has a distinct advantage. An inspection of formulas 25) to 28) will show why, as was already mentioned in the beginning of Chapter III., small motors sometimes fail to act as dynamos. In these machines, or to speak more correctly, in these models of machines, the polar surfaces a b are very small as com- pared to the air space S', and consequently the magnetic resistance of air space is very high. The exciting j)ower is therefore also very high as compared to the strength of field, and it may happen that the electrical energy which is required to produce so large an exciting power is greater than the total electrical energy which can pos- sibly be produced by the armature. In this case the machine fails to act as a dynamo. Since the electro-motive force for a given speed of rotation is proportional to the strength of the field, and since that depends in its turn on the exciting j)ower, it follows that in every dynamo or motor there exists a definite relation between the electro-motive force created in the coils of the armature and the ampere turns applied to the field magnets. As already stated, this relation is of too complicated a nature to permit of its being repre- sented in the form of a mathematical expression rigo- rously applicable to all cases. Approximate formulas have been devised by Frohlich, Clausius and others, but these are not sufficiently accurate for practical use, and I 114 ELECTRIC TRANSMISSION OF ENERGY. have moreover the disadvantage common to all analytical methods of not appealing directly to our senses. In this respect graphic methods are preferable, and in the solu- tions of problems connected with dynamos and motors they offer facilities far beyond those of any analytical treatment. It is possible to represent all the important properties of a machine by curves, and since these pro- perties give it a distinct character by which it differs from others machines of similar type, these curves are called characteristics.^ The relations between current, electro-motive force, speed, horse-power, efficiency, and so forth, can all be represented graphically, but the curve most commonly used is that giving the electro- motive force as a function of the exciting power for a constant speed. In the case of a series dynamo the exciting power is proportional to the current, and if we plot the currents on the horizontal and the corresponding electro-motive forces created in the armature coils on the vertical, we obtain what is commonly known as the in- ternal characteristic. The external characteristic is a curve representing the electro-motive force at the ter- minals of the dynamo, and the distance between corre- sponding points on the two curves represents the loss of electro-motive force occasioned by the internal resistance of the machine. Before passing on to the use of graphic methods for the determination of the working conditions in dynamos and motors, it wdll be useful to show how the characteristic curves of a given machine can be found from its drawing, and conversely how the design of a machine can be pre- ^ This name appears to have been first used by M. Marcel Deprez, in 1881 , in an article in ‘‘La Lumiere Electrique,” although Dr. Hopkinson was the first to make use of graphic methods as applied to dynamos. PRE-DETERMINA TION OF CHAR A CTERISTICS. 115 pared in order that it may give any desired characteristic. The solution of this problem depends, broadly speaking, on our ability to pre-determine the characteristics of a dynamo by calculation.^ The ordinates of an ordinary characteristic as determined by experiment represent terminal pressure for constant speed and variable excita- tion. In order to eliminate the speed, Ave may deduce from this curve another — the characteristic of magnetiza- tion — in Avhich the abscissas represent as before exciting power, and the ordinates represent the total number of useful lines of force passing through the armature. It is this characteristic Avhich we pre-determine from the draw- ing of the machine. The flow of lines Z, as given by formulas 25 to 28, appears to be due to a kind of magneto- motive force (which is represented by the exciting power X) in a similar way as the flow of current is due to an electro-motive force. We have here Ohm’s law trans- lated into the magnetic circuit. There are, however, some important differences. In the first place, the flow of current takes place along well defined paths only, that is to say, through the conductors, and not through the sur- rounding space. The flow of magnetic lines of force, on the other hand, although greatest in the iron parts of the magnetic circuit, is by no means confined to the iron parts alone, but takes place also through the surrounding space. In other Avords, Ave cannot insulate for magnetism as we can insulate for electric currents. The other diffe- rence is that whereas the electrical resistance of a con- ductor is constant, and independent of the current (pro- vided the temperature remains unaltered), the magnetic ^ See the author’s paper on The Pre-Determination of the Characteristics of Dynamos.” (‘‘Journal of the Society of Telegraph Engineers and Electricians,” vol. xv., No. 64, p. 518. 1887.) 116 ELECTRIC TRANSMISSION OF ENERGY, resistance of iron increases with the flow of lines, the increase being small at first, but then more and more rapid, up to infinity, when the iron is saturated. Another difference is that whereas an electric current produces heat in the conductor, and dissipates energy, the continuous flow of magnetism through a piece of iron, does not pro- duce heat, and does not absorb energy ; and it is on account of this property of the magnetic circuit that some scientists have suggested to discard the term magnetic resistance,” since the overcoming of any resistance implies the expenditure of energy, which certainly does not take place in the maintenance of a magnetic field. The energy actually expended in the magnetisation of the field magnet of a dynamo or motor is due, not to any magnetic property of the machine, but simply to the electrical resistance of the magnetising coils. The term magnetic resistance ” has, however, been so generally adopted, and is so convenient, that an attempt to replace it by some more scientific term would lead to confusion, and in the following we shall use it in its conventional sense. To render the electrical representation of the flow of magnetic lines obvious, let us assume that the field magnet limbs in Fig. 53 are replaced by two batteries, the yoke by a wire, and the armature by a certain resis- tance, Rji, which is joined on either side with the ter- minals of the batteries by wires of high resistance. Ra 2 ^ representing the resistance of the interpolar space. If R be the resistance of the battery and connecting wire, and the apparatus be well insulated, the current through the armature will be E R jjr A" R “I” Rol c = PRE-DETERMINATION OF CHARACTERISTICS. 117 This does, however, not represent the conditions of the corresponding magnetic circuit, for in the latter the flow of lines takes place both through and around the iron parts. To correctly represent this condition we must imagine our electrical apparatus submerged in a badly conducting medium. Currents will then flow through this medium, and the current flowing through will be smaller than that given by the battery. It is impossible to say exactly in what manner these currents of leakage are distributed, but it is evident that the most serious loss must occur between the poles, where the difference of potential is a maximum, and that in a given medium and arrangement of battery the total loss through leakage can be considered as approximately pro- portional to the difference of potential between the poles. Let the latter have the value then the current lost, equals X^lp, if by p we represent the average resistance of the suiTounding medium for this particular configuration of battery. We can also, with fair approximation, put the current in the battery — ^2 = ^ 1 + f • Now imagine that by some means we can give to the electro-motive force ( of the battery any value between zero and that value which will produce the maximum current (zi) required. It will then be possible to deter- mine what electro-motive force is required for each value of 2'i, between zero and maximum. To do this we proceed as follows : We determine first the difference of potential at the poles — Xi — 2'j then the loss by leakage — 118 ELECTRIC TRANSMISSION OF ENERGY. X, p then the total current — ^2 = ^ then the loss of electro-motive force occasioned by the internal resistance of the battery — X.2^ — R 2 ' - 2'2 and, finally, the total E.M,F , — X = X, + ^2. The submerged battery is analogous to a horse-shoe magnet, since the latter is always surrounded by a medium which allows the passage of magnetic lines. To the electro-motive force of the battery corresponds the total exciting power applied to the magnet, whilst to the currents 2'j, and Z correspond numbers of lines respec- tively created within the magnet core, passing through the armature, and lost by leakage. Since for small degrees of magnetization the resistances of armature and magnet are not very different from their lowest initial values, which are very small in comparison with the resistance of leakage, it follows that for the early stages of magnetization leakage has very little influence on the exciting power ; in other words, that the creation of leakage or waste field does not materially increase the expenditure of energy required to produce the useful field. For a more intense magnetization, such as found in the usual working conditions of dynamos or motors, the case is, however, very different. The resistance of the armature has now considerably increased, and with it has increased at a still faster ratio A"j, that portion of the exciting power (or magnetic pressure at the PRE-DETERMINATION OF CHARACTERISTICS. 119 field-poles) which is necessary to force the lines through air space and armature. The immediate result is a large increase of the waste field, which^ has to be provided for by the magnet ; augmenting, therefore, the density of lines in the core of the latter considerably beyond the value which would otherwise correspond to the useful field. Hence large increase of magnetic resistance, and consequently also of exciting power. The determination of the exciting power for any given number of useful lines through the armature would now be possible if we knew first what is the resistance of leakage, and, secondly, what is the relation between the induction (lines per square inch of iron) and magnetic resistance for the particular quality of iron used in the machine. This relation can be experimentally deter- mined from samples of the iron, and the leakage can be approximately calculated from the drawing. The latter process is, however, somewhat uncertain, because, with the complicated forms of dynamos, the path of leakage lines cannot be easily defined, and much must be left to what may be called the mechanical instinct of the in- vestigator, according to which he is able to make certain assumptions. The experimental determination of the magnetic properties of the iron used, on the other hand, entails so much laboratory work that very few makers of machines could afford to undertake such investigations for every machine they build ; and the author has there- fore adopted an approximate method which, although not aiming at great scientific accuracy, is in fairly close accord with the results of actual practice. According to his method, the resistance of leakage is assumed to be in- versely proportional to the linear dimensions of the machine and the magnetic resistance of the iron is assumed 120 ELECTRIC TRANSMISSION OF ENERGY. to be a function of the saturation coefficient. Let 2 'i be the number of lines passing through the armature core, and let be the maximum number of lines which the core can accommodate when saturated (the corre- sponding exciting power or magnetic pressure being infinite), then the ratio z^jZ^ = ai is the saturation co- efficient. The magetic resistance, with an induction of would then be given by the product of the initial mag- netic resistance ( formula 25 on page 109) and the term tang (~tr,) 7T 2 which has received the name of tangent function.” As- suming, then, that the tangent function represents cor- rectly the increase of magnetic resistance due to an increase of induction, we have all the data necessary for the pre-determination of the characteristic of magnetiza- tion, and from this for the pre-determination of all the other characteristics of a machine. From numerous ex- periments made by the author with his own machines, and from data supplied him by other dynamo makers, it appears that the iron now obtainable for the construction of dynamos and motors is of a very high quality and saturation may be considered to take place in armature cores when the density of lines has reached 26, and in magnet cores when it has reached 18. The following table has been prepared by Mr. A. T. Snell on the basis of these figures : — PRE-DETERMINATION OF CHARACTERISTICS. 121 Lines per square inch. Tangent Function. Armatures. Fields. 1 1-00 1-005 2 1-01 1-013 3 1-01 1-021 4 1-02 1-041 5 1-03 1-069 6 1-04 1-104 7 1-04 1-147 8 1*08 1-202 9 1-11 1-273 10 M4 1-362 11 1-17 1-489 12 1-22 1-654 13 1-26 1-891 14 1-29 2-250 15 1-40 2-853 16 1-49 4-624 17 1-57 7-707 18 1-79 infinite 19 1-96 20 2-18 21 2-52 22 3-05 23 3-92 24 5-61 25 10-82 26 infinite To ascertain the exciting power required to produce a certain flow of lines in the armature of a given dynamo, we proceed now as follows : We determine from the draw- ing the cross-sectional area of iron in the armature a h and magnets A B, The induction when saturated is re- spectively 26 a by and 18 ^ We assume now a cer- tain useful induction z^y and determine the corresponding 122 ELECTRIC TRANSMISSION OF ENERGY, density, zj {a b). The value of the tangent function is then taken from the table and multiplied Avith the initial resistance of the armature core j- to this the the sum with Zi. This gives that portion of the exciting power which is required to force the useful number of lines Zi through armature core and interpolar space (^i). The corresponding magnetic pressure creates a certain lea^kage, AA^hich is given by the expressions From numerous experiments the author has found that for single upright horse-shoe machines of the type shown in Fig. 40, the resistance of leakage can be represented by the formula whilst for single horse-shoe inA^erted machines (Fig. 39) where the poles are in proximity to the iron bed-plate, the leakage resistance is someAAdiat smaller, and may be represented by the formula The floAv of lines through the field magnet may approxi- mately be taken as the sum of useful and Avaste lines. We determine noAv the density of lines in the field zj A and find the corresponding multiplier from the 680 460 Id z.^ = Zi + PRE-DETERMINATION OF CHARACTERISTICS, 123 multiplied with this figure, which gives the actual re- sistance corresponding to this particular induction. The product multiplied with gives that part of the total exciting power which is necessary to force lines through the magnet. The total exciting power is X = -h ^2- The same calculation performed for a variety of values of and the results plotted, gives the curve known as the characteristic of magnetization. By way of example is here added the calculation for a single horse-shoe upright machine, having the following dimensions : Armature core, 13 in. diam., 14 in. long, 2\ in. deep. Field-magnet, 14 in. bore, 7 in. by 13 in. cross-section. p - 50*3 ; Rcc = 6*27 = *54 R^= 1*33. The density of lines through armature and magnet is denoted by m and M respectively, and A, and are re- spectively the tangent functions for armature and magnet. 124 ELECTRIC TRANSMISSION OF ENERGY, C ^2 j m \M ! Ra X Rf X R(A + aj X 400 7*2 1-04 -56 6-83 2,732 1 — — 54 454 5 1-069 1-42 — 644 1 3,376 550 — — 10 1*14 •61 6-88 3,784 — — 75 625 6-8 1-13 1-5 — 936 4,720 825 — — 15 1-4 -75 7-02 5,800 i — — 115 940 10-3 1-39 1-84 — 1,730 7,530 1,000 — 18*2 1-8 •97 7-24 7,240 1 — 142 1,142 12-5 1-75 2-32 — 2,644 9,884 1,100 — 20 2-18 1-18 7-45 8,195 — — 162 1,262 13-85 2*2 2-92 — 3,700 11,900 1,155 — — 21 2-52 1-36 7-63 8',812 — — 175 1,330 14-6 2-54 3-37 — 4,483 13,295 1,210 — 22 3-05 1-647 7-92 9,600 — — 190 1,400 15-4 3-2 4-25 — 5,950 15,550 1,265 — 23 3-92 2-11 8-38 10,600 — — 200 1,465 16-1 4-2 5-88 — 8,614 19,214 1,300 23-5 4*3 2-32 8-59 11,167 — — 220 1,520 16-7 6 7*98 — 12,130 23,297 1,310 — 1 23-8 4-9 2-75 9-02 11,816 — — 222 1,532 16-8 6-6 8-8 — 13,480 25,296 1,320 — ’ 24 5*61 3-03 9-3 12,270 — — 245 1 1,565 17-2 10 1 13-3 — 20,814 33,084 To show the influence of leakage, the calculation is here also given for a machine of the same dimensions but of the inverted type. In this case p — 34, all the other constructive data being the same. Fig. 55 represents the characteristics of these two machines as plotted from these figures. The characteristic curves of magnetisation give, as already stated, the total flow of lines of force through the armature, provided no other exciting power except that applied to the field magnets is active. This condition is fulfilled if the field magnets are separately excited, and no current is allowed to flow through the armature, the electro-motive force in which is in this case exactly the PRE-DETERMINATION OF CHARACTERISTICS. 125 Fig. 55. C 1 1 \M 1 j Ba X Bf X B(a + X 400 — 1 7*2 1-04 •56 6-83 2,732 — 80 480 5-3 1*075 1-43 — 686 3,418 550 — — 10 1 -14 •61 6-88 3,784 — Ill 661 1 7-28 1*16 1-54 — 1,020 4,804 825 — — 15 1-4 •75 7-02 5,800 — 171 996 10-9 1*48 1-97 — 1,960 7,760 1,000 — — 18*2 1*8 •97 7-24 7,240 — 212 1,212 ; 13-3 2 2-66 — 3,220 10,460 1,100 — — ! 20 2-18 1-18 7-45 8,195 — 241 1,341 1 14-7 2*7 3-6 — 4,827 13,022 1,155 — — 21 2-52 1-36 7*63 8,812 — 260 1,415 15*5 3-34 4-44 — 6,282 15,094 1,210 — — * 22 3-05 1 -647 7-92 9,600 — 282 1,492 16-39 4-9 6-5 — 9,700 19,300 1,245 — 22-6 3-5 1*9 8-17 10,150 — 298 1,543 ' 16-9 7 9*31 — 14,400 24,550 1,260 — i 22-9 3-8 2-05 8-32 10,480 308 1,568 17-23 10 13-3 — 20,860 31,340 1 same as that which can be measured at the brushes. When a current is allowed to flow through the armature, the electro-motive force which we measure at the brushes is 126 ELECTRIC TRANS3IISSI0N OF ENERGY, not exactly the same as that generated in the armature, but either smaller or larger, accordingly as the machine is used as a dynamo or motor. We have thus to distin- guish between three conditions of working, namely (1) no current, (2) dynamo current, and (3) motor current passing through the armature. The first condition can be obtained by simply opening the external circuit, but we can also imagine that the external circuit remains closed, and that some source of electro-motive force is inserted, acting in opposition to the electro-motive force gene- rated in the armature, and that this is so accurately adjusted as to prevent a fiow of current in either direc- tion. We shall then have, so to speak, a static balance between the armature and opposing electro-motive force. And for this reason the author has suggested the name static characteristic,” to any characteristic curve repre- senting this condition of working. If we now imagine the opposing electro-motive force reduced, it Avill no more be able to balance the electro-motive force of the armature, but will be overpowered by the latter, with the result that a current will flow, doing work upon the opposing electro-motive force. The machine now works as a dynamo, absorbing mechanical and giving out electric energy. Any characteristic curve expressing this condition of working, we call dynamic characteristic.” If, on the other hand, we increase the opposing electro- motive force until the latter overpowers the armature electro-motive force, a current will be forced through the armature doing work in overcoming its electro-motive force, in other words driving the machine as a motor. Any characteristic expressing this condition of working of a machine we call motor characteristic.” For certain reasons which will be given presently, the dynamic cha- STATIC, DYNAMIC AND MOTOR CHARACTERISTICS, 127 racteristic must always be lower than the static cha- racteristic, and the same holds good generally for the motor characteristic, but there are cases when the motor curve lies above the static curve. Confining ourselves for the present to dynamos only, it is easy to see why the dynamic curve must be below the static curve. Allowance must in the first place be made for the drop in electro-motive force due to the electrical resistance of the armature. This drop can of course be calculated by multiplying current and resistance. But beyond this, there is a further reduction of electro-motive force, which can be explained as follows. Let us for the moment assume that we work a two-pole dynamo with the brushes set exactly at right angles to the polar diameter, and let us concentrate our attention on say the positive brush, that is the brush where the current leaves the armature. In all the coils on one side of this brush the current flows in one direction, say towards the com- mutator on the outside of the armature, whilst in all the coils on the other side of the brush it flows in the oppo- site direction. There is thus a reversal of current in each coil as it passes under the brush. Whilst under the brush, the coil is short circuited on itself, but as a moment before it was traversed by half the total arma- ture current, there will remain some current flowing during its period of short-circuiting by virtue of the self- induction of the coil. This current becomes gradually weaker, owing to the resistance of the coil, but even if this cause were suflScient to bring the current to zero, it can obviously not reverse it. By the time the coil emerges from under the brush, the previous current in it may or may not yet have died out, but at that instant half the total armature current is forced through it in the 128 ELECTRIC TRANSMISSION OF ENERGY. opposite direction. This causes a violent spark, which can only be avoided by shifting the brush into a more ad- vanced position. When in that position, the coil whilst under the brush cuts through lines of force which induce in it an electro-motive force opposed to the persisting current, and bring it thus quickly to zero. But, these lines of force do more ; they start the current in the opposite direction, so that the coil at the moment of emerging from under the brush carries already the same current as will be forced through it afterwards, whereby the transition from the idle, short-circuited position under the brush to the working position beyond the brush, takes place quite gradually and without sparking. By shifting the brush forward we have killed out the spark, but we have sacrificed some of the lines of force which might otherwise have increased the electro-motive force of the armature, and which, whilst the machine was working on open circuit, have indeed been so utilized. There are therefore more lines of force utilized whilst the machine is working statically, than whilst it is work- ing dynamically, from which it follows that the dynamic electro-motive force must be lower than the static electro- motive force. A further reduction of electro-motive force is due to the fact, that the armature itself becomes magnetised by the current flowing through its coils, and reacts on the field magnets. If it were possible to work with the brushes on the neutral diameter, the poles deve- loped in the armature would stand exactly midway be- tween the main field poles, and would neither strengthen nor weaken them, but the brushes must for the reason already given be shifted forward, which brings armature and field poles of the same sign nearer together, and the result is, that the main field is somewhat weakened. ARMATURE REACTION. 129 which again reduces the electro-motive force. In reality, the phenomena here briefly sketched, and which ^ are technically comprised under the term armature reac- tion,” are not quite as simple as here stated, but it would exceed the scope of the present work to enter more fully into details, the more so as the total effect of armature reaction is in good dynamos very small, amounting often to less than five per cent, of the total electro-motive Fig. 56. Volts. INTERNAL CHARACTERISTIC OF A GRAMME DYNAMO. force. In badly designed machines it may become con- siderable, as can be seen from Fig. 56, which represents the internal characteristic of an A gramme dynamo, tested by M. Marcel Deprez. This behaviour of the dynamo can best be studied with separately excited machines, and Mr. Esson has made very careful trials on the subject, which were published in April, 1884, in The Electrical Review.” The dy- namo experimented upon was a Phoenix ” machine with K 130 ELECTRIC TRANSMISSION OF ENERGY, Pacinotti armature. It was separately excited and kept running at a constant speed of 1,600 revolutions a minute, whilst the current which was permitted to flow through the armature was varied by means of a rheostat. The line E, Fig. 57, represents the internal electro-motive force corresponding to the constant exciting power if there were no reactions. The line Eb represents the electro- motive force which would be found at the brushes if Fig. 57. EXPERIMENT WITH PHCENIX DYNAMO. there were no reaction, and the line EV was that actually observed. The difference of the ordinates of Eh and Eb^ represents the loss of electro-motive force due to self- induction, weakening and distorting of the fleld. The armature reaction in a motor is very similar to that in a dynamo, except that the electro-motive force lost in resistance must be added to, instead of sub- stracted from the internal armature electro-motive force. MO TO R CHA RA CTERIS TIC OF IMPERFECT MA CHINE, 131 If the machine is properly constructed, there will there- fore be very little difference between its dynamic and motor characteristics, both lying below the static cha- racteristic, but if the machine is not so constructed as to give a high efficiency, then it may happen that its motor curve is considerably higher than its dynamic, and even higher than its static curve. The reason is obvious. If a sensible amount of energy is wasted in eddy currents or hysteresis, this energy must be supplied by the motor current in the shape of an increased terminal pressure. Now we find the motor characteristic of electro-motive force by deducting from the electro-motive force at the brushes which can be measured, the calculated electro- motive force necessary to overcome the resistance of the armature. The latter being constant, it follows that the higher the brush electro-motive force, the higher must also be the motor electro-motive force. The author has verified this conclusion by testing a Biirgin machine as a dynamo and as a motor. As was to be expected, the dynamic curve was found to lie below the static curve, but the motor curve, instead of being below the static curve, was found to lie above it. The reason is probably that eddy currents in the corners of the iron wire hexa- gons which form the armature core, and a certain amount of surging of lines over the surface of the pole pieces pro- duced by these corners, absorb a sensible amount of energy requiring an increased electro-motive force in the driving current, and thus making it appear as if the counter electro-motive force in the armature were higher than it really is. The fact that some machines show a high counter electro-motive force, has led certain scientists, and notably Professors Ayrton and Perry, to formulate a theory of electro-motors, according to which the arma- 132 ELECTRIC TRANSMISSION OF ENERGY, ture was credited with some sort of power to increase the strength of the field instead of weakening it, as is actually the case, and it was recommended that motors should have small field-magnets and large armatures. Practical experience has, however, disproved this theory, and the best motors are nowadays designed by the use of the same formulae as the best dynamos. The effect of the current in the armature on the useful field is generally comprised under the term armature reaction,” and has been investigated by the Drs. Hop- kinson, and by Messrs. Esson and Swinburne, who have, however, not given any simple rules by which the arma- ture reaction can be determined in a manner sufficiently easy for every day use. Some engineers content them- selves in making a percentage allowance for the disturb- ing effect of the armature current, varying from four to eight per cent, of the static electro-motive force for ring, and from two to four per cent, for drum armatures. As far as the author is aware, the only attempt for establishing a simple rule for the determination of the armature reac- tion, is due to Herr von Dobrowolsky, the engineer to the Allgemeine Elektrizitaets-Gesellschaft, of Berlin, who proceeds in the following way. Let in Fig. 58, O F represent the ampere turns on the field magnets, and O Ai the ampere turns on the armature drawn to any convenient scale. In a two-pole drum armature contain- ing JVt external conductors and carrying a current Cj, the length of the line O Ai would represent the expression Nt C, 2 2 ’ whilst in a ring armature it would represent the C expression Nt If we now complete the rectangle O Ai F, and draw the diagonal O By, the ratio be- dobrowlsky's triangle. 133 tween the static and dynamic electro-motive force is given by the ratio of length of the two lines O and O F respectively. If similarly the length O repre- sents the ampere turns with the armature current the ratio between the lines O and O F gives the weaken- ing effect on the useful field produced by this armature current. By using the Dobrowolsky triangle, it is thus possible to determine beforehand what field excitation Fig. 58. will be required with various armature currents. The inventor of this method states that the error in pre- determining thus the dynamic or motor characteristic from the static characteristic does not exceed one per cent., and is generally within one-half per cent, of the values subsequently found by experiment, but this refers to machines of the particular type designed by him. An important feature in these machines is the use of a cylindrical iron bush surrounding the armature, and 134 ELECTRIC TRANSMISSION OF ENERGY. joining the poles of the field magnets. There is thus no gap between two successive field poles, but iron all the way round. By this arrangement Dobrowolsky obtains a very gradual shading off in the field lines and a spark- less collection, but at the expense of an increased arma- ture reaction, since the neighbourhood of iron increases the self-induction of the coils being commutated. In machines having no pole-bush the armature reaction is naturally smaller, and the author has found that in these cases the Dobrowolsky triangle gives the dynamic cha- racteristic slightly too low. The error can be rectified by taking -5 to *7 of the armature ampere turns as the basis of the graphic construction. We may now proceed to show the use and interpreta- tion of characteristic curves generally. Fig. 59 shows the internal and external characteristics of a series-wound Siemens dynamo, as given by Dr. Hop- kinson in the Proceedings of the Institution of Me- chanical Engineers, 1879. The dotted curve O repre- sents the electro-motive force at the terminals of the machine, and the curve shown in a full line O that in the armature. The latter is obtained from the former by adding to its ordinates the internal loss of electro-motive force. This is the product of current and internal re- sistance, which latter was in that particular machine 0*6 ohms. Thus at 50 amperes the loss is 30 volts, and it will be seQn from the diagram that the difference between the two ordinates corresponding to 50 on the abscissae is 30. We can also represent the loss of electro-motive force by a characteristic, and since it is always propor- tional to the current, the characteristic in this instance becomes a straight line, O r. The geometrical tangent of the angle which this line forms with the horizontal is HORSE POWER CURVES. 135 evidently equal to the internal resistance of the machine. The ordinates enclosed between O r and O E„ represent the external electro-motive forces, and therefore the in- ternal characteristic, O becomes the external cha- Fig. 59. Volts racteristic if we take O r for the base line instead of the horizontal. By a very ingenious method due to Professor Silvanus P. Thompson these characteristics can also be used to show at a glance the horse-power which corresponds to 136 ELECTRIC TRANSMISSION OF ENERGY, any particular current or electro-motive force. As already shown the horse-power represented by a current c flowing c E . under an electro-motive force E^ is H-P = One horse-power can be represented by an infinite variety of c and E^i but these values must all satisfy the equation 746 = c E,, A curve representing one-horse power will pass through all such points of which the product of their ordinates is a constant, viz., 746. Similarly a curve representing the value of two horse-power will pass through points of which the product of their ordinates equals 1492, and so on. In other words, all the horse-power curves are rect- angular hyperbolas,^ and by drawing a set of these curves across our diagram — as shown in dotted lines — we can determine at a glance what is the horse-power correspond- ing to any point on the characteristic. Thus a current of 30 amperes represents about 3*35 H-P of internal elec- trical energy, and about 2*7 H-P of electrical output or energy delivered into the external circuit. A current of 50 amperes represents a little over 6 H-P internal, and a little over 4 H-P external energy, and so on. In a dynamo the internal characteristic lies always above the external characteristic. In a motor, however, their position is reversed, since the external electro-motive force must necessarily be greater than the counter-electro- motive force developed in the armature coils. Fig. 60 shows the characteristics of the Siemens dynamo men- tioned above if used as motor. 'Not to get the diagram too long the speed has been reduced to 500 Revolutions. The curve O E^ represents the counter electro-motive ^ The scales for volts and amperes being equal. HORSE POWER CURVES. 137 force developed in the armature coils, and the curve O Fig. 60 . which is shown in a dotted line, represents the terminal electro-motive force. The difference between the ordinates 138 ELECTRIC TRANSMISSION OF ENERGY. of the two curves represents the electro-motive force necessary to overcome the internal resistance of the machine. By drawing the straight line O r under an angle, the tangent of which is numerically equal to the internal resistance, but this time below the horizontal and not above it as in the former example, we can regard it as the new base line, and then the curve O becomes the external characteristic. In diagram (Fig. 60) it is assumed that by some means we keep the speed constantly at 500 revs, a minute. Easy as it is to fulfil such a condition in a dynamo, it presents considerable difficulties if we have to deal with a series- wound motor, because its speed depends on a number of factors which to a certain extent may vary independently of each other. The speed depends on the current and electro-motive force supplied to the motor, and on the amount of mechanical work it has to do. In some cases the work itself, that is the product of turning moment and speed, depends on the latter, and thus it will be seen that the relation between these various quantities is of a rather complex nature. It is however easy to represent these relations graphically by the use of speed charac- teristics, which were first published by the author in The Electrician,” of December 29th, 1883. Assume the case that the external electro-motive force is a fixed and con- stant quantity. What will be the relation between speed, power, and efficiency of, say, a series-wound motor? Since E^ is constant at all currents, we have practically an unlimited supply of current such as would be ob- tained from the mains in a system of town supi^ly. The current passing through the motor will depend on its resistance, and on its counter-electro-motive force. The former is constant, whilst the latter increases with the SPEED CHARACTERISTICS, 139 speed. The faster we allow the motor to run the less current will flow through it, and the less power will be absorbed by it. Let in Fig. 61 the speeds be plotted as abscissas, and the electrical horse-power absorbed as ordi- nates, then with a series-wound motor we obtain the curve W W, The exact shape of this curve depends, of course, on the construction of the motor, but its general character will be as shown. The easiest way of flnding the curve experimentally is by attaching a brake to the motor, and loading it with different weights so as to pro- Fig. 61 . SPEED-CHARACTERISTICS OF SERIES MOTOR. duce different speeds. The horse-power absorbed by the brake can at the same time be plotted in the curve w w. If we begin with an excess of load on the brake, which will hold the motor fast, a maximum of current will flow, and a maximum of electrical energy will be absorbed without producing any external work. On the other hand, if we remove the brake altogether the motor will attain a maximum velocity o m, and again no external work will be produced, but in this case very little current will pass, and the electrical energy absorbed will be a minimum. Between these extreme limits of no speed and 140 ELECTRIC TRANSMISSION OF ENERGY. maximum speed external work will be produced, and there is one particular speed, o at which this work will be a maximum. The ratio of the ordinates of W and w can be plotted in a curve, r\ r\, drawn to any convenient scale, and this gives the commercial efficiency of the motor as a function of the speed. There is one particular speed, 0 b, at which the efficiency is a maximum, but this is not necessarily the same speed as that for which the work is a maximum. As a rule it is considerably greater, and in actual work the motor should be so geared that it runs at or about the speed of maximum efficiency. The experimental determination of the most economical speed, as just described, requires the employment of a dynamometer or brake, and if such an apparatus be not at hand, cannot be adopted. In this case a different method can be used, which is fairly reliable, although not quite so accurate as the actual power test. The question to be solved is the relation between speed and current in a given series-wound motor supplied with current at a constant electro-motive force. This question can be solved if we know the internal resistance of the motor and its internal characteristic. Having obtained the relation between speed and current, we can construct the diagram Fig. 61, making a certain allowance for the effi- ciency of conversion. We assume that the motor derives its supply of current from a pair of mains between which a potential difference of 100 volts is maintained. Let, in Fig. 62, O represent internal characteristic for a con- stant speed of say 500 revs., and let the inclined straight line, r, be drawn across the diagram at such an angle with the horizontal that its geometrical tangent is numerically equal to the internal resistance of the motor in ohms, then the ordinates of the line, r, represent the counter-electro- SPEED CHARACTERISTICS. 141 motive forces which must be created in the coils of the armature so that any given current may pass. Thus, at 100 amperes, the counter-electro-motive force must be 40 volts. If the armature revolves at a speed of 500 revo- lutions a minute, we see from the characteristic that its Fig. 62 . Volts 1000 Revs. RBLATTON BETWEEN SPEED AND CURRENT IN SERIES-WOUND MOTOR. counter-electro-motive force is 68 volts, and to bring the latter down to 40 volts, so that a current of 100 amperes may pass, the speed will have to be reduced in the pro- portion of 68 to 40. The speed corresponding to a cur- 40 rent of 100 amperes is therefore 500. — = 294 revolu- 142 ELECTRIC TRANSMISSION OF ENERGY, tions. Similar calculations can be made for other values of current, and the speeds obtained can be plotted in a curve shown in Fig. 62, below the horizontal. At 166 amperes the speed is zero, because the whole of the con- stant electro-motive-force available of 100 volts is re- quired to overcome the internal resistance of the motor, leaving nothing to be opposed by counter electro-motive- force. At 16 amperes the speed is 1000 revolutions, and at smaller currents the speed might be still greater. Theoretically, it should be infinite if no current passes, and this would be the case if the motor were free to revolve without doing any work, and if there were no internal mechanical losses. This, of course, is an im- possible condition, and a limit is set to the speed by the work which must be done to overcome mechanical and magnetic friction. In good motors this is, however, com- paratively small, and consequently the speed of the motor, when running empty, is inconveniently high. This is a great drawback in many cases, especially where motors are required to drive lathes and other machinery offering a variable resistance. The example represented in Fig. 62, applies also to the case where a series-wound motor is worked from a set of secondary cells, having a very low internal resistance, as the electro-motive force is then approximately constant at all currents. To lessen the difference in speed it is usual to insert a rheostat or vari- able resistance into the circuit between the cells and the motor. A maximum of resistance is inserted when the motor is running empty, and as the load increases re- sistance is switched out so as to regulate the speed. At best this is a clumsy device, requiring personal attention, and not very efficient, as with it variations in speed can never be altogether avoided. It is also wasteful, the heat REGULATING THE SPEED. 143 develojDed in the artificial resistance being so much power lost. A better plan is to wind the field magnets of the motor on the compound principle, both main and shunt coils magnetizing in the same direction. This will raise the early part of the characteristic as shown in dotted lines, and will reduce the speed as shown also in a dotted line. This method is not a complete cure for the evil, but it is a palliation of it which in practice proves very successful. To make the motor perfectly self-regu- lating, it would be necessary to let the main coils on the field magnet excite the latter in an opposite sense to the shunt coils ; but then a very valuable quality of the series motor, viz., its great starting power, would be lost. If a motor is employed for railway or tramway work it is very important that there should be an excess of power at starting. This condition is admirably fulfilled by the or- dinary series-wound motor, since the current, the strength of the field, and the statical effort or torque are all maxima when the motor is at rest and decrease as it gathers speed. There is thus an automatic adjustment between speed, power, and resistance. Take, as an example, an electric tramcar worked by accumulators. On a heavy gradient or bad part of the road, the speed is low, allowing a large current to pass through the motor, thus providing the extra amount of tractive force necessary ; on a good level road the speed will increase, less current will pass through the motor, and less tractive force will be developed. But on a downward incline, when no tractive force at all is necessary, the motor, and with it the car, would acquire too high a speed if not checked in some way. This was one of the difficulties encountered in the early forms of Mr. Reckenzaun’s electric tramcar, worked by accumu- lators. Each car w^as provided with two series-wound 144 ELECTRIC TRANSMISSION OF ENERGY. Reckenzaun motors^ gearing by means of a worm and wheel directly with the axles of the car. On a very good level road, and on downward gradients, it was neces- sary to continually handle the brake in order to prevent the motors running too fast. This defect has been removed in the later forms of electric tramcars ; the motors are wound on the compound principle, and thus a certain initial strength of field is maintained whereby the speed of running light is reduced to a safe limit. CHAPTEK V. Graphic Treatment of Problems — Maximum External Energy — Maximum Theoretical Efficiency — Determination of best Speed for Maximum Com- mercial Efficiency — Variation of Speed in Shunt Motors — The Compound Machine as Generator — System of Transmission at Constant Speed — Practical Difficulty. The treatment of problems relating to the electrical transmission of energy is greatly simplified by the use of the curves explained in the preceding chapter, and by other graphic methods, of which we may mention that due to Professor Silvanus Thompson. The problem is as follows. Let a square A B (7 Z> be drawn so that the length of one of the sides shall represent the electro- motive force E of the supply to any convenient scale. Fig. 63, and let the counter-electro-motive force e of the motor be represented by the length A F A G, Draw through F and G the lines F K and G H respectively parallel to A B and A C. The energy supplied to the motor equals the product of electro-motive force F and current C, whilst the work converted into mechanical energy in the armature of the motor equals the product of counter-electro-motive force e and current C, Let B repre- sent the total resistance in the circuit, then C = which in our diagram is represented by the length F C divided by R, The energy delivered to the motor is evidently 146 ELECTRIC TRANSMISSION OF ENERGY. E{E — e) R ^ and that converted in the motor is e {E — e) Now the area of the rectangle F K D C — E {E — e) and the area of the rectangle G B K Ij — e {E — e) \ and since 72 is a constant, we find that these areas — shaded in our diagram — are proportional to the work expended and recovered. Thompson’s diagram can immediately be used to solve Fig. 63. graphically two of the problems which have already been treated analytically in the first chapter (page 39). These are the following: First, what is the condition of maxi- mum work obtained from the motor ? and, secondly, what is the condition of maximum efficiency ? The answer to the first question is easily found by inspecting our diagram. Fig. 63. Since the rectangle G B K L, which represents the work of the motor, is inscribed between the diagonal A D and the sides A B, D B ; the question resolves into that of finding which of all possible rectangles inscribed within these lines has a maximum of area. This is evidently a square, the sides MAXIMUM COMMERCIAL EFFICIENCY, 147 of which are half as long as those of the external square. In this case the work expended is represented by a rectangle of half the area of the external square, and the efficiency is therefore 50 per cent. 1 We have : Work expended ^ ^ . „ Work recovered ^ . „ Efficiency = 0*50. As regards the second question it will readily be seen that the discrepancy in the area of the two rectangles, Fig. 63, is the greater, the nearer the point L is to or in other words, the smaller the counter-electro-motive force. In the same measure as the latter increases, point L is pushed further towards Z), and the areas of the two rect- angles become more and more equal. The efficiency, therefore, tends towards unity as the counter-electro- motive force of the motor tends towards the electro-motive- force of the source of supply of electricity. This state- ment has already been made in the first chapter, and it is theoretically quite accurate ; but from a practical point of view it requires some qualification. It will be seen that when the counter-electro-rnotive force of the motor ap- proaches very closely the electro-motive force of the supply, the current becomes very small, and the work expended and converted becomes also very small. Now the work converted in the motor is not all available in the shape of external mechanical energy, and it may well happen that in this case, after the resistance of mechanical and magnetic friction has been overcome, no margin remains for useful external work. The com- mercial efficiency would therefore be Zero, although the theoretical efficiency is a maximum. To put the matter 148 ELECTRIC TRANSMISSION OF ENERGY. in another way : a certain minimum of current is required to overcome the friction of the motor, quite apart from any external resistance. It has been shown that with a constant field the torque of the motor depends only on the current which passes through the armature, and is independent of the speed. We may apply this law with sufficient approximation to the present case and assume that at all speeds the current which is required to over- come the internal friction of the motor is constant. Let y represent this minimum of current, which will just keep E — e the motor alone going, then — ^ 7 ] is the current doing useful external work, and the commercial efficiency is E-e 7) = e It ~E E-e R __ Ee — — e Ry ” “ - Ee To find the condition under which >7 becomes a maximum we put = 0 and obtain a e {E -ey = E Ry 29). This formula is cai)able of graphic representation. Let in Fig. 64 O A represent the current 7 , which is required to keep the motor revolving at or near its normal speed when no external work is being done, and let O jfi?represent the electro-motive force E of the source, which we suppose to be constant for all conditions. This would be practically the case if the source of current were a self-regulating dynamo, or a set of secondary batteries having a very low internal resistance. The area of the rectangle O AG H represents the number of watts required to overcome the MAXIMUM COMMERCIAL EFFICIENCY, 149 friction of the motor at its normal speed when doing no external work, and if the motor be shunt-wound, or com- pound-wound for constant speed, its strength of field will not greatly vary when external work is being done, and we may with a reasonable degree of approximation re- gard the area of the rectangle O A G H to represent the internal loss of energy in the motor under all conditions. Draw O 72 at such an angle to theMiorizontal that its geometrical tangent is numerically equal to the total electrical resistance of the motor and the line, then S A Fig. 64. represents the loss of electro-motive force corresponding to the current O A, M D represents the loss correspond- ing to the current O 75, and so on. Produce O N = S A and complete the rectangle O N P H (dotted in the diagram). The area of this rectangle is evidently equal E 72y, and if we produce a square O B K L of equal area, the side O L will be equal to the square root of E 72y, and will, according to equation 29, represent E — Hence it follows that if we so load the motor that its counter-electro-motive force e = 7TZ, it will work with maximum commercial efficiency. The energy obtained at the motor spindle is represented by the area of the 150 ELECTRIC TRANSMISSION OF ENERGY. rectangle G F M the energy expended at the source of electricity is represented by the area of the rectangle O D F and the ratio of the two is the commercial efficiency. In the preceding chapter it was shown how, by the use of an absorption dynamometer, the speed for maximum commercial efficiency can be found experimentally ; it was also shown how, in the case of a series-wound motor, this determination can be made with a fair degree of approximation even without the use of a dynamometer. We can now employ the relations just found to make this determination for shunt or compound-wound motors also, without requiring the use of a brake. This may be ex- plained by an example from actual practice. One of the author’s dynamos (shunt-wound and designed to feed sixty glow lamps) was used as a motor. The electro-motive force of the source, which was a compound-wound dynamo, was 100 volts, current through motor when running empty was 4 amperes, speed 1,100 revs., and resistance of line and armature ’2 ohm. We have now By = and \/ F By— .y/80 = 8*94. To obtain best efficiency the motor must therefore be so speeded that its counter-electro-motive force e — 100 — 8*94 e = 91 volts. When running empty the counter-electro-motive force is 100-0*8 = 99*2. The best working speed is therefore 91 1100. — — = 1010 revolutions, y y ^ The current passing at that speed is 45 amperes, of which 4 amperes are required to overcome the internal friction of the motor, leaving 41 amperes to produce useful external work. By gearing the motor to the speed of PRACTICAL EXAMPLE, 151 1010 revolutions a minute, we shall therefore obtain 41 X 91 ^ 7 ^ = 5*07 H-P, actually available on the motor spindle. But it is not always possible to keep the motor running exactly at the right speed, especially if the load should vary, and in this case it becomes important to know how far on either side of the best speed a variation may take place without seriously reducing the efficiency. F or the motor above cited we find the following figures : — 1010 revs. 5*07 H-P. ^ = 91 c = 45 82*87o Com. effic* 1065 „ 2*07 „ = 96 = 20 76*7 944 „ 8-20 „ = 85 = 75 80*0 It will be seen from this table that a shunt-wound motor is fairly self-regulating, the range of speed between no load and full load being only about 157o the present instance. It should be here remarked that the motor de- scribed is intended for a working current of 45 amperes, and should not be loaded to more than 5 H-P for con- tinuous work. This reduces the extreme variation in speed to something under 97o* To show the influence of the resistance of the armature on the best speed and efficiency, a table is added, calculated for the same motor and the same electro-motive force, but with an additional resis- tance of *3 ohm in the circuit of the armature, making R = *5. 950 revs. 2*82 H-P. ^ = 86 c = 28 73*57o Com. effic. 860 „ 4*30 „ = 77 = 45 70*5 1000 „ 1*96 „ = 90 = 20 72 0 In practice, however, the additional resistance would not be placed in [the circuit of the armature, but in the line, where, indeed, it is unavoidable if the transmission of energy has to be made over a considerable distance. 152 ELECTRIC TRANSMISSION OF ENERGY. By inserting the resistance into the armature circuit only, we have not disturbed the condition under which alone formula 29) gives the best speed, viz., that the strength of the field shall be the same for all currents and speeds. This condition might be fulfilled even in the case of a transmission to a considerable distance if we excite the field of the motor separately or by a pair of separate wires from the distant source, but in practice such an arrange- ment would be too complicated and, as we shall see pre- sently, it would have no advantage in point of constancy of speed over the simpler plan of exciting the field of our shunt-motor direct from the line which brings the working current. The effect of an increased resistance in the line is in the first instance to lower the electro-motive force at the terminals of the motor. With a constant strength of field this would naturally lower the speed of the motor, but if its field magnets are not excited to the saturation point, the reduction of electro-motive force at the ter- minals of the motor will result in a reduction of the strength of the field, thus allowing more current to pass through the armature by which its torque and speed is increased until its counter-electro-motive force again balances the reduced electro-motive force of the supply. The variation of speed will therefore be smaller than would at first sight appear. But a little consideration will show that the gain in speed due to the increased armature current can never quite compensate for the loss of speed due to the reduced electro-motive force, and thus a pure shunt-wound motor, if fed from a source of con- stant electro-motive force can never be perfectly self- regulating. It must run faster when the load is thrown off, and it must run slower if more work is put on it. W e found the same to be the case with the pure series- THE SHUNT MOTOR. 153 wound motor^ but in a more marked degree. In this respect the shunt motor is preferable, as will be seen from the above tables (page 151), as its speed when running empty is only slightly higher than when loaded, whereas the speed of the series motor when running empty is excessive. On the other hand, the shunt motor has no starting power, since its armature, when at rest, forms a short circuit of very low resistance. To start a shunt motor it is necessary to arrange the switch in such manner that the field becomes excited before the current is allowed to fiow through the armature, and to avoid excessive sparking or heating of the armature, in cases where the motor has to start with the load on, additional resistances must be placed into the armature circuit, which are again cut out as soon as the motor has attained some speed. We shall now investigate the problem in what manner the electro-motive force of the source of supply must be varied in order to produce constant speed in a shunt- wound motor working under a varying load. Not to com- plicate the problem too much, we assume that the field magnets of the motor are, with the normal electro-motive force, excited to a very high degree, so that any slight variation in the magnetizing current cannot produce any material difference in the strength of the field. Under this condition the counter-electro-motive force in the armature of the motor will vary directly as the speed ; and since the latter is to be constant, the former will also be constant for all loads. Let y represent the armature current if the motor runs without load, let c be the current when there is a load, and let e be the constant counter-electro-motive force, then (c— y) e represents the external mechanical energy ; and since e and y are both 154 ELECTRIC TRANSMISSION OF ENERGY, constants, a variation of external energy, or, as we call it, a variation in the load of the motor, makes it necessary to vary the current c through its armature. This is done by raising the electro-motive force E of the supply if the load increases, and lowering it if the load decreases. Let R E — e be the resistance of line and armature, then c = — — — R and^ e + c R. We neglect as very small the amount of current required for the shunt on the field magnets. The equation shows that to maintain a constant speed of the motor the electro-motive force of the source ought to increase with the load. Its lowest value, when there is no load, will \>q E = e y R, and its highest value will be when load, and consequently current, are both maxima. The difference between the lowest and highest value will be the less, the smaller the resistance R of line and arma- ture, but it can never entirely vanish, for that would re- quire a line and an armature of no resistance. From the above considerations it will be seen that two shunt-wound dynamos can under no circumstances form a system of transmission of energy at constant speed of the receiving machine, because the electro-motive force of the generator — which we suppose to be driven by some prime mover at a constant speed — decreases as the current given out in- creases, whereas the motor requires exactly the opposite relation between these quantities. A shunt motor might be made to run at a constant speed by using an over- compounded dynamo for the generator. The principle of the compound-wound dynamo, or, as it is also called, of the self-regulating dynamo, is so well known that a few Avords only of explanation Avill suffice. Let the field magnet of a dynamo machine be wound with tAvo coils, one of fine high resistance Avire coupled THE COMPOUND DYNAMO^ 155 direct to the brushes, and the other of stout low resistance wire, coupled in series with the brushes and the external circuit. If the latter be open, no current passes through the main or series coils, and the magnetism of the machine is entirely due to the exciting power of the shunt coils. If the machine is properly designed, this amount of mag- netism should produce an internal electro-motive force exactly equal to that which it is desired to maintain at all currents in the external circuit, provided the dynamo is driven at a constant speed. If a current is permitted to flow through the armature, the electro-motive force measured at the brushes is naturally somewhat less than that created within the armature coils, on account of losses through resistance and self-induction, the loss in- creasing with the current. To compensate for this loss it is necessary to increase the internal electro-motive force, and this is accomplished by an increase in the strength of the magnetic field. This is brought about automatically by the main current itself, which assists the shunt current in exciting the field magnets. In a correctly compounded machine the increase of magnetization due to the main coils is sufficient, and no more than sufficient, to keep the external electro-motive force constant at all currents which can safely be passed through the machine. We say the machine is accurately compounded for constant terminal pressure. Now it is easy to see that we can overdo the thing, by putting on somewhat finer shunt wire, which will lower the electro-motive force when the machine works on open circuit or on a circuit of high resistance ; and by in- creasing the number of main coils so as to make the ex- citing power of the main current preponderate over that of the weaker shunt. In this case the increase of internal 156 ELECTRIC TRANSMISSION OF ENERGY. electro-motive force will more than counterbalance the loss through self-induction and resistance, and the result will be that up to a certain limit the external electro- motive force will rise as the current increases. Such an over-compounded machine could therefore be used as a generator, the receiver being an ordinary shunt machine, and we would thus obtain a system of transmission of energy at constant speed. Theoretically this is quite correct, but in practice there arises a difficulty due to the fact that the polarity of a compound machine can easily be reversed, especially if the influence of the main coils is considerably greater than that of the shunt coils. The author has attempted to establish such a system of transmission of energy at constant speed, but failed for the above reason. The failure was, however, more instructive than would have been the case had the system worked with theoretical per- fection, and an account of it was published at the time in The Electrician ” (April, 1885), of which the following is an abstract : — A series-wound dynamo, when used as a motor, runs in the opposite direction to that in which it has to be driven when used as generator. To make the machine run in the same direction (call it forward), the coupling between fleld and armature must be reversed. With a shunt machine this is not so ; the coupling between field and armature remains the same when used as a motor, and it runs always forward. The shunt machine used by the author was driven by a current from an over-com- pounded dynamo, the shunt of which was weak as com- pared to the main coils, and when the motor was doing little or no external work it behaved in a most erratic manner, running backward and forward alternately. At EXPERIMENT WITH SHUNT 310 TOR. 157 every reversal excessive sparking took place at the brushes of both the motor and generator, and it was clear that both machines were overstrained and would speedily come to grief if the circuit were not interrupted. To explain what takes place under these circumstances we will start with the assumption that the generator is kept running at a constant speed, and that the motor is switched on whenever power is required. This is the usual practice where motive power is used at intervals for industrial purposes. Since the leads from the generator remain always charged, the moment we switch the motor on, a large current will pass through its armature and a small current through its magnets. As the motor is at rest there is no counter-electro-motive force to oppose the flow of electricity through the armature, and the result is a momentary excess of current. The immediate effect of this is to start the armature revolving at a high speed be- fore the magnets have had time to become fully excited, for it must be remembered that an armature will revolve in a non-excited fleld, though with considerable waste of current. The speed required to produce a given counter- electro-motive force is the greater, the weaker the excita- tion of the fleld, and hence the motor starts off at a much faster speed than it would have in regular Avork with its magnets fully excited. On account of self-induction in the shunt field magnet coils, Avhich is considerable, the magnets require some time to become fully excited, and whilst the strength of the field is growing the armature is gathering speed and storing mechanical energy. When at last the field magnets are saturated, the armature of the motor has attained such a speed that its counter- electro-motive force not only equals, but exceeds the dif- ference of potential maintained between the leads by the 158 ELECTRIC TRANSMISSION OF ENERGY. generating dynamo, and the current is forced back through it. For the time being the motor acts as a generator, the energy stored mechanically in its revolving armature being returned to the circuit in the form of current. This reverses the polarity of the compound dynamo (its shunt coils being weak, as stated above), and now both the generator and the armature of the motor are working in series, the generator assisting instead of opposing the current started by the motor. At this moment we have the following state of things : — The field magnets of the motor have just attained their maximum of magnetization with their original polarity ; the polarity of the generator has been reversed, and an excessive current, in an oppo- site direction to that which produced motion, fiows through the armature of the motor. Consequently the latter is quickly brought to rest, and started backward at a high speed. It now opposes a certain counter-electro- motive force to the current from the generator, but it is not an increasing force as before. It is a decreasing one, because the original excitation of the motor field magnets is gradually vanishing, by reason of the reversal of polarity in the main leads, from which these shunt coils are fed. Just as it took a certain appreciable time of several seconds for the magnets to become excited, so does it take time for them to lose their magnetism. Even- tually there arrives a moment when all the original polarity in these magnets has vanished, and when, there- fore, the force impelling the armature to run backward has also ceased, though there is still an excessive current passing through it. A moment later the armature comes to rest, and begins to run forward again at a high rate of acceleration, when the whole cycle of phenomena just de- scribed is repeated, but this time with a current in the EXPERIMENT WITH SHUNT MOTOR. 159 reverse direction to the first. The third cycle will start with a current in the same direction as the first, the fourth cycle will start with an opposite current, and so on.” A similar phenomenon was observed by M. Gerard- Lescuyer, who used a Gramme series-wound dynamo as a generator, and a magneto machine as a receiver. He called the phenomenon an electro-dynamic paradox, and a description of it will be found in The Engineer ” of Sept. 17, 1880. CHAPTER VI. Classification of Systems according to Source of Electricity — Transmission at Constant Pressure — Motors mechanically governed — Self- Regulating Motors — Transmission at Constant Current — Difficulty of Self-Regulation — Motor for Constant Current made Self- Regulating — Application to Transmission over large Areas— Continuous Current Transformator — Transmission betw^een two Distant Points — Loss of Current by Leakage — Theory — Commercial Efficiency — Conditions for Maximum Commercial Efficiency — Self- Regulation for Constant Speed — Practical Example. It will be necessary to distinguish between different sys- tems of electric transmission of energy, according to the source of electricity. An almost endless variety of cases may present themselves in different applications of elec- trical transmission, but three systems are of special in- terest, because most frequently occurring in practice. These are the following : — 1. Transmission of energy from primary or secondary batteries at short distances to one motor only. 2. Transmission of energy from one or several dynamos to a number of motors placed upon the same circuit, but working independently of each other. 3. Transmission of energy between two distant points by means of one generator and one motor. We may also make another classification according as the motors are intended for a constant or variable load, or a constant or variable speed. Generally speaking, the systems of transmission coming under heading 1) are not required for a constant load, nor is it of any great impor- SYSTEMS OF ELECTRIC TRANSMISSION, 161 tance that the speed should remain constant under a variable load. We shall not enter into a minute descrip- tion of these cases here, as the investigation of electric tramways and railways, worked by accumulators, will alford ample opportunity of entering into details. System 2) is that presenting most difficulties on account of the condition that all the motors must be independent of each other. The case is further complicated by the requirement that each motor should run with the same speed when empty or loaded. A moment’s consideration will show that the last condition is an absolute necessity if we would make the electric transmission of energy of real practical use to small domestic industries. The artisan or small manufacturer would have his motor con- nected to a common system of service leads, and when- ever he required power he would switch the current on to his motor. In doing so he must not disturb any other work which, at the same time, may be done elsewhere from the same service mains, such, for instance, as light- ing or working other motors ; and further, his motor should always run at the same safe speed, whether it is giving him little or much mechanical energy. Most operations requiring the use of tools as turning, planing, &c., can only be properly performed at a certain fixed rate of speed, and the machinery must be kept going at that rate at all times. System 3) presents difficulties of a different nature. Since we have to deal only with one generator and one motor, it is easier to make each fit the other, and as a rule the load is fairly constant, so that regularity of speed is not difficult to obtain. In this case the difficulty lies more in the necessity of proper insulation of line and machinery. Generally speaking, the system is required M 162 ELECTRIC TRANSMISSION OF ENERGY, for long distance transmission, and to obtain an econo- mical arrangement, both as regards first cost and com- mercial eflSciency, the use of a high electro-motive force is necessary. This entails some danger to human life, and some difficulty in maintaining an efficient insulation. Both these points can, however, be satisfactorily dealt with, if proper care is used in the design and execution of the work. As regards the danger to human life in- volved in the use of electric currents of high pressure, this is generally greatly overrated. It is quite possible for a man who with both hands should touch the positive and negative wires in a non-insulated part, to be killed or severely injured if the pressure is over two or three thou- sand volts, but the accident can be rendered almost im- possible if due precaution is taken. A circular saw if only lightly touched whilst revolving will cut a man’s finger off, and what can be more dangerous than a pair of powerful spur wheels? Yet we have found means of protecting life very effectually from destruction by purely mechanical means, and the experience of the past fe’w 3 ^ears has shown that equally efficient protection can be provided from the electrical danger. System 2) is best described as electric transmission and distribution of energy from one central station to several distant points. This distribution can be made on the parallel or on the series system. In the first case the electro-motive force (or pressure) between the positive and negative mains must be kept constant, and the motors are connected all in parallel from the mains ; in the second case the current passing through the mains must be kept constant, and each motor, when at work, is traversed by the same current. The pressure at the station must be the greater the greater the number of motors at work. PERIODIC GOVERNOR. 163 In the first case the pressure is kept constant, but the current delivered into the mains must be the greater the greater the number of motors at work. We have thus to distinguish between distribution at constant pressure and distribution at constant current. Electric Distribution of Energy at Constant Pressure. We must now inquire into the theoretical conditions of this case. It will be evident at the outset that for econo- mical reasons any attempt to obtain constancy of speed by the use of artificial resistances can only lead to a partial and not very satisfactory solution of the problem, and had better not be made if other means are at hand. This happily is the case in the present instance. We have two means by which we can without waste regulate the power of the motor to the work and yet keep it running at a constant speed. First we may apply a mechanical device by which the current is periodically cut off in proportion as the work is cut off, and, secondly, we may apply an electric device in the shape of special winding of the field magnets of the motor by which the torque exerted by the armature is automically regulated so as to correspond to the mechanical load. As regards the first system, Professors Ayrton and Perry have in a paper on Electro-Motors and their Government ^ shown how this can be done. They say : The method of cutting off the power as hitherto employed has this serious defect^ that instead of the power cut off being directly in propor- tion to the work cut off, the arrangements have been such ^ “ Journal of the Society of Telegraph Engineers and Electricians,” No. 49, vol. xii. 1883. 164 ELECTRIC TRANSMISSION OF ENERGY. that either all power was cut off or none, so that the motion of the motor was spasmodic, just as in an ordinary gas-engine, which suffers from the same defects, that full charge of gas or no charge are the usual only alternatives. An electro-motor governor of this type, which may be called a ^ spasmodic governor,’ consists merely of a ro- tating mercury cup into which dips a wire, which makes in this case contact with the mercury, and so completes the circuit when the speed is slow, but which, on account of the hyperbolic form assumed by the surface of the mer- cury as the speed rises, ceases to dip into the mercury at high speeds, and so breaks contact.” Later on the in- ventors say : The first improvement we made in govern- ing consisted in replacing the ^ spasmodic governor ’ by a ^ periodic governor.’ With our periodic governor the power is never cut off entirely for any length of time, but in every revolution power is supplied during a portion of the revolution, the proportion of the time in every re- volution during which much power is supplied to the time during which less is supplied depending on the amount of work the motor is doing. Our periodic governor, then, differs from the spasmodic governor in the same way that a good loaded steam-engine governor differs from the ordinary governor of a gas-engine. One of the ways of effecting this result is as follows : a brush. A, Fig. 65, lies on the rotating piece, B A, the cylindric surface of which is formed of two conducting portions connected with one another through any resistance, and the brush, A, is moved along the cylinder B K under the action of the governor balls. When the brush A is touching the contact part jB, the motor is receiving current directly, but Avhen it rests on the part A, the motor receives cur- rent through the resistance which is interposed between PERIODIC GOVERNOR, 165 B and K, If the governor balls fly out^ the brush is moved along so that there is contact with K during a greater part of the revolution than before ; and if the governor balls come together, the speed of the motor being too small, the brush is moved in the opposite direction so that it makes contact with B for a longer time during each revolution. If the motors are in series, we arrange that the periodic governor shunts the current periodically, in- stead of introducing resistance. In this case the connec- tions are as follows : B is made of wood, while K is made Fig. 65. of metal. K is connected to one end of a shunt coil, the other end of the shunt being connected to one of the ter- minals of the motor and A is connected to the other ter- minal of the motor. If, then, A rests on 5, the shunt is inoperative and all the current passes through the motor ; whereas, if it rests on the shunt is in operation, and part of the current only passes through the motor.” It will be seen that both these governors invented by Pro- fessors Ayrton and Perry, have partially, at least, the fault of depending on artificial resistances whether they 166 ELECTRIC TRANSMISSION OF ENERGY. be used for parallel or series work. The loss of energy thus occasioned can be reduced by making the resistance high for parallel^ and low for series work, and on purely theoretical grounds it could even be entirely prevented by making the resistance infinite, that is, breaking the circuit altogether during a portion of each revolution when working in parallel. But this would produce an unequal turning force, and would also entail destructive sparking between the brush, and the contact pieces B and K. Even if the resistance between B and K or that of the shunt coil between K and one terminal of the motor is fairly low, there must be some sparking ; and the inventors say in their paper that with any such governors it is difficult to entirely prevent sparking, and that on this account motors wound so as to be self-regu- lating without any mechanical device are preferable. Broadly speaking, the self-regulating motor is the con- verse of the self-regulating dynamo wound for constant pressure. In a properly compounded dynamo the pres- sure at the terminals must remain constant, although the resistance of the external circuit may vary between wide limits, causing an inversely proportional variation in the external current. The power required is approximately proportional to the current. The machine works, there- fore, under these conditions : Speed constant — Electro- motive force constant — Current variable — Power re- quired to drive the machine also variable, but propor- tional to current. Now, in a self-regulating motor the conditions are : Electro-motive force constant — Speed constant — Power variable — Current required to drive the motor also variable, but proportional to power. It has already been pointed out that in a general way dynamo and electro-motor are convertible terms ; and SELF-REGULATING MOTOR. 167 although there are cases when it is impracticable to work a motor as a dynamo, it is always pei^fectly easy to work a dynamo as a motor. From this general convertibility it is reasonable to expect that a properly compounded dynamo can without any alteration in the connection be- tween its field magnet coils and armature, be used as a self-regulating motor, the only condition being that it shall be supplied with current at a constant electro-motive force. When speaking of a self-regulating motor in the sense that its speed of rotation shall automatically be kept constant whatever variation might occur in the load or mechanical resistance which the armature of the motor has to overcome, it must be understood that this refers only to such cases where the load varies between zero and a maximum not beyond the capability of the motor. If we throw an excess of load on to the motor, it will pull up or slacken speed, and thus cease to be self-regulating, just as the electro-motive force at the terminals of the best compound-wound dynamo will be lowered if we allow an excess of current to fiow. But within a reason- able limit of load in the case of the motor, and a reason- able limit of current in the case of the dynamo, both machines can be made self-regulating, and this result is obtained by the same means, that is to say, the same winding which will make the dynamo give a constant electro-motive force, will make the motor run at a con- stant speed. This result might be expected on the ground of the general convertibility of these machines, but since it is of great practical importance, special pi’oof is desirable. This can be easily obtained from our for- mulas in Chapter III. According to equation 7) the torque exerted by an armature current, in a field of Z lines, is in absolute measure : 168 ELECTRIC TRANSMISSION OF ENERGY. ZNt C, 7T It is independent of the speed, and since Nt is constant for any given motor, the torque or turning moment exerted by the armature is directly proportional to the product of the strength of field and armature current. By increasing either or both these factors we are able to overcome our increased load. Since the electro-motive force is supposed to be constant, it is evident that a varia- tion in load must be compensated mainly by a variation in current. Assuming that the ends of the shunt coils are coupled to the terminals of the motor — not to the brushes — we have, retaining the notation of Chapter III., the following equations : = E,- E, = E, -(r^ + r,) C^. The counter-electro-motive force E^ is, according to equa- tion 5), expressed in volts by E,= Z NtnlO ^ 5) + 0 C^ = ZNtnlO-\ Now the condition under which the motor is to be used is that the electro-motive force at its terminals E^^ shall be kept constant. We have, therefore. Constant Et = (r^ + r^) + Z Nt n 10“®. Since the speed n must also remain constant if the motor is to be self-regulating at all loads, the only variables are Ca and Z, which have to satisfy the above equation. In other words, we may regard the field Z’ as a function of the armature current and the condition that the motor be self-regulating is brought down to this, that the strength of its field shall depend on, and vary in a SELF-REGULATING MOTOR. 169 certain manner with the current passing through the motor. Z — X 10-® Nt n We see by this equation that Z will be the smaller the greater and since is almost directly proportional to the mechanical load of the motor, we arrive at this, at first sight, startling result : that the heavier the work we impose upon the motor, the weaker must be its field. It might have been thought that as additional load is thrown on, we ought so to arrange matters that the magnetism of the field magnets becomes strengthened, and able to exert an increased magnetic pull on the armature. But a moment’s reflection will show that the effect of such an arrangement would be to reduce the speed. The mag- netic pull exerted by the field magnets upon the armature does not depend on the strength of magnetism in the field magnets only, but is the product of that quantity and the current in the armature coils. An increase of pull may therefore be brought about either by making the field stronger, or by increasing the current in the armature, or by both means combined. If we make the field stronger, we not only increase the magnetic pull exerted on the armature, but we also increase the counter-electro-motive force, as will be seen from equation 5), page 82, and thus check, or at least reduce, the flow of current through the armature at the very moment when we want most power. The motor would thus run slower until its reduced counter-electro-motive force again allows a current to pass of sufficient strength for the work imposed on the motor. If, on the other hand, w^e seek the increase of power by allowing more current to pass through the armature, we do not increase the counter-electro-motive force, but we 170 ELECTRIC TRANSiMISSION OF ENERGY, have a slight increase in the loss of electro-motive force due to the resistance of the armature. To compensate for this slight increase of loss, it is necessary to weaken the field somewhat for heavy currents, and thus bring about the reduction of counter-electro-motive force by an amount corresponding to the increased loss of electro- motive force due to the resistance of the armature. If the motor runs without doing external work is almost zero, and we have the strongest field, Nt 71 [ which is entirely due to the shunt coils. Let now a load be thrown on. The immediate effect will be to slightly re- duce the speed. The counter-electro-motive force which previously was nearly equal to will thereby become somewhat reduced, thus allowing a considerable current to pass through the armature and the series coils of the magnets. This again accelerates the armature until the normal speed is reached. The direction of winding of the series coils must be evidently such that the main working current tends to demagnetize the field magnets. Now in an ordinary compound-wound dynamo, the series coils are wound and connected in such a way that the main current tends to increase the magnetism produced by the shunt coils. If we use such a dynamo as a motor, the current in the shunt coils will remain the same as before, the current in the armature will flow in the re- verse direction, and therefore produce motion — instead of resisting it, as is the case when the machine is worked as a dynamo ; and the current through the series coils will also flow in the reverse direction, thus tending to weaken the field magnets. It will be seen that these are precisely COMPOUND DYNAMO USED AS MOTOR. 171 the conditions which our theory indicates as necessary, in order to make the motor self-regulating, and we find that it is correct to say that a compound-wound machine can be used either as a self-regulating dynamo or as a self- regulating motor. There may be slight differences in the exact proportion between shunt and series coils in both cases, but the general principle of compounding is the same for either purpose. A question of considerable practical importance is that of the relation between the weight of the motor and the maximum of mechanical energy it can give out. Since that maximum must be given out when the field is weakest, whereas in a non-self-regulating motor the arrangements can always be so made that the maximum is given out when the field is strongest, it is evident that, for a given power, the self-regulating motor must be heavier. This is certainly a drawback, and it becomes necessary to know what price, in the shape of increased weight, we have to pay for the advantage of automatic regulation. Our formula for Z enables us to form a rough estimate of this increase in weight. The difference between the initial value of Z and the minimum value is due to the product (r^ -f- r^) The greater this product, the more must the field be weakened, and the smaller is the maximum of power obtainable with a given weight of motor. It is therefore of importance to keep the product (r^ -f as small as possible, and since (7^, which we must consider as the primary source of power, cannot be reduced, it is evident that the re- sistance of series coils and armature should be as small as possible. Now, in a good modern motor, the loss of electro-motive force occasioned by the resistance of these parts, varies between 5 and 10 per cent, of the electro- 172 ELECTRIC TRANSMISSION OF ENERGY. motive force applied at the terminals ; take 7 per cent, as a fair average, and we find that if the initial field is represented by, say, 1,000 lines, the field at full work will contain 930 lines. Now if the motor were not self-regulating, the field at full power would contain 1,000 lines, and thus be able to develop about per cent, more mechanical energy. If, on the other hand, we wish the two motors to develop the same maximum of mechanical energy, the field magnets of the self-re- gulating motor Avould require to have 7l per cent, more cross sectional area. Since series and shunt coils act differentially, a larger amount of copper is also required. This excess would probably amount to about 2i per cent, of the total weight, so that in all the self-regulating motor will weigh 10 per cent, more than an ordinary motor which is not self-regulating. This does not seem too high a price to pay for the safety and general comfort of a self- regulating motor, and the experience gained in American and Continental towns having central electric light stations from which current is supplied to a network of mains on the parallel system has proved that it is perfectly practicable to utilize the same mains for distributing mo- tive power to artisans and small manufacturers by supply- ing them with such self-regulating motors. Electric Distribution of Energy at Constant Current. This problem is not of so easy solution as the distribu- tion of energy at constant pressure, and the difficulty is a fundamental one. It lies in this, that there is no direct connection between the speed of a motor and the current which flows through its armature. There is a direct con- nection between speed and electro-motive force, and. DISTRIBUTION AT CONSTANT CURRENT. 173 therefore, self-regulation is possible without the use of any external apj^liance in the shape of a mechanical governor or other apparatus which controls the power. But where the current is constant, some kind of external governor is necessary. This follows also immediately from M. Marcel-Deprez’s experiments cited in Chapter III, page 91. We have seen that the speed was totally independent of the current, the latter remaining throughout the range of each experiment practically con- stant, whereas the speed was in some cases increased five- fold, by simply increasing the electro-motive force of the source. When a number of motors are coupled in series, as would be the case in a general system of distribution, the difficulties are much increased. To test this matter experimentally the author has placed three precisely similar motors (series-wound) in series into the same circuit. The current was supplied by a dynamo, and the three motors were loaded by brakes to as near as may be the same amount. It was then found quite im- possible to keep all three motors going for any length of time at the same speed. The least irregularity in the current, or the least variation in the friction of the brakes, would cause first one and then the other motor to come to rest, whilst the speed of the remaining motor increased to a dangerous extent. Professors Ayrton and Perry have in the paper above mentioned proposed to make motors self-regulating if worked by a constant current in the following way ; The field magnets. Fig. 66, are wound differentially with a fine wire coil, which is a shunt to the armature only, and a thick wire coil which is in series with the armature and main current. The armature and shunt coil consti- tute a shunt motor, the armature and main coil a brake 174 ELECTRIC TRANSIMISSION OF ENERGY. generator which is intended to absorb any surplus power if the load is thrown off. As far as the author is aware the system has not been tried in actual practice, and there are theoretical reasons for expecting that it would not work. From equation 7) it will be evident that the field must be strongest when the load is greatest. Now suppose that the differential winding could be so propor- tioned that for a given load the field is exactly of the right strength to produce the normal speed. Now let a very slight additional load be thrown on. The immediate Fig. 66. effect will be to slightly reduce the speed, and in conse- quence of the reduction in speed the magnetizing current in the fine wire coil will also be reduced. The field will thus be slightly weakened. This will further reduce the speed and again weaken the field, and so on, until the armature comes to rest. At that moment the magnetizing influence of the main coils, which is in the opposite direc- tion to that of the shunt coils, will alone exist, and the field magnet instead of presenting a A7pole to the arma- ture, as shown in the illustration, will present a S pole to it. The tendency must therefore be to reverse the SELF-REGULATING CONSTANT CURRENT MOTOR. 175 motion, and thus the slight addition of load has not only brought the armature to rest, but actually caused a tendency to run backwards. Whether it will run back- wards depends on the relative magnetizing power of the main and shunt coils. In a subsequent paper read by the same authors before the Physical Society on May 26, 1888, they suggest to make motors self-regulating for constant current by omitting the series coils on the field magnets altogether, and inserting into the armature circuit a storage battery, the electro-motive force of which helps the current on. As far as the author knows no practical test of this system has been made, and it is easy to see that the objection of insta- bility, which was pointed out above for the first arrange- ment, also applies in this case. Constant current motors are extensively used in the United States on arc light circuits, but in all cases the regulation of speed is effected by some additional mechanism which either shifts the brushes or alters the exciting power. The attempt to make a con- stant current motor self-regulating without such additional devices has very little chance of practical success. An arrangement devised by the author, and which seems to promise somewhat better to fulfil the condition of constant speed, is shown in Fig. 67. A is the armature of a series-wound motor mounted upon a spindle, to which is also attached the armature a, of a small series- wound dynamo which has no other work to do but to supply current for demagnetizing the field magnets of the motor. The main current magnetizes them in the direction, say, from B to C, the auxiliary current from the dynamo acts in the direction/ to g, and tends to de- magnetize them. 5 c is the field magnet coil of the dynamo. Now for each dynamo working on a closed 176 ELECTRIC TRANSMISSION OF ENERGY. circuit of constant resistance^ as in the present case, there exists a critical speed at which it will begin to give a current of some strength. Below that speed it gives hardly any current, and above that speed it gives almost at once the full current. The motor should be so geared as to run at the critical speed of the little auxiliary dynamo. If now an additional load be thrown on, the immediate result will be to reduce the speed of the motor, thereby causing the armature of the dynamo to run Fig. 67. ^fVWVl^ /[/WVAA ^-QJ\S\f\f\f\ C below its critical speed. The dynamo will thus partly or entirely lose its current and the demagnetizing influence which previously has kept the fleld below its full strength, will to a greater or lesser degree be withdrawn. The strength of the fleld will thus be increased, and an additional magnetic pull will be brought to act on the armature, by which it can overcome the increased load. In case the load be entirely thrown off, the motor will have a tendency to race, but this tendency will be im- mediately checked by the auxiliary dynamo, the current SELF-REGULATING CONSTANT CURRENT MOTOR. 177 from which increases considerably with a very slight increase of speed. Its demagnetizing influence is thus enormously increased, and the field of the motor is weakened to such an extent that there is just power enough left to drive the dynamo but no more. To make this arrangement successful it is necessary that the field magnets of the auxiliary dynamo be made of very soft iron, so as not to retain any considerable amount of per- manent magnetism, which would alter the critical point as between an increasing and decreasing speed. The more sensitive and unstable the dynamo can be made, the better. For this reason it is also necessary to place the two armatures a considerable distance apart on the same spindle, so that the field magnets of the motor may not induce magnetism in the field magnets of the dynamo, and thus disturb the critical point. In practice it would probably be found necessary to place a bearing between the two armatures, and that could easily be so shaped as to act as a screen between motor and dynamo. According to the classification made in the beginning of this chapter we have now to consider System 3), which comprises the transmission of energy betwen two distant points by means of one generator and one motor. Let £'3, and jE'^, represent respectively the electro- motive force in the armature, at the brushes and at the terminals of the generator, and let e^y and e^y repre- sent the same for the motor. Let R^y represent the resistance of the armature, and magnetizing coils of the generator, and r^, represent the same for the motor, then we have, according to the equations 15) to 22), if both machines are series-wound, the following rela- tions ; N 178 ELECTRIC TRANSMISSION OF ENERGY. Generatoe. E^ = E^+ CR^. JE^ — CR^* Et ^ C{Ra + Rm)> Motor. + O- C being the current sent by the generator into the line, and c being the current received by the motor. If the insulation of the line were perfect, these two currents would be equal ; but in practice some small leak of cur- rent from the positive to the negative circuit, when the line extends over several miles, might take place and then we must assume ^ C>c. The loss of current C — c represents, as far as the generator is concerned, a waste of energy expressed by the product Ef {C — c) watts. As far as the motor is concerned, this leak not only re- duces the current which is available at the receiving station, but it has also the effect of reducing the available electro-motive force beyond the value corresponding to the current c. It will be clear that unless the leak occurs close to the generator, part of the line will have to carry a current larger than c, and thus the loss of electro-motive force due to the resistance of the line must be greater than the product of that resistance and the motor-current €. If the line is throughout its entire length equally well insulated, each unit of its length will have the same in- sulation resistance, which should be very high in com- parison to the conducting resistance itself. In a perfect line it should be infinite, but, as remarked above, this may not be obtainable in an overhead circuit going many miles across country. Let ^ represent the conducting re- LOSS OF CURRENT BY LEAKAGE. 179 sistance of the line, and let i denote the insulation resis- tance between the positive and negative lead for unit length. If the distance from the generator to the motor be /, the total insulation resistance as measured on i a Wheatstone bridge would be Knowing this from actual measurement, it might be thought that by the application of Ohm’s law we could easily find the leak, C — c, by simply dividing the electro-motive force be- tween the wires by the insulation resistance. This would, however, not be correct, for the simple reason, that the electro-motive force between the wires is not a constant, but diminishes in a certain ratio as we approach the dis- tant end of the line, the actual law by which this diminu- tion takes place depending not only on the resistance of the line and the current, but also on the insulation resis- tance itself. The question is therefore not so simple as it at first sight appears. An approximate solution suffi- ciently accurate for practical purposes is the following : Let £ represent the electro-motive force between the leads at the distance x from the generator ; let the dis- tance be increased to x + dx and the leak of current cor- responding to length dx be dc, the drop in electro-motive force corresponding to that length being ds. Then the following equations evidently obtain : — ds = c j dx. — dc = ~ dx. z From these equations we obtain 180 ELECTRIC TRANSMISSION OF ENERGY, and by integration — Constant. To find the constant we apply the formula to the home end of the line^ where 2 = c = C, and obtain between that and the far end the relation E,^-e;= i i{C^ from which we find c ^ J c^- I (E;-e;). This gives the current arriving at the motor, but in a somewhat inconvenient form. To simplify the expression we develop the square root into a series, and since the second term is very small in comparison to the first we can neglect the second and subsequent powers. 1 — 2 gi C c= C- Now E^ — = [Ef 4” ^t) (Et — e^) and -j = J, the total insulation resistance of the line. Hence The leak of current is E, + e,\\ E, — e, C c 30 ). E f Of . Now -E— — - is the average electro-motive force between ( E “f“ ^ \ 1 ^ jj represents the current which under that average electro-motive force LEAKAGE AND ECONOMICAL SPEED. 181 would flow through J, the total insulation resistance. This It 7 ^ £ current, multiplied by — gives the actual leak. ^ C will be observed that g (7, being the product of a resistance and a current, represents a difference of potential, and in this case it represents the electro-motive force which would in a line of perfect insulation be required to drive the full initial current C through the circuit, supposing the far ends were in metallic contact, g C represents, there- fore, the loss of electro-motive force if there were no leak. The actual loss, is naturally somewhat greater, and thus the quotient between the two must always be greater than unity. From this it follows that the loss of current due to leakage along the line is slightly greater than the figure we obtain by dividing the average electro- motive force by the total insulation resistance. Where the insulation resistance is very high, and the conduct- ing resistance very low, the leak will with sufficient accuracy be expressed by C-o = (^)i but when the conditions are less favourable formula 30) should be used. It is necessary in this place to briefly consider the in- fluence of the leak on the total efficiency of a system of electric transmission, especially with reference to the most economical speed of the motor. In text books, and in scientific articles on the subject, the assumption is gene- rally made that the insulation of the line is perfect. This may be so in some favourable cases, but a general theory must include all cases ; it should, therefore, take the leak into account. As far as the writer knows, this has only been done by Professor Oliver Lodge in his treatise on 182 ELECTRIC TRANSMISSION OF ENERGY. the transmission of power by dynamo-machines, published in The Engineer,” 1883. It is also generally stated that the efficiency is the greater, the nearer the counter- electro-motive force of the motor approaches the electro- motive force of the generator. It has already been pointed out that this is quite wrong (see Chapter V., page 142), even if the motor be worked by a current of constant electro-motive force, such as would be the case if the generator were a self-regulating dynamo placed close to the motor, and connected with it by leads of practically no resistance and perfect insulation. But when the leads have considerable resistance, and especially if their insulation is not absolutely perfect, the statement above referred to and which is carefully perpetuated by successive writers, becomes still more erroneous. From equation 30) it will be seen that the leak is the greater, the greater At the same time an increase of has the effect of checking, or at least diminishing, the working current c, thus reducing the amount of energy received. Since the energy lost by leakage increases with the counter-electro-motive force, whilst the energy actually given out by the motor at first increases with the counter- electro-motive force up to a certain point, but beyond it decreases again, it will be clear that high efficiency can- not be obtained by allowing the counter-electro-motive force to approach too near to the electro-motive force of the generator. In the following investigation we shall assume for the sake of simplicity that there is absolutely no leak in the line. The results obtained will, therefore, be to some extent inaccurate, but they can be rectified by using equation 30). Thus with a perfect line we would obtain certain values for C = c and ; and the generator would have to give the current and electro- ELECTRICAL EFFICIENCY. 183 motive force thus determined. Now assume that, after a certain time, the line begins to leak. This will reduce the energy received by the motor, and consequently also that given out by it. It is evident that this loss can be compensated by running the generator at a higher speed ; in other words, by increasing E, and C beyond their original values. A similar plan we follow in the mathe- matical investigation. We assume at first that the insu- lation of the line is perfect, and we are thus enabled to use formulas of great simplicity. This gives a certain set of conditions for the generator. If the line is in reality in as perfect a state as assumed, the problem is solved. If, however, the line leaks, we rectify the values for and C by using equation 30). This gives a new set of conditions for the generator, and the mechanical energy necessary for actuating the generator must be calculated for these new conditions. The condi- tions of the motor are not altered thereby. The electro-motive force lost in the line is which must be equal to the difference of electro-motive forces at the terminals of generator and motor E, = gc + e,. The internal electrical energy of the generator is c that of the motor is c and the proportion between the two is the electrical efficiency of the whole system. Electrical efficiency = By combining this expression with the above equations we find also : Electrical efficiency — t .... 31). ^ "h + Rri) It is evident that whatever may be the resistance of the 184 ELECTRIC TRANSMISSION OF ENERGY. line g, or in other words whatever may be the distance to which the energy has to be transmitted, we can always obtain the same electrical efficiency by suitably vary- ing c and e^. The higher the counter-electro-motive force of the motor, the greater is the electrical efficiency. Now there are two means by which we can raise the counter-electro-motive force. The one is by increasing the speed, the other by employing machines containing a large number of turns of wire (iW) on their armatures. The first expedient is limited by the mechanical diffi- culties generally attendant on the use of excessive speeds, and the latter by the difficulty that the internal resistance of the machines is the greater the more turns of wire they contain. This, in itself, would not effect the result if the electro-motive force would increase in the same propor- tion as the resistance of the machine. But this is not the case. If a given size armature core be wound with many turns of fine wire, and a precisely similar core with such a number of turns of stouter wire, that both windings fill exactly the same space, the weight of copper contained in the armature wound with stout wire must always be some- what greater than in the other, because the space wasted by the insulating covering on the wire is less. It is clearly not admissible to reduce the thickness of insulation in the same ratio as the gauge of the wire. A minimum thickness is absolutely necessary for the safe handling during the process of manufacture, and moreover the finer wire is intended for an armature of higher electro-motive force and should for this reason alone have rather better in- sulation than the thick wire, which is intended for a lower electro-motive force. A good practical rule is to employ a covering of cotton about 8 mils for wires of all sizes up to about 120 mils. The diameter of the covered wire is COMMERCIAL EFFICIENCY. 185 thus by 16 mils greater than that of the single wire. Now it can be shown that the energy wasted in heating the wire of the armature is inversely proportional to the weight of copper employed, and therefore, with the arma- ture of stouter wire, the same electrical output can be obtained at a smaller cost of energy wasted in heating the wire. The same holds good for the field magnet coils. The dynamo wound with stouter wire, will, therefore, be the most economical of the two, as its internal resistance will be relatively small as compared to its electro-motive force. Inversely, if we wind the machines (generator and motor) with very fine wire in order to obtain a high electro-motive force, we increase their resistances, 72^, in a somewhat quicker ratio, and thus lower their efficiency, taken apart from the line. As regards the line resistance the higher the electro-motive force the better, and it will be evident that taking these two things into consideration there must be one particular value for the electro-motive force for which the electrical efficiency becomes a maximum. This value can be found in each given case by assuming different windings for generator and motor, and calculating their electro-motive forces and resistances. By inserting the data thus obtained succes- sively into equation 31) it can easily be seen which is the best. We suppose that the resistance of the line is given. The electrical data thus obtained can only be regarded as a first approximation to a solution of the problem, be- cause they were obtained on the basis of the highest elec- trical efficiency, whereas the question of importance is the actual or commeixial efficiency. It is sometimes as- sumed that the commercial efficiency of dynamos and motors bears a fixed proportion to their electrical effi- ciency, and if that were so we could obtain the actual 186 ELECTRIC TRANSMISSION OF ENERGY. efficiency of our system of transmission by multiplying equation 31) with that fixed proportion. But this would not be correct. It is evident that the commercial effi- ciency of a motor cannot be a fixed quantity, but must depend on the power given out, being, generally speak- ing, the higher the nearer the work done by the motor approaches to the maximum for which it is designed. This relation is best expressed in the manner adopted in Chapter V., by assuming that a certain minimum of cur- rent, 7 , is necessary to overcome the mechanical and mag- nectic friction of the motor, and that all the power corre- sponding to the difterence between this minimum and the actual working current is available for external work. Similarly we assume that a certain minimum of current, multiplied by the internal electro-motive force of the generator, represents the mechanical energy absorbed by mechanical and magnetic friction. We have, therefore, the following relations : Generator. Woi’k absorbed, W= {c -j- g) E„. % Motor. Work given out, w = (6— r) e^. Put + -S = i?, and r^-= r, then for series- wound generator and motor we have Ea = e„-\- c R + r), ^ = (<^ + ^) + c (f + 72 -f r)) And the commercial efficiency of the whole system is \c +g) _ / c —y \ ” \c + ^/e» + c (? + 72+ r) 32 ). COMMERCIAL EFFICIENCY. 187 A question of practical importance is that concerning the working conditions under which^ in a given system of transmission, the commercial efficiency becomes a maxi- mum. As already shown, the first condition for attaining this object is to work the generator at as high a speed as mechanically safe. We shall therefore assume that its electro-motive force is a constant and as high as pos- sible. The variables are the current c, and the counter- electro-motive force of the motor. If we allow the motor to run too slowly it will allow a large current to pass, but this will entail a considerable waste of energy in heating the line and the two machines. If we speed the motor too high, this waste will be very small, but the high counter-electro-motive force will only allow very little current to pass, and in this case the work done by the motor will be small, thus again lowering the commer- cial efficiency. Between these two extreme cases there must evidently exist one current and one counter-electro- motive force for which the commercial efficiency becomes a maximum. To find these values we form the first dif- ferential quotient and equate it to zero. Thus the most favourable current will be found by the equation dv] and the most favourable counter-electro-motive force will be found by the equation dn de^ = 0 . Writing for the sake of brevity E for E^, and e for and R for the sum of the resistances ; R r, the first equation gives, (c+ff) (E-2Rc + yR) -c(E + yR) + R y E '=■ 0^ 188 ELECTRIC TRANSMISSION OF ENERGY. c being the only unknown quantity. Kesolving the equation we find / c = - 9 / +-;^ (5^ + y) + V y 33). It will be seen that the quadratic equation has two roots or values for the one being positive the other negative. The latter implies that the current travels in the opposite direction, in which case the motor would become the generator and vice versa. This does not concern us here, as it applies to cases where the receiving machine is larger than the generating dynamo, an arrangement which no practical engineer would employ. We have, therefore, only to deal with the positive root, viz.. c - 9 V 9^ + ^{9 + y) + y 9 34 ). Having thus determined c, we find the counter-electro- motive force of the motor. e — E -Re 35). To obtain a maximum of commercial efficiency the motor must be so speeded that its counter-electro-motive force attains the value E — Rc. By using the equation d y\ d e = 0, we can also obtain di- rectly the most favourable counter-electro-motive force. The solution gives again two values for e, one smaller than E, the other larger than E, The latter corresponds to the case when motor and generator have changed places and need not be further considered for reasons above stated. The former value for e is alone of impor- tance ; it is given by the formula. e = E Rg-^{E RgY - {E -{■ Rg) {E - Ry ) . 36). COMMERCIAL EFFICIENCY. 189 This equation does not clearly show that e must in all cases be smaller than but on developing the expres- sion under the square root on the rights we also obtain, e = E-\- Rg- ^ g y E R [g y) . 37). The same expression is obtained by inserting, E -e into formula 34). It is evident that the square root in 37) must under all circumstances be numerically greater than R g^ and therefore e must under all circumstances be smaller than E. Now, according to the orthodox theory found in text books, maximum efficiency is obtained for E ^ e. This could only be if g '=^ o and y = o ; that is to say, if the dynamo would absorb no energy whatever when working an open circuit and if the motor could be kept running idle without the expenditure of electrical energy. Both these conditions are evidently absurd. Since the formulas 32) to 37) have a somewhat com- plicated appearance, it might be as well to elucidate their application by a practical example. We will assume that in a given system of transmission the generator can be worked at the safe limit of 1,000 volts and 20 amperes, and under these conditions has a commercial efficiency of 80®/o. Its internal resistance is 5 ohms. Its external electro-motive force at maximum output would there- fore be 1,000 — 20 X 5 = 900 volts. To produce 900 volts and 20 amperes with a machine having 80% efficiency, requires the expenditure of 18,000 X = 22,500 watts. Of this amount 20,000 190 ELECTRIC TRANSMISSION OF ENERGY. watts represents the internal electrical energy developed in the armature, and 2,500 watts represents the energy necessary to overcome the mechanical and magnetic friction of the dynamo. At 1,000 volts this energy is represented by a current of 2*5 amperes. A similar cal- culation applied to the motor gives, say, 1*5 amperes. W e have, therefore, ^ = 2-5 7 = 1-5. Let us assume the distance between generator and motor to be one mile, and the circuit to consist of two miles of copper wire *134 inch in diameter. At 98 per cent, con- ductivity the resistance of the line would therefore be 6*2 ohms. Allowing 3 ohms for the resistance of the motor, we have R = 14*2. These are all the data necessary for solving the problem as to current and counter-electro-motive force for maxi- mum efficiency. Equation 34) gives immediately c = 14*5 amperes, and 35) or 36) gives e = 790 volts. The maximum commercial efficiency attainable under these conditions is from equation 32) _ 14-5 - 1-5 790 ~ 14-5 + 2*5 1,000 60 per cent. Assuming then that the generator be kept running at such a speed, that its electro-motive force is kept at the safe limit of 1,000 volts, we must, in order to obtain the maximum possible return of 60 per cent, of the power expended, so gear and speed the motor that it will oppose an electro-motive force of 7 90 volts to the current. The PRACTICAL EXAMPLE. 191 strength of the latter will then be 14*5 amperes, and the energy actually given out by the motor will be 7QO (14*5 — 1*5) = 13*8 horse-power. 746 ^ To show how a departure from these conditions affects the efficiency and power developed, the following table is added : Counter-Electro- motive Force, Current. Commercial Effi- ciency, per cent. Power obtained from motor, H.P. 790 14-5 60 14 876 8 54 7-7 716 20 58-6 18 A fflance at this table will show that for currents either larger or smaller than 14*5 amperes, the efficiency is less than 60 per cent., but that the falling off is limited to a few per cent., whereas the power transmitted may vary considerably. This is a very valuable property of electric transmission of energy, as it allows a variation of power between wide limits, without serious sacrifice of efficiency, and thus renders the system very elastic. The great importance of this point will become apparent when we compare electric with hydraulic transmission. In the latter the motor consumes always the same quantity of water, whatever work it be doing ; and since the pressure is constant, the efficiency falls very low if the motor is working under its normal power. To remedy this, Mr. Hastie has introduced a water-motor with variable crank- radius, the latter being automatically adjusted to the work done by a spring. A contrivance of this nature, although extremely ingenious, adds considerably to the cost and complication of the machine and represents an additional chance of break-down. On the other hand. 192 ELECTRIC TRANSMISSION OF ENERGY. electricity can be used without any separate contrivance for regulation, and has thus a great advantage over hydraulic transmission. The system of transmitting energy by means of two series-wound dynamos has the other advantage of being almost perfectly self-regulating as regards the speed of the motor. It has been shown how a motor intended to be worked by a constant current can be made self-regu- lating, that is, can be arranged to run always at the same predetermined speed, whatever load may be thrown on it. It has also been shown how motors can be made self- regulating, if supplied with current at constant pressure. In the first case, the electro-motive force must increase as the load increases ; and, in the second place, the cur- rent must increase as the load increases, one or the other being kept automatically constant at the generating station. But with a series-wound dynamo, neither the current nor the electro-motive force are constant, but vary in a certain dependence on each other. It might thus, at first sight, seem as if the problem of making the motor self-regulating were thereby rendered very much more difficult. This is not the case. The evil, if we may so regard it, in the dynamo becomes of itself the remedy in the motor. Let, in Fig. 68, O E represent the ordinary charac- teristic of the series-wound generator, the curve being drawn for a constant speed of, say, 1,000 revolutions a minute. Let O e represent the characteristic of the motor also for the speed of, say, 1,000 revolutions. The counter-electro-motive force developed in the armature of the motor at that speed is therefore represented by the ordinates of the curve O e. Thus to a current O C will be opposed an electro-motive force C B, to sl current TWO SERIES MACHINES. 193 O (7j will be opposed an electro-motive force Ci and so on. In the dynamo the electro-motive force corre- sponding to the current O C is C D, and that correspond- ing to the current O C\ is D^. Draw O R under such an inclination to the horizontal that the tangent of the angle R O X represents to the scale of the diagram the numerical value of the sum of the resistances {R r g) of dynamo, motor, and line, then the electro-motive force lost in overcoming these resistances is for the current O (7, evidently C A, for the current O Ci, Fig. 68. and so on. The ordinates between the straight line O R and the characteristic curve O E represent, therefore, the counter-electro-motive forces which must be developed in the armature of the motor at various currents. If the current is O (7, the counter-electro-rnotive force is A Z>, if the current is O the counter-electro-motive force is A^ Z>i, and so on. Now the counter-electro-motive force of the motor, if running at a constant speed of 1,000 re- volutions a minute, is given by the curve O e, and it will be seen that if the ordinates of this curve are for every current equal to the ordinates contained between O R o 194 ELECTRIC TRANSMISSION OF ENERGY. and O then the motor suits perfectly the requirements of the generator^ and it will run at a constant speed. The motor will run at that speed whether the current be O Ci or O Cy provided that and C A = B D. The solution of the problem consists, therefore, in the proper choice of motor and dynamo, so that their charac- teristics fit each other as nearly as possible, as explained. Beyond this, no other precaution or apparatus is neces- sary to make the system perfectly self-regulating. Even if the characteristics should not fulfil the condition C A — B D over their entire range, it will, as a rule, not be difficult to find two points, Cy and (7, tolerably far apart, for which the condition is fulfilled, and between which the deviation of one curve from the form demanded by the other is very trifling. The system will, therefore, be practically self-regulating between these limits. Several years ago the author had occasion to practically test the soundness of this theory. He had occasion to use electric transmission of energy within the limits of an engineering works, for the purpose of supplying with power the pattern-makers’ shop, which on account of its location could not be reached by any mechanical trans- mission. The power required by the wood-working ma- chines in that shop, including band and circular saws, was, of course, very variable, and it became a matter of the greatest importance to keep the main shaft — from which the different tools were worked by belting — re- volving at a constant speed. This object was attained by the method just described. The generator was a Btirgin dynamo, driven at a constant speed from the main engine in another part of the works, and the motor was also a Biirgin dynamo, but Avound for a lower electro-motive force. There was a considerable distance between the TiFO SERIES MACHINES. 195 two characteristics O E and O Fig. 68, and to find two points, O Cl and O (7, for which the condition C A = B D should be fulfilled, it was necessary to increase the inclination of the line O R by placing a little addi- tional resistance into the circuit. This, of course, entailed some small loss of energy, but was in no way a fault of the system. It was occasioned simply by the necessity of using the two dynamos which happened to be at hand. If the machines could have been designed for this very purpose, no additional resistance would have been re- quired, and the automatic regulation would have been equally good. Within the last few years this method of obtaining constant motor speed in electric transmission of energy has been very extensively applied by Mr. C. E. L. Brown, of the Oerlikun Engineering Works, Switzerland, in the various plants established by that firm on the Con- tinent, and, by careful design of the machines, Mr. Brown has succeeded in reducing the maximum variations in the speed of the motor between running idle and fully loaded to as little as 2 per cent. Equally good work has recently been done in this direction by Herr von Dobrowolsky, and by the courtesy of this gentleman the author is able to give particulars of a transmission plant which he has seen tested in the works of the Allgemeine Elektrizitaets Gesellschaft at Berlin in the beginning of this year. The self-regula- tion under extreme variations of load was so perfect as to form the best possible proof of the correctness of the above theory, according to which a certain relation between the characteristics of the two machines insures constant speed of the motor at all loads, provided the generator is driven at constant speed. The plant in question consists of a 20 K. W. generator of the 196 ELECTRIC TRANSMISSION OF ENERGY. Edison type, working with a normal current of about 25 amperes, a motor of the same type, and a line of 1.25 ohms resistance. The generator is speeded at 1,000 and the motor at 885 revolutions per minute. The general Fig. 69. dimensions of the iron parts of the machines can be seen from Fig. 69, which is drawn to scale. The full lines refer to the generator, and the dotted lines to the motor. In the latter, the magnet cores are slightly smaller in diameter, and the yoke is also lighter. The armature SELF-REGULATING TRANSMISSION PLANT. 197 core in both machines is 11|- inches in diameter and 121- inches long, but whereas in the generator only five per cent, of the space is taken up by paper insulation, the space thus taken up in the motor amounts to twenty-five per cent. The winding of the armature is the same in both machines, and consists of 780 external conductors, resistance *657. The field winding consists of 553 turns on each limb of the magnet, both in the generator and motor. The resistance of the field coils in the generator is T53 ohms, and in the motor 1*44 ohms, the reduction in resistance being due to the smaller diameter of magnet Fig. 70. cores. In the generator there is also a shunt of 20 ohms, arranged across the terminals of the field coils. This shunt is a resistance coil made of wire doubled back on itself, so as to have no self induction. It will be obvious that by the addition of such a shunt any slight error in the design of the machines can be rectified, that is to say, a more perfect agreement between their characteristics obtained ; but according to Dobrowolsky the shunt does more than this, it increases the rapidity with which one machine corresponds to the other. If the load on the motor be suddenly diminished, the current will be as 198 ELECTRIC TRANSMISSION OF ENERGY. suddenly checked ; but very rapid variations of current through the fields are resisted by the self-induction of the magnets. There must thus always be a tendency to irregularity in the speed owing to a kind of surging of energy to and fro between the two machines, which according to the degree of disturbance will take a longer or shorter time to die out. By the adoption of the shunt, which acts as a kind of electro-magnetic damper or dash- Fig. 71. pot, the kick from the magnets is principally taken up by the shunt, and the machines are thereby able to attain more quickly their steady working condition. Fig. 70 shows diagrammatically the arrangement of circuits. As it might break down the insulation if the line were opened by an ordinary switch, the stopping of the current is effected by means of a switch S, which is con- nected with the terminals of the field coil on the gene- rator. By closing this switch the field is caused to vanish, and the current gradually stopped. A liquid STARTING DEVICE. 199 rheostat serves for starting the motor. This is shown in Fig. 72. Fig. 71. It consists of a series of vessels filled with five 200 ELECTRIC TRANSMISSION OF ENERGY, per cent, soda solution, into which dip iron electrodes A attached to a cross beam which can be raised and lowered by means of a crank, rack, and pinion. The electrodes are shaped as shown at In the vessels there are fixed strips of iron of the form shown, with knife contacts at either end of the apparatus, and when the movable electrodes are placed right down the knife contacts short circuit the electrodes. To prevent creeping the upper parts of the iron plates are painted with some heavy mineral oil. The iron is not attacked by the cur- rent. For pressures up to 200 volts, Herr von Dobrowol- sky uses in his liquid rheostat two vessels only, and for pressures up to 800 volts he uses four vessels. The curves. Fig. 72, show the results of tests with the transmission plant above described. The inclined straight line represents the loss of electro-motive force due to the resistance of the circuit, and the two curves represent dynamic and motor electro-motive force of the two machines respectively. The line at the top of the diagram shows the speed of motor actually observed at various currents. CHAPTER VII. The Line — Helation betv^een Capital Outlay and Waste of Energy — Most Economical Size of Conductor — Formula for Maximum Current — For- mula for Mean Current — Tables for Finding Most Economical Size — Heating of Conductor — Table for Kise of Temperature. Both as regards first cost and economy of working, the line forms a very important item in any extended system of electric transmission of energy. We have to consider two separate cases. The one, where energy from a central station is transmitted to and divided between a number of small working centres all grafted upon a net- work of conductors forming the main circuit, and the other, where all the energy is conveyed to a single receiving station along a pair of conductors without any ramifications. The first case would occur in a system of town supply where electricity is furnished for lighting and power purposes, and where the lamps and motors are all connected in parallel to the mains. The second is that occurring when energy from an hitherto inaccessible source is conveyed to a convenient point of application, the distance being considerable. Whatever particular form of transmission and distribution the system may have, it will be clear that the first cost of the conductors, and the annual expenditure represented by the energy wasted in heating the conductors, follow opposite laws. To economize energy it is necessary to employ leads of low resistance, and, therefore, of considerable cross-sec- 202 ELECTRIC TRANSMISSION OF ENERGY. tional area. To reduce the first cost we would, on the other hand, employ leads of small weight — that is, of small sectional area. We see that first cost and the subsequent working expenses are both governed by the area of conductor chosen, but whilst the former increases with the area, the latter decreases as the area increases, and it is evident that in each system of electric trans- mission of energy there must exist at least one particular area of conductor for which the sum of interest on its first cost, and annual cost of energy wasted, becomes a minimum. It is also evident that, notwithstanding any other consideration, this particular size of conductor must be adopted, being the cheapest in the long run. The determination of this, the most economical size of conductor is somewhat complicated, and must be made specially for each case, regard being had to the following : 1. The rate of interest to be charged on capital outlay ; 2. The cost of one horse-power-hour at the terminals of the generator ; 3. The number of hours per annum that the maximum energy is required, and the number of hours that three-fourths, one half, and one quarter this amount is required ; 4. The cost of unit weight of the conducting material ; 5. The cost of insulation ; 6. The cost of supports if an overhead line, or troughs if an underground line ; 7. The cost of labour in laying. If it be permissible to consider the capital outlay as pro- portional to the total weight of conducting material, then for a given line we have the relation p K = k a p, where K is the total cost of the line, k a constant and p the annual rate of interest. The resistance of the line is inversely proportional to the area a, and the energy wasted equals resistance multiplied by the square of the current. Let q represent the cost of one electrical horse- THOMSON'S LAW. 203 power-hour at the terminals of the dynamo^ and let t represent the number of hours per annum during which the current c is flowing — there being always the full amount of energy transmitted — then we have the annual value of energy wasted, W= - qtc^ a w being a constant. The total expenditure will be a ,^dKp dW minimum it + —n — = 0 , da da This gives K = ^ — ^ , and a ^ w q t a = c — . p k By inserting this value into the equations for K and W we find p K c V w q t k p and W = w q t k p Hence p K = or the most economical area of conductor, will be that for which the annual interest on capital outlay equals the annual cost of energy wasted. This law is commonly known as Sir William Thomson’s law, and was first published by him in a paper on The Economy of Metal Conductors of Electricity,” read before the British Asso- ciation in 1881. It should be remembered that this law in the form here given only applies to cases where the capital outlay is strictly proportional to the weight of metal contained in the conductor. In practice this is, however, seldom correct. If we have an underground cable, the cost of digging the trench and filling in again will be the same whether the cross-sectional area of the 204 ELECTRIC TRANSMISSION OF ENERGY. cable be one tenth of a square inch or one square inch ; and other items, such as insulating material, are if not quite independent of the area, at least dependent in a lesser degree than assumed in the formula. In an over- head line we may vary the thickness of the wire within fairly wide limits without having to alter the number of supports, and thus there is here also a certain portion of the capital outlay which does not depend on the area of the conductor. It would, therefore, be more correct to write K — Ko -1- ka^ where Ko represents that part of the capital outlay which is constant and independent of the area of the conductor. This addition on the right-hand side of the formula makes no alteration in the differential equation, for d Ko d a = 0 . We obtain, therefore, again. a ■=■ c ^ w q t but the value oi p K altered. p K — p Ko + w qt k p W = \/ w q t k p. The interest on capital outlay, and the annual cost ot energy wasted are now in the relation p K — p Ko + W. They are no longer equal, but the interest on capital out- lay must be greater than the annual cost of energy wasted. By writing the above equation in the form p {K- Ko) = ir, we find that the most economical area of conductor is that THOMSON'S LAW. 205 for which the annual cost. of energy wasted is equal to the annual interest on that portion of the capital outlay Avhich can be considered to be proportional to the weight of metal used. Professor George Forbes, in his Cantor Lecture, on The Distribution of Electricity,” delivered at the Society of Arts, in 1885, called that portion of the capital outlay which is proportional to the weight of metal used, ^^The Cost of Laying One Additional Ton of Copper,” and he showed that for a given rate of interest inclusive of depreciation, and a given cost of copper the most economical section of the conductor is independent of the electro-motive force and of the distance, and is pro- portional to the current. These facts can also be seen from the above formula a — c f w q t since the square root is a constant for each case, and since neither distance nor electro-motive force appear in the expression for which is simply proportional to c. Having in a given system of electric transmission settled what current is to be used, we can, by the aid of Sir William Thomson’s law, proceed to determine the most economical size of conductor. To do this we must know the annual cost of an electrical horse-power inclusive of interest and depreciation on the building, prime mover, and dynamo, we must know what is the cost of laying one additional ton of copper, and we must settle in our mind what interest and depreciation shall be charged to the line. These points will serve to determine the constants of our formulas, and then the calculation can easily be made. To avoid the labour of going through these 206 ELECTRIC TRANSMISSION OF ENERGY. figures for every special case. Professor Forbes has pre- pared and published in his Cantor Lectures, some ex- tremely useful tables which are reproduced on pages 207 and 208. Table A refers to the cross-sectional area of conductor required to carry a current of 1,000 amperes if the annual cost of one electrical horse-power varies from £5 to £20. Table B refers to the cost of laying one additional ton of copper, and interest and depreciation on it. The use of the tables will best be seen by an example. Say we have to transmit 50 amperes, the an- nual value of one-horse power is £10, and the cost of the line is £110 per ton of copper plus a constant. We shall also assume that it has been decided to charge 7^7o interest and depreciation on the line. We look in Table B horizontally along the line opposite 7J till we come to the vertical column headed £110. We find thus the figure *424. We now look in Table A horizontally along the line opposite £10 until we find again *424 or the nearest figures to it. In the present case *441 and All. The heading of the vertical column corresponding to this figure gives the area of conductor necessary for 1,000 amperes. We find thus that the conductor should be between 2*8 and 2*9 square inches — say average 2*85. But since our current is 50 and not 1,000 amperes, the 50 area of conductor will have to be - x 2*85 = *1325 square inches. If we were to adopt a larger conductor the system would be less economical, because the capital outlay would become too great, and if we were to adopt a smaller conductor the system would be less economical because the waste of energy would be too great. CORRECTION FOR VARIABLE CURRENT. 207 Table A . — Section per Thousand Amperes in Inches. 1 1*1 1*2 1*3 1*4 1-5 1-6 1-7 1 1*8 1*9 2 ! £ % 5 1*628 1*356 1*147 *980 *857 *746 •658 : *585 . 523 *471 *426 o 6 1*954 1*627 1*377 1-176 1*028 *895 •790 ^ *702 628 *565 *511 7 2*279 1*898 1*606 1-372 1*199 1-044 *921 *819 733 *660 1 *596 u 8 2*605 2*169 1*836 1-568 1*370 1*194 1*053 •936 838 *754 ! *682 w 9 2*930 2*441 ; 2-065 1-764 1-542 1-343 1*185 ’ 1-053 ; -942 *848 , *767 10 3*256 2-712 ' 2*295 1-960 1*713 1*492 1-316 1 1-170 1-047 *942 *852 11 2*983 2-524 2-156 1*885 1*641 1-448 1-287 1*152 1*037 •937 12 I 2-754 2*352 2*056 1-790 1-580 1 1-404 1-256 1-131 1-022 a at 13 1 2*548 2*227 1-940 1-711 1-521 1-361 1*225 1-108 14 2*398 2-089 1-843 1-638 1-466 ' 1-319 1*193 15 2-238 1-975 1-755 1-570 1*414 1-278 16 2-106 1-872 1-675 1*508 1-363 o 17 1-990 1-780 1-602 1-448 O 18 1-884 1*696 1-534 cS 19 1 1*790 1-619 <1 20 1 1-704 2*1 2*2 2*3 2*4 2*5 2-6 2-7 ' 2-8 2-9 3^0 3*1 1 3-2 ; h ' f £ 5 •386 •355 *324 *298 •276 *255 •237 ■221 •206 •192 *180 •170 i , 6 *463 •426 *389 *358 *331 *305 *284 •265 •247 •230 *216 •203 a 7 *540 *498 *454 *417 *386 *356 *331 •309 •288 •269 *252 •237 K 8 *618 *569 •518 •477 *442 •407 *378 •353 *329 •307 *288 •271 O 9 *695 •640 *583 •536 *497 •458 *426 397 •370 •346 *324 •305 10 *772 *711 •648 *596 *552 •509 •473 441 *411 •384 *360 •339 a 11 *849 *782 *712 *656 *607 *560 •520 •485 *452 •422 *396 •373 12 *926 *853 *778 *715 •602 *611 •568 529 *493 •461 •432 •407 1 13 1*004 •924 *842 •775 •718 *662 •615 •573 *534 •499 •468 •440 s 14 1*081 *995 •907 *834 •773 *713 *662 617 *575 •538 *504 •475 M-h o 15 1*158 1*066 *972 •894 •828 *764 *710 •662 *617 •576 •540 •509 03 16 1-235 1*137 1*037 •954 *883 *814 *757 •706 •658 •614 •576 •542 O "O 17 1*312 1*208 1*102 1-013 *938 *865 •804 •750 •699 •653 *612 •576 18 1*390 1*279 1*166 1-073 *994 •916 •851 •794 •740 •691 •648 •610 cS P 19 1*467 1*351 1-231 1-132 1-049 *967 •899 •838 •781 •730 *684 •644 c ^ fl <3 20 1*544 1-413 1-296 1-192 1*104 1*018 *946 882 •882 •768 *720 *678 3*3 1 3*4 3-5 3-6 3*7 3*8 3*9 4 4-5 : 1 ! 5*5 6-0 ' £ 5 *160 •150 1 o & 6 *192 •180 •170 0) 7 •224 •210 *199 •188 M 8 *256 •240 *227 *215 •203 o K 9 *288 •270 *256 *242 *229 *217 10 *320 •300 •284 •269 *254 *241 •229 .1 11 •352 •330 *312 *296 •279 *265 *252 •240 B 12 *384 *360 *341 *323 *305 *289 *275 •262 •206 a 13 *416 *390 •369 *350 •330 •313 *298 •283 •224 *181 5 14 *448 *420 •398 *379 •356 •337 *321 *305 *241 •195 *162 o 15 *480 *450 •426 *404 •381 *362 *344 *327 *258 *209 *174 *147 16 *512 •480 *454 •430 •406 *386 •366 *349 *275 *223 *186 •156 o 17 *544 •510 •483 •457 •432 •410 •389 *371 *292 *237 *199 •166 o 18 *576 •540 *511 *484 *457 *434 *412 *392 •310 *251 *209 •176 0 19 •608 •570 5*40 *511 *483 •458 *435 *414 •327 *265 *220 •186 Ci (3 <3 20 •640 •600 *568 *538 *508 •482 •458 *436 •344 *279 *232 *196 208 ELECTRIC TRANSMISSION OF ENERGY. Table B. — Cost of Laying one additional Ton of Copper. £60 £65 £70 £75 £80 £85 £90 £95 £100 £110 £120 •154 i 1 I -167 •180 •193 •206 •219 •231 •244 •257 •283 •309 •231 1 •251 •270 •289 •309 •328 •347 •366 •386 •424 •463 •308 1 •334 •360 •386 •411 •437 •462 •488 •514 •565 •617 •385 ' •418 •450 •482 •515 •546 •573 •610 •643 •707 •772 •463 •501 •540 •578 •617 •656 •694 •733 •771 •849 •926 •616 ! •668 •720 •771 •824 •875 •925 •976 1-029 1-131 1-235 •771 •835 •900 •964 1-028 1-093 1-156 1-221 1-285 1-415 1-543 boo S 2 S c3 'ti cS 13 o C to •- S ? o S A ! £130 1 £140 £150 £200 £250 £300 £350 £400 1 £450 1 5 •334 -360 •385 •514 •643 •772 •900 1-029 1157 1 •501 i -540 •579 •771 •964 1-157 1-350 1-543 1-736 1 10 •668 ^ -720 •770 1-029 ' 1-286 1-543 1-800 2-057 2-315 •835 , -900 •964 1-285 1-607 1-929 2-250 2-572 2-893 1 15 1-003 1-080 1-155 1-543 : 1-928 2-314 2-700 3-086 3-471 1 20 1-336 1-440 1-540 2-056 2-571 3-089 3-600 4-115 4-629 125 1-671 ! 1-800 1-925 2-572 3 125 3-857 4-500 5-143 5-786 W e have calculated the area of our conductor under the supposition that the maximum current of 50 amperes would be flowing during all the hours per annum that the installation is at work. In other words, we have assumed that the motor when at work should always give full power. This will, in practice, seldom be the case. Whether we want the current for propelling railway cars, or producing the electric light, or working lathes and other tools generally, or giving power for a whole mill, the amount of energy required at various times will be different. It has been shown that energy can be trans- mitted in either of three ways. First, by keeping the current constant and varying the electro-motive force of the generator in accordance with the demand for power at the receiving station. Secondly, by keeping the electro-motive force constant, and varying the current in accordance with the demand for power. Thirdly, by CORRECTION FOR VARIABLE CURRENT. 209 varying both current and electro-motive force. In the first case, where the current is constant, the above for- mula for the most economical area of conductor is at once applicable whatever may be the difference in the energy transmitted at various times of the day or year. In the two other cases, however, a correction must be applied to the formula in order that account may be taken of those hours when a reduced current is passing, and when the most economical area of conductor would be smaller than that corresponding to the full current and maximum energy transmitted. This correction must evidently be applied in this form : — We make our calculation not for the full current but for the reduced current, the reduction being the greater the greater is the number of hours during which a reduced current is passing as compared to the number of hours during which the full current is passing. At first sight it might seem as if this reduced or mean current could be determined by simply dividing the total number of ampere hours per annum by the number of hours per annum. This would, however, not be correct, for the reason that the energy wasted varies not with the current itself, but with the square of the current. Let ^1, 3^3, ^4, represent the number of hours per annum during which one quarter, one half, three quarters of the full current, and the full current is respectively passing through' the conductor ; then the total of horse-power hours wasted per annum is evidently To find that mean current, C4, which flowing during ^ + ^25 ^3 + ^4 p 210 ELECTRIC TRANSMISSION OF ENERGY. hours per annum will cause an equal waste of energy, we put t = W a ”' ’ and obtain + ^2 "b ^3 + ^4* The value of the mean current, must be used in the determination of the size of conductor in order that the annual cost of energy wasted and the interest and depre- ciation on that part of the capital outlay which is propor- tional to the weight of conductor used should be equal. To facilitate the calculation Professor Forbes gives the following table : — tj. ^•2,' tg. t,. Ratio. 0 0 0 1 1-000 0 1 0 1 0-790 1 0 2 1 0-744 1 1 1 1 0-685 2 2 1 1 0-604 4 0 0 1 0-500 The figures in the column headed Patio ” are those with which the most economical area for the maximum current must be multiplied to obtain the most economical area for a varying current. In our previous example we found that the conductor should be T325 square inches, provided that the full current of 50 amperes be always HEATING OF CONDUCTORS. 211 flowing. Say that our transmission will be at work 4,000 hours per annum, but that the full current will only flow for 1,000 hours, the remaining 3,000 hours being equally divided between quarter current, half current, and three- quarter current. In this case q = 1, = 1, 2^3 = 1 and = 1 ; and we see from the fourth line in the table that the area of our conductor must be reduced to ‘685 of the area for full current. We should, therefore, have to employ a conductor of *0907 square inches area, or say a wire *34 inches in diameter. The heat generated in the conductor by the passage of the current must be carried away in the same measure as it is generated, if the temperature of the conductor is to remain at a safe limit. An undue increase of temperature is objectionable for three reasons. In the first place it spoils the insulation, reducing the insulating power of the material and rendering it so soft as to allow the conductor to sink through it. This is a very important point, and should be guarded against with great care. It has been proposed to lay underground cables in iron troughs, com- pletely filled with a bituminous compound, in which the cables are to be imbedded. This arrangement would serve excellently well for keeping the cables dry, but it has the great drawback that any compound of that nature does not behave as a rigid body, but rather as a very thick fluid. It is w^ell known that a stick of sealing wax fas- tened by strings to a card which is suspended vertically, will in course of time bend as if it were a flexible ribbon, a nd the strings will cut through it. This process goes on even in a cold room, but is accelerated by heat. Now in our underground conductor some amount of heat is always developed, and if we trust the bituminous compound to support the cables we shall find them, after a certain time. 212 ELECTRIC TRANSMISSION OF ENERGY. at the bottom of the trough, and short circuits will be- come very probable. The second reason why an increase of temperature is objectionable is, that the resistance of the cable becomes greater, causing an additional waste of energy. The third reason is that an undue increase of temperature may cause a fire risk at the points where the cables are attached to buildings. It becomes thus a matter of great importance to de- termine beforehand what rise in temperature is to be ex- pected in each given case, and if that rise should be found to be greater than appears safe, provision must be made to increase the rate at which heat is carried off. This can generally be done by increasing the superficial area of the conductor. Say we have one circular conductor of 1 square inch area, and find that with 1,000 amperes flowing it would become too hot. Now by splitting up this conductor into 10 separate wires each one-tenth of a square inch cross-sectional area we have not altered the total amount of energy transformed into heat, but we have increased the surface exposed to the cooling action of the surrounding air in the ratio of 1 : \/ 10, and there- fore the ten thin wires can dissipate more than three times the heat, as compared with the single thick wire. Pro- fessor Forbes gives a table — reproduced on page 202 — from which it can be seen at a glance what current a wire can carry when the rise in temperature is 9° and 26® above that of the surrounding air. HEATING OF CONDUCTORS. 213 Carrying capacity and heating or Wires. 6 CQ Diameter in inches. Section in square inches. lb. copper per 100 yards. Resistance per 100 yards at 15-5o C., or 60o F. Wire heated to 9o above temp, of air. Wire heated to i 260 above | temp, of air. | B. A. Units, Legal Ohms. Current. Loss in volts per 100 yds Current. 1 Loss in Volts ^ per 100 yds. 2-26 4-0 4608- •000634 •000627 1514- 1 •998 ' 2490- 1-7.301 2-11 3-5 4032- •000724 •000716 1384- 1-042 2246* 1.783* 1-95 3-0 3456- •000845 •000835 1228- 1-079 1995* 1-847 1-78 2-5 2880- •00101 •000999 1072- 1-125 1739- 1-931 1-69 2-0 2304- •00127 •00126 913* 1-205 1482- 2-059 1-38 1-5 1728- •00163 •00161 732* 1-288 1188- 2-187 1-13 1-0000 1152- •00253 •00250 542*1 1-426 880-0 2-446 1*00 •7854 904-78 •00326 •00324 451-5 1-513 732*7 2-586; •75 •44178 508-93 •00584 •00578 293-0 1-746 475-5 2-991 •707 •39250 452-16 •00653 •00651 268-0 1-800 435*2 .3-081 •500 •19635 226-18 •01307 •01292 159-4 2-139 258-6 3-6.59 0000 •454 •1618 186-39 •01567 •01550 138*1 2-248 224-1 3*8.54 000 •425 •1419 163-47 •01786 •01766 125-1 2-323 203-0 3-977 00 •380 •1134 130-64 •02236 •02211 105-7 2*456 171-4 4-201 •354 •09842 113-37 •02605 •02576 95-1 2-546 154-3 4-.357 0 •340 •09079 104-60 •02824 •02793 89-4 2-594 145-1 4-442 1 •300 ^ -07068 81-42 •03637 •03597 74-0 2*755 120*1 4-722 2 •284 •06334 73-00 •04048 •04003 68*63 2-850 114-4 4-880* 3 •259 •05268 68-68 •04867 •04823 59-60 2-981 96-7 5-102: 4 •238 •04448 51-61 •05764 •05700 52-37 3-102 85-0 .5*315! 5 •220 •03801 43-78 •06740 •06665 46*5 3-213 75-5 5*521 6 •203 •03236 37-27 •07937 •07849 41-27 3-360 66-97 5- 7621 7 •180 •02544 29-3 •1006 •08943 34-31 3-553 55-69 6-084 8 •165 •02138 24-62 •1199 •11857 30-25 3-728 49-09 6-384 9 •148 •01720 19-81 •1492 •14643 25-28 3-878 41-03 6-630 10 •134 •014102 16-25 •1812 •17919 22-12 4-133 35-90 7-075 11 •120 •011309 13-00 •2267 •22418 18-74 4-366 30-41 7-473 12 •109 •009331 10-74 •2748 •27175 16-25 4-589 26-38 7-857 13 •095 •007088 8-16 •3617 •35769 13-09 4-866 21-25 8-3.32 14 •083 •005410 6-23 •4739 •47064 10-84 5-280 17-58 9-031 15 •072 •004071 4-68 •6298 •62281 8-723 4-646 14-14 9*652 16 •065 •003318 3-82 •7727 •76412 7*481 5-931 12*14 10-169 17 •058 •002642 3-04 •9711 *96032 6-371 6-354 10-24 10-771 18 •049 •001885 2-07 1-3598 1-3447 4-876 6-813 7-95 11-691 19 •042 1 -0013854 1-60 1-8502 1*8297 3-883 7-385 6-31 12-658 20 •035 •0009621 1 1-108 2*6650 2-6354 2-953 8-087 4-80 13-866 Legal volts 96 per cent, conductivity copper. Heating of bare copper wire, emissivity = *00025 C.G.S. units. CHAPTER VIII. Circuits for Electric Transmission — Circuits for Electric Distribution — Relative Importance of Insulation — Aerial Lines — Insulators — Attach- ment of Conductor to Insulator — Joints — Couplings — Material for Aerial Lines — Estimate for Aerial Line — Protection from Lightning — Under- ground Lines — Edison Mains — The Three- Wire System — Various Sys- tems of Underground Conduits — Lead-Covered Cables. The question whether the line should be carried over- head or be placed underground, depends on a number of local circumstances, but as a rule it will be more economi- cal and sufficiently safe to use aerial conductors for the transmission proper of energy, whereas for its distribution underground cables are preferable, and in some cases indispensable. The time is fast approaching, and in America may be said to have already arrived, when no further addition to the vast network of overhead tele- graph and telephone tvires in towns will be permitted, and it is quite certain that no exception in favour of wires containing, so to speak, a large store of potential energy, will be made. Electric Light and Power Companies have realized this state of affairs from the beginning, and where they have come forward with definite proposals for a general supply, they have always arranged for their distributing plant to be placed underground. The case is different when electric transmission over a long dis- tance, and possibly across country, is involved. Here the danger from breakage of an overhead wire can be almost entirely avoided by placing the supports at frequent in- tervals — a precaution not always possible in towns where the width of streets and places often necessitates an ex- AERIA L LINES . 215 cessively long span from one support to the other — and if a wire should break, the chances of anybody being hurt are infinitely smaller than in the crowded streets of a town. We have already seen that power can only be economically transmitted over a long distance by the employment of a high electro-motive force, and hence the proper insulation of the line becomes a matter of the utmost importance. If, in a town district supplied with current at, say, 100 volts, a small leak of a few amperes should take place — and Mr. Edison’s experience in the New York Central Station seems to show that such leaks do occasionally occur — the loss of energy, as compared to the thousands of amperes sent out from the station, is very trifling, but if an equal leak should be developed in a circuit of two or three thousand volts, it might very easily absorb all the current which the generating dynamo can pour into the cables, and no energy at all could be obtained from the motor. In an overhead line faults of insulation are not so easily developed, and if they occur, are more easily discovered and repaired than in any of the underground systems, which as yet have hardly had any prolonged trial to show their practical value for high pressure ; and for this reason it will be safe to assume that aerial conductors will be almost generally used for the transmission of electric energy at high pres- sure over long distances, and that underground conductors will generally be used for the distribution of electric energy at low pressure. Aerial Lines. The conductor is generally a naked wire or cable of copper, iron, phosphor bronze, or silicon bronze, but slightly insulated conductors are sometimes also used. The 216 ELECTRIC TRANSMISSION OF ENERGY, insulation gives some protection against short circuits, which might otherwise be caused by other wires, branches of trees, or other bodies falling across the leads, and it has also the advantage of increasing the cooling surface of the conductor, thus reducing the temperature. At first sight it might seem surprising that a wire coated with insulating material, which is necessarily also a bad conductor of heat, should become less heated than a naked wire. But such is the fact ascertained by experiments, and explained on the ground that quiescent air is the INSULATORS. 217 very worst possible conductor of heat, whereas the ma- terial of the insulation, although relatively to metal a bad conductor, is a good one relatively to air. If the wire be exposed to wind, then air, although a bad con- ductor, carries heat off very fast, because each molecule of air as it becomes heated by contact with the wire is carried away and replaced by a new and cool molecule, and in this case the insulated wire has no advantage over the naked wire. The conductor is supported on porcelain insulators in the manner of telegraph lines, but to obtain a high de- gree of insulation they should be of the double-bell type, as shown in Fig. 73, or of the type known as fluid insulators in which the inner bell is formed into a cup which is filled with oil to prevent surface leakage. The material should, when fractured, show a uniformly fine and dense grain free from pores and holes ; it must be perfectly white, and contain no cracks or other flaws. The glaze must be perfectly white, and cover the whole external and internal surface. The thread must be even and well defined, with- 218 ELECTRIC TRANSMISSION OF ENERGY, out having broken parts. The stem. Fig. 74, is cylindrical and roughened up. It is taped with yarn, served with linseed oil, and then screwed into the thread, or it may be sulphured in. To test the insulator electrically it is Fig. 75. placed upside down, the inner space is filled with acidu- lated water, and it is then immersed to near the rim in a bath of acidulated water. If the insulation is perfect, it must be impossible to pass a current from the liquid on the inside through the insulator to the liquid on the outside. Fig. 76. The wire may be attached to the insulator either on the groove at the top or at the side, the latter if there should be a bend in the line occasioning a considerable lateral strain. The method of attachment in both cases will be seen from Figs. 75 and 76, Avhere the views «, 5, c, d and a b c, represent respectively the different stages of the process. JOINTS, 219 Since wire and cables can only be obtained and carried to the place of erection in limited lengths, it is frequently necessary to make joints. A joint should not only be as strong as the wire or cable itself, but it must have an absolutely perfect contact, as otherwise the passage of Fig. 77. the current would heat and ultimately destroy it. It is also desirable to avoid the use of other metals than that of the conductor, so as to prevent electrolytic action. The use of solder is, of course, a necessity, and must be Fig. 78. exempt from this rule ; but it is not advisable to use iron couplings for a line of copper, or any other combina- tion of two different metals. With thin wires a strong joint is made, as shown in Fig. 77, which explains itself. Fig. 79. To improve the contact, the middle portion is soldered over. Fig. 78 shows another form of joint suitable for thin wires, which can easily be bent. A is one wire, Bi B the other ; the ends A and B^ are left long enough to allow of being lapped round the middle portion of the joint until they meet, and are then twisted together, as shown in Fig. 79. If the wire is too thick to allow of its being easily 220 ELECTRIC TRANSMISSION OF ENERGY, twisted into a knot, the joint shown in Fig. 80 is some- times used. The two ends of the wire are bent short at right angles, and placed side by side, so that the ends point outwards. In this position they are held by a clamp whilst being served with a layer of binding wire Fig. 80 . of the same material as the conductor. When the space between the two ends is completely filled by the binding wire it is soldered over. Cables may be joined either by careful splicing or by couplings. A very neat coupling has lately been intro- Fig. 81 . duced by Mr. Lazare Weiller; it consists of a double hollow cone (Fig. 81 ) with an opening in the middle. The end of the cable is inserted at one end, brought out at the cen- tral opening, then doubled over and pushed back again through the opening. A pull is applied to the cab.e as Fig. 81 a. if to draw it out of the coupling, and this has the effect of jamming the end of the cable tightly in the cone. The end of the second cable is treated in the same manner, and to secure perfect contact, and prevent any slipping back, melted solder is poured into the central opening. Fig. 81 a shows the coupling in section and the cables in MATERIAL FOR AERIAL LINES. 221 place. As a suitable composition for the solder, Lazare W eiller recommends two parts of block tin to one part of lead. The wire cable and the coupling are both made of silicon bronze^ and thus electrolytic action is avoided. Material for Aerial Lines. There are two requirements with regard to the material suitable for aerial lines which are to a certain extent contradictory. The specific resistance of the material should be very low, and its tensile strength very high. Now copper has of all metals which can practically be used the lowest resistance, but its breaking strain is com- paratively low, and, therefore, the supports must be placed at frequent intervals and a considerable sag must be allowed in order to prevent the wire from being overstrained. The first circumstance increases the cost of installation, and the latter is objectionable because the probability of the wire coming accidentally into contact with neighbouring objects when swayed by the wind is increased. Iron, and especially steel, offers in this respect an advantage, but it has the drawback of being but a poor conductor of electricity. The conductivity of wrought iron is only about 17 per cent, of that of pure copper, and the conductivity of cast steel is only 10 per cent of that of pure copper. If cast steel wire be used for the line, the total weight which must be sup- ported by the insulators will, therefore, be between nine and ten times as great as that of copper wire of equal conductivity, and although the supports may be farther apart, each individual support must be much stronger than would suffice for copper wire. Thus there is no saving to be obtained by the use of the stronger material. As a 222 ELECTRIC TRANSMISSION OF ENERGY. way out of this difficulty it has been proposed to use steel wire coated with electrolytically deposited copper. It was thus hoped to obtain a conductor which would combine the tensile strength of steel with the high con- ductivity of pure copper. This expectation was^ how- ever^ not realized, and the compound wire has never come largely into use. The reason is not far to seek. Roughly speaking, the best steel has about three times the tensile strength of copper, and pure copper has about nine times the conductivity of steel. Imagine now a wire composed of equal parts of copper and steel, its tensile strength, even assuming that the electrolytically deposited copper takes its fair share of the strain, will be one-half plus one-sixth, that is 66 per cent, of that of a steel wire of equal diameter, the weight being about 5 per cent, greater on account of the greater specific weight of the copper. The conductivity of the wire will be twice that of a steel wire of equal diameter, or 66 per cent, that of a copper wire of equal diameter. We have, there fore, the following relation : — Pure copper wire, breaking strain 35,900 lbs. per square inch, conductivity ..... 100 Steel wire, breaking strain 119,000 lbs. per square inch, conductivity . . . . . .10 Compound wdre, breaking strain 78,000 lbs. per square inch, conductivity . . . . .66 If we express the merit of a wire by the product of its breaking strain and conductivity, we find that the com- pound wire is only 30 per cent, better than the copper wire, and in practice this apparent gain will be, to a great extent, counterbalanced by the increased cost of manufacture. Since the invention of the compound wire MATERIAL FOR AERIAL LINES. 223 great strides have been made in the production of certain alloys which combine great tensile strength with a fairly good conductivity. The first in the series was phosphor bronze^ having a breaking strain of 100,000 lbs. per square inch, and a conductivity of 26 per cent. Lately, Mr. Lazare Weiller introduced his silicon bronze, the con- ductivity of which is 97 per cent, of that of pure copper, and the breaking strain of which is half that of the best steel. For purposes of transmission of energy and for electric lighting he also makes silicon copper wires, for which he claims a conductivity lying between that of pure copper and that of silver ; the breaking strain, how- ever, is not given.^ The following Table, taken from Herr Grief’s book, but reduced to English measure, gives weight and resistance of this wire : — ^ J. B. Grief, “ Silicium-Bronze-Leitungen.” Wein, Seidel und Sohn. 224 ELECTRIC TRANSMISSION OF ENERGY, Silicon Copper Wire, Diameter in Millimeter. ; Diameter in English Miles. Weight per Mile, lbs. Resistance per Mile, ohms. 0-30 1-2 2-2 365-00 0-40 1-6 3-9 205-00 0-50 2-0 6-2 131-00 0-60 2*4 9-0 91-20 0-70 2-8 12-2 67-00 0-80 3-1 15-8 51-20 0-90 3-5 20-0 40-50 1-00 3-9 24-7 32-80 1-20 4-7 35-6 22-70 1-25 4-9 38-6 21-00 1-50 5-9 55-6 14-60 1-75 6-9 76-0 10-72 2-00 7-9 99-0 8-20 2-25 8-9 125-0 6-50 2-50 8-9 155-0 5-25 2-75 10-8 187-0 4-34 3-00 11-8 223-0 3-65 3-25 12-8 260-0 3-12 3-50 13-8 303-0 2-68 3-75 14'8 347-0 2-35 4-00 15-8 396-0 2-04 4-25 16-8 446-0 1-82 4-50 17-8 500-0 1-62 4-75 18-8 556-0 1-45 5-00 1 19-7 620-0 1-31 The following Table gives the relation between break- ing strains in lbs. per square inch and conductivity for different materials : — MATERIAL FOR AERIAL LINES. 225 Material of Conductor. Breaking strain. Conduc- tivity. Pure Copper . 39,500 100 Phosphor Bronze . 101,000 26 Silicon Bronze, Mark A 63,200 97 „ „ Mark B 79,500 80 Swedish Hammered Iron 50,700 16-5 Swedish Bessemer Steel . 56,200 16-0 Siemens-Martin Steel 59,000 13-3 Cast Steel 133,000 10-5 The choice of material for an overhead line must depend on many local circumstances. As a general rule copper is preferable to iron or steel, and phosphor bronze, or silicon bronze, is preferable to pure copper. To show the saving in capital outlay which can be effected by the use of the latter material in comparison to iron, Herr Grief, in the book above mentioned, gives comparative estimates for a telegraph line of 1,000 kilometers (625 miles) in length. Although the line is longer than any which will probably ever be used for electric transmission of energy, and the wire is smaller than would generally be required for this purpose, these estimates have still some practical value as affording a ready means of com- paring the two materials, and for this reason they are here added. The reader can see from these Tables how esti- mates for overhead lines are made up and what propor- tion the different items bear to the total cost. It will be noticed that the cost of the wire itself is not so great an item in the total cost as to greatly influence it. Although the silicon bronze wire costs nearly twice as much as an equivalent iron wire, there is yet a saving in the total capital outlay effected by its use. There are lighter supports and a lesser number of them required ; the cost Q 226 ELECTRIC TRANSMISSION OF ENERGY. of carriage and labour is reduced, and the subsequent cost of maintenance is also less. Estimate I. — Galvanized Iron Wire, 5 millimeter diameter. Wire, 156,000 kilos., at 304/ per 1,000 kilos. . 47,424 Carriage for wire, 16/ per ton .... 2,496 Insulators, 15,000, at 160/ per 100 . . . 24,000 Fixing insulators and attaching wire, at 8/ per kilometer ....... 8,000 Poles, 25 per kilometer (5 single and 10 double), at 9*60/ each ...... 240,000 Carriage for poles, 80/ per 100 ... 20,000 Erecting poles, 160/ per 100 . . . . 40,000 Joining double poles, 1*60/ per pair . . . 16,000 Total 397,920 Estimate II. — Silicon Bronze Wire, 2 millimeters in diameter. Wire, 28,000 kilos., at 3,200/ per 1,000 kilos. . 89,600 Carriage for wire, 16/ per ton .... 448 Insulators, 12,000, at 80/ per 100 . . . 9,600 Fixing insulators and attaching wire at 3*2/ per kilometer ....... 3,200 Poles, 16 per kilometer (8 single and 4 double), at 8/ each ....... 128,000 Carriage for poles, 80/ per 100 .... 12,800 Erecting poles, 160/ per 100 .... 25,600 Joining double poles, 1*60/ per pair . . . 6,400 Total 275,648 PROTECTION FROM LIGHTNING. 227 It will be seen that the total saving in favour of the more expensive silicon bronze wire is very considerable. It should also be remembered that this wire is hardly at all affected by the atmosphere, and that it retains nearly its full market value after the equivalent iron wire has been rendered valueless by rust. Protection from Lightning. Overhead lines, whether used for electric lighting or Fig. 82 . transmission of energy, are exposed to the effects of light- ning, which may not only destroy the line, but also the dynamos or motors at either end. To protect the plant, various methods are in use, all of which are more or less modifications of the lightning protectors used in tele- graphy, and which are based on the principle that a lightning discharge can leap a small break in a circuit to earth, which to the working current is impassable. Fig. 82 228 ELECTRIC TRANSMISSION OF ENERGY, shows a lightning protector for the line. The vertical point acts as an ordinary lightning protector, so as to minimize the risk of the line being struck. Should this nevertheless happen at some place between two protectors, then the current will travel along the wire, and leap across to the horizontal point, and thus be conducted to Fig. 83.^ earth before it can reach the machinery at either end of the line. Another arrangement intended for the same purpose is that invented by Professor Thomson in con- nection with his system of arc lighting over long dis- tances. It is, of course, equally applicable to long lines ^ The Author is indebted t<> tlie Editor ot*‘‘ The Electrical Review ” for the use of this illustration and of Fig. 84. UNDERGROUND LINES. 229 used for the transmission of energy. This protector. Fig. 83, permits a discharge to earth from both the posi- tive and negative line, and it moreover automatically ruptures any short circuit of the lines which may be thus started by the lightning current. It must be remembered that the electro-motive force employed in the Thomson- Houston system is very high (up to 2,000 volts), and that if an arc is formed between the metal parts of the protector, this electro-motive force would be high enough to sustain it after the lightning stroke has passed, and thus not only damage the protector, but possibly also burn up the dynamo. To prevent this a device is em- ployed which automatically ruptures the arc. In Fig. 83 G is the dynamo working the line C ; L and are plates of metal in connection with the line, and e and are similar plates in connection with earth at jF, E^. There is a small interval between e and X, and between and Z’, which, when a lightning discharge falls on the line, is easily leaped by it. The stroke is thus carried to earth. To rupture the arc formed, a magnet ikf, Fig. 84, is employed. The plates L and E approach closely only at their lowest parts, and above are spread out as shown. The effect of the magnet, whose poles are flattened out as shown, is to repel upward the arc that may be formed between the plates L and E. The arc at once rises to the wider space and is thus broken, or so to speak, blown out magnetically. Underground Lines. A large number of systems of underground cables have been either proposed or tried, but as yet it cannot be said that anything like finality has been reached. The mains have been, and are still the most serious difficulty 230 ELECTRIC TRANSMISSION OF ENERGY. of electrical distribution. Edison was one of the first to grapple with the problem, and may be said to have found a solution which, if not perfect, at least has the merit of working. He originated the system of placing two half- round conductors into an iron pipe, the space between Fig. 84. the conductors, and between them and the pipe being filled by a bituminous compound, which was poured in when liquefied by heat. The main was made in 20- feet lengths, the copper strips protruding at each end for convenience of jointing. The joints were made by solder- ing, which proved a very troublesome operation, because UNDERGROUND LINES. 231 the thick copper strips carried the heat away almost as fast as it was applied to the joint. To minimize this trouble it would have been advantageous to employ longer tubes, but that was found impossible, as the streets of New York, like those of any large town, are so cut up by gas, water, and drain-pipes, that no straight line of any length can be obtained. A short coupling-tube was placed over each joint connecting the iron pipes. Very soon after the installation was started troubles arose. The unequal nature of the ground, and the strains arising from the heavy traffic on the streets caused the pipes to be bent or broken, the conductors were thereby strained and worked through the bituminous compound until they came in contact and formed a short circuit, and the pipes were often accidentally damaged by the tools of workmen engaged upon some gas, water, or sewer work. The light wrought-iron piping at first employed was by degrees exchanged for strong steam-piping which could not easily be broken through by a pickaxe, and greater flexi- bility was given to the whole system by replacing the rigid couplings by ball-and-socket joints which permitted the mains to follow more or less any subsidence in the ground which might take place. The copper strips before being inserted in the pipe, were each taped singly, then served spirally with cord, laid together with their flat sides, and again wound spirally with cord. This prevented their coming in contact with each other, or with the sides of the pipe, even if the insulating compound should give way. Where a bend in the main is necessary, cast-iron elbows, as shown in Fig. 85, are introduced. It is impossible to speak of underground conductors without making a digression to explain the so-called “three-wire system.” It has already been pointed out 232 ELECTRIC TRANSMISSION OF ENERGY. that electric distribution, to be perfect, must insure the independence of one motor from the other, and this can best be done by keeping the pressure in the mains con- stant, however the demand for current may shift and vary in different parts of the system, and at different times. Now, if we have a single pair of cables, one of them con- nected to the positive and the other to the negative terminal of the generator, it will be clear that on account of the resistance of these cables the current can only arrive at the motor after having undergone some loss of pressure, and this loss will be the greater the greater the current, or in other words, the more motors are at work Fig. 85. on the same main at the same time. As far as motors only are concerned the slight difference in electro-motive force thus occasioned would not be of any serious conse- quence, but as electric lamps must usually be fed from the same mains, it becomes very important to keep the pressure as nearly constant as possible. This can be attained by using a conductor of large size in comparison with the maximum current it has to carry. But we have already seen that the area of the conductor is determined by economical considerations, and any undue increase of the weight of copper laid down in the mains would render their cost prohibitive. The absolute variation in the pressure, both as regards distance from the generator and time, is thus a fixed quantity, and all we can do is to THE THREE-WIRE SYSTEM. 233 lessen the relative importance of this variation by working at as high a pressure as possible. A variation of five volts up and down of a standard pressure of, say, fifty volts, is a very serious matter, involving a difference of 20 per cent, between the highest and lowest pressure ; but if we can increase the standard to 200 volts the difference will be decreased to but 5 per cent., an amount which in practice will be found tolerable. Now, glow lamps as usually made require about 100 volts, and if our electric mains are to serve both for light and power purposes we must keep a pressure of 100 volts between them. If reliable lamps of 200 volts could be obtained we would be able to reduce the weight of mains to one-fourth of what is necessary at 100 volts, but since such lamps are as yet not to be had, we must look for some expedient which will allow us to use 100 volt lamps and at the same time work our supply at 200 volts. This is done by the three- wire system patented by Dr. J. Hopkinson in 1882. In this system two dynamos are used coupled in the following way. The negative terminal of the first dynamo to the negative main ; the positive terminal of the first dynamo to the negative terminal of the second dynamo and to a main called the balancing wire ; the positive terminal of the second dynamo to the positive main. The lamps and motors are attached in as nearly as possible equal propor- tion across the negative main and the balancing wire, and across the balancing wire and the positive main. If all are at work no current will pass along the balancing wire to the dynamos, but if some of the motors or lamps on either one or the other side of the balancing wire be switched oflF, then a differential current will pass along that wire to or from the dynamo whose circuit happens for the time being to do the greater part of the work. 234 ELECTRIC TRANSMISSION OF ENERGY. Since it is extremely unlikely that all the lamps or motors to one side of the balancing wire will be switched off at the same time, this wire can be considerably smaller than any of the mains ; probably half the area will, in practice, be found sufficient. In this case the amount of copper required for the three-wire system working at 200 volts, as compared with the two-wire system working at 100 volts, is as follows : Each main carrying only half the current need only be half the weight. The balancing wire, carrying at most one-quarter the current, need only be one-quarter the weight. Therefore the total weight of copper will be I + I -f- 5 in the three-wire system, and 1 + 1 in the two- wire system, or the former saves 37 per cent, as compared to the latter. This calculation is made under the sup- position that in either case the area of conductor is de- termined according to Sir William Thomson’s law of best economy. It however frequently happens that the size of wire must be increased from that given by this law in order to limit the variations of pressure, and as the in- creased voltage makes a greater absolute variation per- missible, the saving effected by the three-wire system may be greater than here stated. The Edison Company claim that the saving is about 60 per cent. Mr. Edison has proposed to still further reduce the size of the balancing wire by providing each consumer with a switch which can be set to either one or the other main, the theory being that consumers will naturally set their switches so as to get the higher pressure, if there should be a difference due to unequal distribution, and thus mutually assist each other in getting the standard pressure, and relieve the balan- cing wire from having to carry any considerable current to or from the dynamos. As a further improvement. Professor Forbes has constructed an electro-magnetic apparatus THE THREE-WIRE SYSTEM. 235 which will automatically set the switch to one or the other main as soon as a certain difference in pressure should be exceeded. The three conductors — viz., the two mains and the balancing wire — are laid in one wrought- iron tube. Each conductor is wrapped with a layer of insulating tape and then wound spirally with a rope impregnated with some insulating compound liquefied by heat. The pitch of the spiral is large as compared with the diameter of the rope, so that the windings of the several conductors fit between each other when the conductors are placed together. The three conductors are also bound together by a final spiral winding of rope around them all and then inserted into the tube. Molten insu- Fig. 86. lating compound is poured into the tube at high tempe- rature, ready access being afforded the liquid throughout the length of the tube along the spiral paths formed by the rope winding. The joints between successive lengths of these electric tubes” are made by means of special coupling-boxes with ball-and-socket connection shown in Fig. 86. Instead of soldering the ends of the conductors immediately together, short flexible couplings are employed, consisting of a piece of cable having cast or brazed on its end suit- ably shaped composition sockets, in which are drilled holes to fit the size of copper conductor, and to which they are soldered in laying the line. Three such cables are required for an ordinary joint. A modified coupling 236 ELECTRIC TRANS3IISST0N OF ENERGY. box is used for the connection of branch circuits to the main circuits. The construction of this will be clear from the illustration, Fig. 87, and after what has been said about the ordinary coupling box need not further be explained. Another kind of junction, being essential to the system, must be here mentioned. It is the so-called ^^junction safety catch box,” designed for connecting so-called feeders ” with certain points in the network of mains. By employing feeders the same result is obtained as if the generator, which may be quite outside the district Fig. 87. supplied, were actually placed at the feeding centre ; and if we employ a sufficient number of feeders, each con- nected to its own dynamo, the pressure at the terminals of which can be varied so as to keep the pressure at each feeding centre constant, we have virtually transferred each dynamo to its feeding centre, that is, into close proximity to the points of consumption. The equaliza- tion of pressure throughout the district supplied is thereby much facilitated. At the Edison installation in New York there are twenty separate feeders. The junction safety catch box is a large round box, the top of which is flush with the surface of the street. It has a loose out- JUNCTION SAFETY CATCH BOX. 237 side cover and an inside cover which is bolted on to make a water-tight joint with the box. The central part of the box is occupied by three pole pieces corresponding to^ and connected with^ the three conductors in the feeder, these pole pieces being gun-metal rings, each having as many radial projections as there are mains connected with the box. The pole pieces, as well as all the other electrical fittings within the box to be presently described, are insulated from each other and from the box. The radial projections terminate in plane, polished, gold-plated sur- faces, one inch square, which are arranged symmetrically around the centre of the box and equi-distant from it. The terminals of the mains are arranged in three other circles, each respectively on the same level as the corre- sponding polar ring, but of larger diameter, and the safety fuses, also provided with gold-plated terminals, connect each pole piece with the terminal of the main radially opposite to it. The electric tubes containing the mains enter the box two feet below the surface of the street, and the connection between them and the box is made water-tight. If from any reason — a short circuit or a heavy leak — the current through any of the mains be- comes too strong, the safety fuse connecting that main with the feeder melts and interrupts the current until the damage has been repaired and a new fuse inserted. Another system which has already passed the first experimental stages is that of the American Sectional Underground Company. It is intended for the accommo- dation of telephone, telegraph, electric light, and electric power wires, all in the same duct, but separated from each other by shelves. The largest size duct yet made is a cast-iron pipe of rectangular section, ten inches by fifteen inches, costing £3,000 per mile when laid. The pipe is 238 ELECTRIC TRANSMISSION OF ENERGY. provided at every street-corner with a man-hole large enough for one or more men to enter for the purpose of hauling in the wires and making the necessary connec- tions. There are further, at convenient distances along the line of conduit, hand-holes for tapping the wires for house-to-house supply. The connection with the house wires and mains are made at the nearest man-hole, and the house-wires are run along an upper shelf in the duct devoted for that purpose until the hand-hole is reached, where they are taken out and across to the house. It is claimed as a special feature of this system, that through the interposition of shelves the telephone wires are guarded from induction from the electric light and power wires. On the other hand, it seems doubtful whether the insulation of heavy cables, after they have been dragged along the shelves, will remain perfect. To provide against water the man-holes are provided with sumps connected to the street drains, and open gratings are placed at cer- tain man-holes, by which means sweating and condensa- tion in the duct itself are prevented. The Brooks Underground Conduit , — The conductors are laid in wrought-iron pipes with suitable splice-boxes, hand-holes, and outlets. To protect the pipes from oxida- tion they are laid in a wooden trough, into which hot pitch is poured so as to completely envelop the pipes. The conductors are made up into bundles, soaked in hot mineral oil, and drawn into the pipes in 2,000 feet lengths. A heavy mineral oil is then forced into the conduit for the purpose of excluding moisture and increasing the insulation. To show the efficacy of this oil as a means of insulation, Mr. Brooks, at the Philadelphia Exhibi- tion, showed the following experiments. Two wires were attached to a Holz induction machine, and their extre- UNDERGROUND LINES. 239 mities dipped into the oil. They were so placed as to be ^ inch apart in the oil and inch apart at the surface. On turning the machine the spark passed through the 1^" of air-space at the surface, but not through the oil, although there the leaping distance was much smaller. In all the systems above described the leading idea is to provide for the conductor, in the first instance, an insu- lation so perfect and of such thickness that moisture cannot get to the metal of the conductor, and so cause leakage. The Continental Underground Cable Company prefer to use for their conductor a cheap and thin insulation, but they endeavour to surround it with an atmosphere of perfectly dry air under pressure. This is done by build- ing conduits with walls of asphalte blocks or other anti- moisture material, and providing them with iron supports and semicircular wrought-iron troughs for the accommo- dation of these lightly insulated mains. The mains are hauled through from one man-hole to the other by means of a cord, which has, in the first instance, been sent through the conduit by a carriage propelled by an electric motor. Very light wires can be laid direct by this carriage. It is proposed to close, as her- metically as possible, the whole of the conduit, and to force air into it which has been deprived of all moisture by being passed over some chemicals. Safety-valves are fitted at the end of the conduit where this* air may escape if the pressure rises beyond a safe limit. As far as the author is aware this system has not yet been practically applied. Lead Covered Cables . — Perhaps the most simple, and certainly a very efficient way of keeping cables dry, is that of surrounding each cable with a continous sheath 240 ELECTRIC TRANSMISSION OF ENERGY. of lead. The cable, after having received the usual insu- lation, is passed through a machine which, by hydraulic pressure, surrounds it with a cylindrical coating of lead free from any open joints, flaws, or other imperfections through which moisture might enter. The cable, thus protected, can be laid either directly into the ground or into a trough made of brickwork fllled with loose sand, and then covered over by flags or brickwork. If special protection is required a second lead sheathing is put over the first. The following Table gives the weight of single and double lead-covered cable of different cross-sectional area. ' Section in square inches. Weight in pounds per mile. Ohms per mile. Single Covering. Double Covering. •007088 1,150 1,714 6-36 •009331 1,234 1,819 4-84 •011083 1,430 2,092 4-05 •011309 1,489 2,209 3^99 •013193 1,516 2,235 3-40 •014102 1,663 2,400 319 •015280 1,795 2,613 2-94 •016960 1,817 2,653 2-65 CHAPTER IX. Possible Applications of Electric Transmission of Energy — Best Field for it is Long Distance Transmission — Comparison with other Systems — Herr Beringer’s Investigation — Hydraulic Transmission — Pneumatic Transmission — Wire-Eope Transmission — Comparative Tables of Effi- ciency and Cost — Practical Conclusions. If we would judge fairly the merits of a new invention, we should not only look upon it by itself, but compare it with all that has gone before and might be superseded by it. This is especially the case if we have to deal with a new thing that has many rivals, and the electric trans- mission of energy is precisely in this position. Ever since man began to use tools worked by other than manual power, he had to emj)loy some system of transmission of energy, and as a natural consequence the number of systems is not only very large, but each has in the course of time been brought to great perfection. Electric trans- mission has therefore to compete with a host of mechani- cal devices, and it becomes important to compare it with them. Some enthusiasts predict that in the near future all belts, pulleys, shafts, ropes, and cog-wheels will be superseded by electric wires and motors. Thus Mr. Walker, in ^‘The Electrician,” of Jan. 8, 1886, says: How easily and how quickly, and with little occasion for repairs, can two cables be laid, in almost any position, for mines, ironworks, docks, factories, as compared with shafting, ropes, steam-pipes, compressed air. I have 242 ELECTRIC TRANSMISSION OF ENERGY, every confidence that the day will come when shafting and belts in factories will be looked upon as a barbarism, and people will wonder however they endured it so long.” As a matter of fact, there are already several engineering works where electric transmission of power is largely used. Messrs. Ducommun in Mulhouse, the Foundry of Cannon in Ruelle, the Ecole Industrielle in Saint-Cha- mond, the ironworks in La Buire, the workshop of Sir David Salomons, and a private workshop of Sir W. Arm- strong, in all these instances electro-motors are used. In the works of the Societe Gramme, and in those of the Compagnie Electrique, there is not a single shaft worked by belt, all the tools being coupled direct to small electro- motors which are supplied with current from one generator driven by the main engine. In the Thomson Houston Company’s and some other electrical engineering works in America electro-motors are also largely employed for driving the different machine tools. It has been objected that the total efficiency of these installations reaches but fifty per cent. — that is, only half the power of the engine is actually obtained at the tools. This estimate is pro- bably below the mark, but even if it were correct it must be remembered that there is no necessity for heavy walls, columns, or other supports to carry overhead shafting, no attendance is required for keeping the bearings in proper order, and, above all, there is no intricate mass of belting, the maintenance of which is expensive, and which often presents a source of danger. Moreover, transmission by shafts and belting has generally a lower efficiency than fifty per cent., because of the great weight of machinery which must at all times be kept in motion, irrespective of the number of tools at work at any moment, and irrespec- tive of the load on each tool. With electric transmission, DIFFERENT SYSTEMS OF TRANSMISSION. 243 on the other hand, no power is consumed, and none is transmitted for those tools which are idle, and the power transmitted to the tools at work is always proportional to the amount of work they are doing. It is this peculiarity of electric transmission, that the power consumed is always proportional to the work performed, which renders it on the whole more economical than some other and purely mechanical devices. Whether electricity is ultimately destined to supersede shafting, pulleys and other gear now commonly used for transmission of energy over short distances is a question which only enthusiasts, or those imperfectly acquainted with the technical part of the subject, can be bold enough to answer. The practical engineer is content to go step by step, and to solve those problems which appear most promising before he attacks those less certain of success, and viewed in this light it would seem that long distance transmission offers a better field for the application of electricity than short distance transmission. The reason is not that it is easier to transmit energy electrically over longer distances, but that the difficulties of employing purely mechanical means are so great as to make com- petition easier. The difficulties of all systems of trans- mission increase with the distance, but for electricity not so much as for mechanical means, and consequently the advantages of electric transmission become more appa- rent at long distances. There are four systems of importance suitable for long • distance transmission: — The electric transmission of energy. The hydraulic transmission of energy. The pneumatic transmission of energy. The transmission of energy by wire rope. 244 ELECTRIC TRANSMISSION OF ENERGY. We do not mention steam as a means for transmitting' energy over long distances, because it has not come into extensive use except in some American towns, where, however, the economical results obtained are scarcely satisfactory enough to regard steam as a serious com- petitor. It is, moreover, unnecessary to consider this system specially, as the investigation of pneumatic trans- mission would with certain small modifications be also applicable for steam instead of air. We also exclude from our list the transmission of energy by means of coal gas, taking place daily between the distant gas-works and numerous gas-engines working from the mains in the centre of the town. In this case we transmit, strictly speaking, energy, but it is energy in a latent form, and not potential energy, as in the systems above mentioned. We also leave out of consideration the idea to transmit energy by means of a revolving shaft extending over the whole distance, as something quite impracticable. Even if the cost of such a plant, and the difficulty of providing suitable bearings would not be the formidable obstacles they are, the system would yet be quite unsuitable for long distances, because the friction of the long shaft in its journals would absorb too much power. A simple calcula- tion shows that with bearings perfectly in line, and good lubrication, a wrought-iron shaft of uniform thickness and two miles in length, cannot be turned from one end because the resistance of friction is greater than the tor- sional moment which can safely be applied. If the shaft be one mile in length, fifty-five per cent, of the power which can safely be applied, is required to overcome its own friction, leaving only forty-five per cent, to be re- covered at the distant end, whereas in a shaft 100 feet in length only one per cent, is wasted in friction. These COST OF POWER. 245 figures show that transmission of energy by shafting is only economical when applied to short distances, and for our present purpose it need therefore not be further considered. Keturning now to the four systems which can be fairly considered to be competitors for long distance trans- mission, the choice of one or the other of them must to a great extent depend upon local circumstances. If the latter are equally favourable to all the four systems, then the factors which will determine the choice are : The power to be transmitted, the number of hours per annum during which the plant is at work, the price of one-horse poAver hour at the generating station, and its commercial value at the receiving station, and finally the distance of transmission. Herr Beringer, in his interesting work ^ on this subject, has made the attempt to compare in a general way the electric transmission of energy with the other three systems above mentioned. In making the com- parison it is, of course, necessary to start with certain assumptions which should, as nearly as possible, represent the average conditions of actual practice. Thus Herr Beringer assumes as a basis for his calculations two sources of energy, steam and water. As regards the former he adopts Grove’s valuation of one horse-power hour, which costs in Small steam-engines . . . 3*80 pence. Medium size steam-engines . 2*63 ,, Large steam-engines . . 1*02 ,, If the engine is at work for 300 days per annum, and for ten hours during each day, the annual cost of one ^ Beringer, Kritische Vergleichung der Elektrischen Kraftiibertrag- ung.” Springer, Berlin, 1883. 246 ELECTRIC TRANSMISSION OF ENERGY. horse-power when using large engines w^ould, therefore, be £12 14^. As regards water-power, Herr Beringer adopts Meissner’s estimate, according to which water- power costs one-fifth to one-tenth of steam-power. Under the above conditions he fixes the price of one annual horse-power obtained by water at £2 16s. A similar calculation for gas-engines, assuming that they require 30 cubic feet of gas per horse-power hour, and that gas costs 3s. 6d. per 1,000 cubic feet, brings the price of one horse-power hour up to 2*7 pence. This is inclusive of interest and dejireciation and of attendance. Since both gas-engines and small steam-engines can be erected in almost any locality, it would obviously be useless to transmit the power of these prime movers to any dis- tance. A system of transmission wdll only pay if the cost of the power received at the distant end is less than the price which would have to be paid for its production there, and consequently we need only take those cases into consideration Avhere large steam-engines or water- wheels, producing power at a cheap rate, are employed at the generating station. It has already been pointed out that the theoretical economy of electric transmission increases with the pres- sure employed. On the other hand, the difficulties of producing suitable machines, of maintaining the insula- tion of the line, and the risk entailed in a failure of insu- lation are all greater with a system where a very high pressure is used, and for these reasons it seems advisable to fix the pressure at a moderate limit. Herr Beringer assumes 1,500 volts as a limit sufficiently high to insure economical transmission, and yet not too high for safe working. He estimates the cost of electric transmission when 5, 10, 50, and 100 horse-power are required at the COST OF POWER, 247 receiving-station, and in all cases for distances of 100, 500, 1,000, 5,000, 10,000, and 20,000 meters distance. The prices taken for dynamos and motors are rather higher than the present market prices, and thus his estimates are slightly less favourable to electricity than they need be. Thus a dynamo to give 8 electrical horse-power output is valued at £200, whereas such a machine of approved construction can now be had in the open market for about £100. The comparison between electric transmission and mechanical systems is, therefore, less advantageous to the former than it would be if present market prices had been taken for the basis of these estimates ; but the fault is one in the right direction, and shows that Herr Beringer was not biassed in favour of the electrical system. The con- ductor is, in all cases, supposed to be bare copper wire carried overhead on poles and insulators, and separate wires for the out-and-home circuits are used. The diameter of the wire is calculated on Sir W. Thomson’s rule for greatest economy. In all twenty-four estimates were made, and the result is given in the following table, the figures in which represent the capital outlay in pounds sterling per horse-power obtained at the receiving station. The cost of the prime mover and that of buildings, boilers, chimney and hydraulic works are not included, as account is taken of these items in the estimate of the annual value of one horse-power as produced by the prime mover. The cost of foundations for dynamos and motors, that of mea- suring instruments and switches, are, however, taken into account : — 248 ELECTRIC TRANSMISSION OF ENERGY, Maximum Horse- Power Trans- mitted. Capital Outlay reduced to One Horse-Power transmitted over a distance of 100 m. 500 m. 1,000 m. 5,000 m. 10,000 m. 20,000 m. 5 75 78 81 108 142 210 10 52 54 56. 77 103 154 50 40 41 42 55 69 100 100 32 33 35 45 59 87 In estimating the working expenses the author allows 14 per cent, of the capital outlay for interest and depre- ciation, and he assumes that the commercial efficiency of electric transmission over the distances of 100 500 1,000 5,000 10,000 20,000 metres is -69 *68 -66 *60 .51 *32 For each horse-power obtained at the receiving station there must consequently be produced at the generating station — 1-45 1-47 1-51 1*67 1*96 3*12 horse-power. Allowing 1*02 pence as the price to be paid per horse- power hour at the generating station, the prime mover being supposed to be a large steam engine, the cost of one horse-power hour at the receiving station is 1-48 1*50 1*54 1-70 2*00 3*18 pence, which, added to the figures representing interest and de- preciation, gives the following values for one horse-power hour obtained at the receiving station. HYDRAULIC TRANSMISSION. 249 Energy obtained by large Steam Engine. Maximum Horse- power Trans- Price of One Horse- Power Hour in Pence transmitted over a distance of mitted. 100 m. 500 m. 1,000 m. 5,000 m. 10,000 m. 20,000 m. 5 2-25 2-33 2-41 2-87 3-29 5-20 10 1-98 2-07 2-14 2-53 3-10 4-85 50 1-87 1-94 1-99 2-28 2-74 4-25 100 1-79 1-85 1-91 2-18 2-63 4-08 In compiling this table it has been assumed that the plant is at work for 3^000 hours per annum. If water-power has to be transmitted, it will in most cases be advantageous to keep the plant at work night and day, because, in so doing, a maximum of horse-power hours is obtained from a given plant. Under these con- ditions the price of one horse-power hour in pence is as shown in the following table : — Energy obtained by Water Motor. Maximum Horse- Power Trans- mitted. Price of One Horse- Power Hour in Pence transmitted over a distance of 100 m. 500 m. 1,000 m. 5,000 m. 10,000 m. 20,000 m. 5 •35 •36 •37 •44 •52 •84 10 •27 •28 •29 •36 •47 •71 50 •23 •24 •26 •29 •37 •55 100 •20 •22 •23 •26 •32 •50 The hydraulic transmission of energy presents many points of resemblance with the electric transmission. We have at the generating station a force pump which delivers water under high pressure into a pipe leading to the 250 ELECTRIC TRANSMISSION OF ENERGY. receiving station, where part of the energy is recovered by a water motor. To minimize the loss of energy due to the friction of the water against the pipe, the velocity of flow should be as small as possible, or, in other words, the diameter of the pipe should be as large as possible. In thus trying to increase the economy of the system, we incur a larger capital outlay ; and applying Sir William Thomson’s rule also to this case, we And that there exists for every given set of conditions, one particular diameter of pipe for which the sum of the annual value of energy wasted, and the annual interest on capital outlay becomes a minimum. So far the analogy with the electrical system is perfect. But there enters another element into the calculation which to a certain extent modifies the law. It has been shown that the most economical size of a copper wire depends only on the current, but not directly on the pressure. It depends on the pressure in- directly, inasmuch as with an increased pressure a given amount of energy can be transmitted by using a reduced current, and this again involves a reduction in the size of the conductor. But if the current be a fixed quantity, an increase of pressure does not entail a greater outlay for conducting material. It might very slightly increase the cost of the installation by necessitating a more careful insulation ; but the difference in expenditure for a good and a perfect insulation is not a great item. With hydraulic transmission the case is different. An increase of pressure does entail a greater outlay for conducting material, because the thickness of metal in the pipe must be increased as the pressure is increased ; therefore the cost of the conductor depends not only on the quantity of water to be transmitted, but also on the pressure. A natural limit is thus set to the increase of pressure by PNEUMA TIC TRANSMISSION. 251 financial considerations which are absent in the case of electric transmission. The pressure is also limited by technical considerations. In the first place the water motor is an engine with bearings and other moving parts, which can only work satisfactorily, and without heating or undue wear, if the pressure between the moving sur- faces in contact remains within reasonable limits ; the same as in any other engine. In the second place, if the pres- sure be increased beyond a certain limit depending on the tensile strength of the material of the pipe or working cylinder, no increase in the thickness of metal can save these parts from bursting, as every engineer knows. Such a limit to electric pressure does not exist in dynamo machines or electro-motors. It is perfectly conceivable to employ any desired j^ressure provided we increase the thickness of insulation sufficiently. Another point where there is a vital difference between electric and hydraulic transmission of energy is that the electro-motor has nearly the same commercial efficiency, whether fully loaded or not, whereas the water-motor when working light has a very much smaller efficiency than when fully loaded. Hydraulic transmission can, therefore, only be employed w^here the question of effi- ciency is of secondary importance, or where the amount of energy to be transmitted is not subject to variations. All the above remarks, save the last, apply also to pneumatic transmission. But since the friction of com- pressed air against the walls of the pipe is less than that of water, the pneumatic system can be used over greater distances than the hydraulic system. It is also possible to obtain a higher efficiency with varying loads by pro- viding the receiving machine with a variable expansion gear. There is, however, another drawback peculiar to 252 ELECTRIC TRANSMISSION OF ENERGY. the employment of an elastic fluid. It is well known that heat is generated in compressing air. To prevent the air- pump from becoming too hot, the air in the act of being compressed must at the same time be cooled, which is done either by surrounding the compressing cylinder with a water-jacket, and also by circulating water through the interior of the compressing piston, or by injecting a spray of water at each stroke of the piston. The latter is by far the more effective plan. It has been adopted by M. Colladon in his air-compressors used in the works at the Gothard Tunnel, and is adopted in the Popp ” system of compressed air supply at Paris. The air, which is always charged more or less with moisture (irrespective of the water injected), upon expanding and performing work in the receiving engine, becomes reduced in tem- perature, and thus there is danger that snow will be deposited in the valves and passages of the engine. This danger will be the greater the more expansively the engine works — that is to say, the more economical we wish to render the system. To prevent the engine from becoming clogged by snow and ice, it is therefore neces- sary to apply heat in some form, and that is done either by injecting hot water, or by passing the air through a stove before allowing it to enter the engine. It need hardly be pointed out that the complications thus introduced at the generating station and at the receiving station, and the increased working expenses entailed, are very objection- able features of pneumatic transmission, counterbalancing to a great extent the economical advantage it has as compared to hydraulic transmission. It may here be mentioned that the efficiency of the Popp system at Paris, as determined by Professor Kennedy, is only about 50 per cent., and that only under very special circumstances. WIRE ROPE TRANSMISSION. 253 The transmission of energy hy wire rope invented in 1850 by M. Him is the most simple, and, up to reason- able distances, the most economical of all the known means of transmitting energy. The system is so gene- rally known that a detailed description need not be given in this place. Suffice it to say that the principal sources of loss of energy are, 1. Friction in the bearings of the rope pulleys ; 2. Air resistance ; 3. Stiffness of ropes. The loss occasioned through the slipping of the rope on its pulleys is so small that it may be neglected. The pulleys are placed about 100 yards apart. Greater dis- tances are sometimes used, but are avoided where pos- sible, as the sag of the rope requires too much head room, the influence of temperature in contracting and ex- panding the rope becomes too great, and the handling of the rope for renewal or repairs becomes too difficult. From data obtained with wire-rope transmissions actually in- stalled over distances varying between 100 and 1,000 meters, Herr Beringer calculated the efficiency of the system as follows : — Distance 100 500 1,000 5,000 10,000 20,000 meters. Efficiency *96 *93 *90 -60 *37 *13 The rapid falling off in efficiency for the longer dis- tances is due to the large number of intermediate stations necessary on account of the span being limited to 100 meters. The author in the book above mentioned gives a large number of estimates for transmission by water, air, and wire rope, calculated in the same manner as for electric transmission. We need in this place not follow him into all the details, but must be content to note the final results of his calculations as being more directly of 254 ELECTRIC TRANSMISSION OF ENERGY. interest to our subject. The following comparative table shows the commercial efficiency of the four rival systems for different distances of transmission : — Commercial Efficiency, Distance of Transmission. Electric.^ Hydraulic. Pneumatic. Wire Pope. 100 m. •69 •50 •55 •96 500 m. •68 •50 •55 •93 1,000 m. •66 •50 •55 •90 5,000 m. •60 •40 •50 •60 10,000 m. •51 •35 •50 •36 1 20,000 m. •32 •20 •40 •13 I It will be seen that for distances less than 5 kilometers (about three miles) transmission by wire rope is more economical than that by any other system. For distances greater than 5 kilometers the electric transmission is most economical. As regards capital outlay, the wire- rope system is also for short distances more advantageous than electric transmission, the limit being at about 3 kilometers (a little under two miles). Beyond that the electrical system is the cheapest, as will be seen from the annexed table : — ^ Since these tables were compiled great improvements in dynamos and motors have been made, and the efficiency of several long distance trans- mission plants now at work is over 75 per cent. COMPARISON OF CONDUCTORS, 255 Capital Outlay in Pounds Sterling reduced to one Horse’-Power, Maximum Horse- power trans- mitted. System of Trans- mission. Over a distance of 100 m. 500 m. 1,000 m. 5,000 m.j 10,000 m. 20,000 m. r Electric 75 78 81 108 142 210 5 J 1 Hydraulic 41 66 97 358 610 1280 1 1 Pneumatic 73 96 210 600 1090 2060 I 1 Wire Rope 6-5 31 61 305 760 1220 I r Electric 52 54 56 77 103 154 10 J 1 Hydraulic 30 45 65 220 416 806 1 Pneumatic 60 72 88 213 369 680 1 1 Wire Rope 5-1 23 47 231 460 925 j r Electric 40 41 42 55 69 100 50 -s 1 Hydraulic 16 21 30 91 170 325 1 Pneumatic 31 36 42 88 147 265 1 1 Wire Rope 1*8 7-2 14 69 136 272 j r Electric 32 33 35 45 59 87 100 1 Hydraulic 14 20 28 88 164 310 1 Pneumatic 26 30 34 67 109 192 1 1 Wire Rope 1*1 4*3 8-4 41 81 162 The table shows that for short distances the cost of ‘electric transmission is very considerable as compared to that of the other systems. The reason for this is that the prices of dynamos and motors have been rather over- estimated, as already mentioned. For long distances this is not so noticeable, as the conductor forms the more important item, and especially since an electric wire is cheaper than an equivalent hydraulic or pneumatic tube. If we compare the conductors only we find that for the transmission of 10 horse-power a copper wire of 127 mils diameter is equivalent to a water-pipe of diameter, or to an air-pipe of 3|-" diameter, or to a wire rope of diameter. The proportion between the cost of these con- ductors calculated for equal distances is as 1-4 : 34-1 : 27*8 : i 256 ELECTRIC TRANSMISSION OF ENERGY. The conductor with hydraulic transmission costs there- fore twenty-five times as much, and with pneumatic transmission it costs nearly twenty times as much as with electric transmission. These figures prove that as far as capital outlay is concerned, the electric system has the greatest advantage where the conductor is long, that is, where the energy has to be transmitted over a long distance. It would, however, not be correct to compare the four systems on this basis alone. The comparison must be made on the question of capital outlay combined with efficiency, in other words, the figure of merit for each system is the price which has to be paid for one horse- power hour at the receiving station. Price in Pence of One Horse-Power Hour obtained at the Receiving Station. COMPARATIVE MERIT. 257 Cost of Steam Power reduced at Re- ceiving Station. o 00 GO 2*63 cq 9 CM 0 & f 1 ^ ^ 1 r ^ 1 r ^ 's Water Power Transmitted over a distance of 1 10,000 m. 20,000 m. 1 I Oi O CO CG ^ .-1 O t ^ ^ O 00 CO -rt 1 O CO 00 ^ 9-^09 0 CO 05 9999 o o o »o »P o (M 1— 1 »0 ^ 05 1 1 rH CO uo oq CO CO t'- cq M 00 00 9999 1 S o o lO* 'ch 00 iO CO (M CO lO 00 CO CO 05 00 05 05 CO rfH 00 cq Tt ^ CO CO CO CO 00 cq CO cq s o o o 00 00 O CO O CO 05 t>. uo OI CO ^ (N CO cq 00 CO cq oq cq — . CO 05 r-( cq cq 500 m. CO 00 »>• 05 CO CO TjH rH GO O 00 (M CO CO -H Tl- 00 ^ cq ^ oq rH cq CO 0 cq r-H cq rH OOT 1 O 05 O rH CO tT r-H \ lO ^ 05 (N (N CO o 1 CO 50 cq 05 cq r-( cq 0 0 CO cq 00 cq 1 — ( cq 0 Steam Power Transmitted over a distance of a o o o' : cq 1 o o cq o ^ O r- 05 CO . iO 1 — < 1 00 tT 0 CO 0999 CO Tf 05 10,000 m. 1 05 O CO o cq o »p CO o 05 o 1 O C 5 lO o I »p CO CO 00 1 Tf F- < Tf f— ( ^(^99 cq CO Tt< 1 CO 50 0 CO 9 9 9 9 cq ^ CO CO 5,000 m. b- (M io m 00 »p (N i) o CO 00 00 o o o ^ »o (M 4}^ 00 0 Tt< 9 9 00 9 cq ^ CM 00 CO 1 — t 9999 cq cq cq cq a o o o 2- 41 3- 15 3*30 1-88 Tf 05 O rH 00 r- cq cq (N 05 0 00 0 9 9 9 9 ^ ^ b\ ^ ^ 00 05 cq 9909 F^ fIh cq F^H 500 m. CO CO uo CO (» 05 (N . CO cq 00 9900 r-i r-( cq fIh 05 01 0 t>- CO 0 0 'r-< h\'r-K System of Trans- mission. Electric Hydraulic Pneumatic Wire Pope Electric Hydraulic Pneumatic Wire Pope Electric Hydraulic Pneumatic Wire Pope Electric Hydraulic Pneumatic Wire Pope Maxi- mum Horse power trans- mitted lO o 0 50 100 j s 258 ELECTRIC TRANS3IISSI0N OF ENERGY. The smaller this price, the better the system. A glance at the annexed table will show that the cost of one horse- power hour increases in all systems with the distance, but with electric transmission the increase is not so rapid as with the other systems. The table also shows that up to a distance of 1,000 meters (five-eighths of a mile), wire- rope transmission is better than electric transmission, but above that limit the electrical system is better. Hy- draulic and pneumatic transmission are in some few cases better than electric transmission, but then the wire rope is again better than either, so that there does not seem to be a field for the application of the hydraulic or pneu- matic system, except in cases where the other two systems are for some local reason inadmissible, or where the water and air may be of further use after the power has been obtained from them. This, for instance, is the case with the pneumatic transmission employed in the building of tunnels. Here it is an absolute necessity to force air to the end of the workings for ventilating purposes, and pneumatic transmission is adopted in preference to any other system which would require some special ventilating plant being erected. The last column on the right in the table gives the cost of one horse-power hour in pence, obtained from a steam- engine placed at the receiving station, in which case the transmission becomes unnecessary. It is evident that it will always pay to place a steam-engine if the power from it can be had at a cheaper rate than it can be brought from a distant source. If the source be water-power transmitted electrically, then the local engine is more expensive in all cases comprised within the limits of the table ; but if the source is a steam-engine then it depends on the distance and the amount of power required, whether COMPARATIVE MERIT. 259 a local steam-engine can produce the power more cheaply or not. Say we require ten horse-power at the receiving station. We can obtain this power at the expenditure of 2s. 2\d. per hour with a small steam-engine erected there. Now if a large steam-engine, working very econo- mically could be found within a radius of say two miles from the place where we require ten horse-power, w^e might utilize this engine to drive a generating dynamo and transmit the energy electrically. The hourly cost of ten horse-power actually obtained from the electro-motor would then be about 2s., or ten per cent, less than the power obtained from a local engine. In this case it would barely pay to use electric transmission. If the distance were a little more than two miles it would certainly not pay. On the other hand, if the power required is small, say under five horse-power, then electric transmission shows a considerable advantage. Thus five horse-power produced in a small local steam-engine cost l^. 7d. per hour, whereas we might transmit the same power over a distance of two miles at a cost of 1^. 1^/., which represents a saving of 6d. per hour. Summarizing the results detailed in the table, we come to the following conclusions : — 1. It pays to transmit cheap water-power ; by wire rope if the distance is less than a mile, and electrically if the distance is a mile or more. This applies to all powers. 2. It pays to transmit cheap steam-power if the amount of energy required at the receiving station does not ex- ceed ten horse-power. If the distance is less than a mile use wire-rope transmission ; for distances of one mile and upwards, up to two or three miles, use electric transmis- sion. Beyond this limit a small local steam or gas engine is preferable. CHAPTER X. Classification of Dynamo Electric Machines — The Edison- Ho pkinson Dy- namo — The Thomson-Houston Dynamo — The Immisch Dynamo — The Laurence, Paris, and Scott Dynamo — The Manchester Dynamo — The Elwell-Parker Dynamo — The Crompton Dynamo — The Andrews Dy- namo — The Siemens Dynamo — The Goolden Dynamo — The Phoenix Dynamo — The Kapp Dynamo — The Brown Dynamo — The Victoria Dynamo — The Giilcher Dynamo. In the foregoing chapters we have dealt with the general principles of electric transmission of energy, and with the general conditions to be fulfilled by the generator and receiver, without, however, limiting the in- vestigation to any special type of dynamo machinery. It will now be necessary to confront the subject from a more practical point of view by entering in detail into the types of dynamos and motors at present in use. In so doing the author must point out that the present book is not intended to teach how dynamo machinery should be designed and practically constructed. This is a subject so vast, that its treatment could well fill two such volumes as the present one, aud will therefore not be attempted in these pages. The question is rather, how existing types of dynamos and motors can best be utilized for the electric transmission of energy, and for this purpose it suffices to give a descriptive account of those types of dynamo electric machines which have been found practi- cally successful. Both as regards generator and receiver the machines can be classified in the manner adopted DYNAMO ELECTRIC 3IACHINES. 261 in Chapter IV. according to the type of armature em- ployed. We distinguish three types of armature : — I. The Drum, where the wire is wound along the sur- face and over the ends of the core. II. The Cylinder, where the wire is wound along the surface and through the internal space of the core. III. The Disc, which only differs from the cylinder by the proportions of the core, the diameter being large in comparison to the length. It may be well to point out in this place that the drum requires less wire than the cylinder to produce the same electro-motive force, because the end connections in the former are generally shorter than the internal connections in the latter type ; but when constructed to give a high electro-motive force it has this practical defect, that at the ends the diflferent wires cross each other in many layers. This is objectionable for two reasons. In the first place, wires between which a great difference of potential exists, are brought close together, whereby the liability to short circuits is increased, and in the second place repairs are very difficult, because in order to reach any particular wire, all those coils which are wound over it must first be removed. In the cylinder and disc arma- tures, on the other hand, neighbouring wires on the outside as well as on the inside are never at a great difference of potential, and each coil can be removed and replaced without disturbing the rest of the winding. Each of the three types mentioned above can be further subdivided, according as the core has a smooth surface, or is provided with teeth projecting through the winding. It would, however, not serve any useful pur- pose to make such a classification in this place, as this is 262 ELECTRIC TRANSMISSION OF ENERGY. a subject which properly belongs to a treatise on the con- struction of dynamos, rather than one on their application for the transmission of energy. It will suffice to give here brief descriptions of some of the more important machines, without making any attempt to present a systematic and exhaustive list of all the various dynamos in practical use. The Edison Hopkinson Dynamo . — This machine is an improvement upon the original Edison machine, which was the first dynamo made of large size. The armature was of the drum type, and its core was formed by iron discs mounted upon a spindle, and held together by iron bolts. The field was produced by a compound magnet consisting of a number of long wrought-iron cylinders of comparatively small diameter, abutting at one end against the pole pieces and at the other against the yoke, formed of a massive cast-iron block. It will be clear that this arrangement was defective on account of the high magnetic resistance occasioned by the small cross-sectional area of the magnet cores, and also because the total length of magnetizing wire required for a number of small magnets was greater than that which would have sufficed for a single magnet of a cross-sectional area equal to the sum of the areas of the small magnets. Since the specific magnetic resistance of cast iron is considerably greater than that of wrought iron, the cast-iron part of a magnetic circuit should be larger than the wrought-iron part ; and if this be not the case in any particular point — as, for in- stance, at the butting joint between magnet cores and pole pieces and yoke above mentioned, — the lines of force will be throttled at that point, thus reducing the strength of the field. Another defect in the original Edison machine was that the bolts employed to hold the core of EDISON-HOPKINSON DYNAMO. 263 the armature together were not insulated from it^ and be- came, therefore, the seat of strong local currents, which created heat and absorbed energy. Dr. Hopkinson has in his design removed the various defects here enume- rated, and produced machines giving for the same size of armature about double the output as compared to the original type. The Edison Company have subsequently also adopted this improved design. One of these im- proved dynamos is illustrated in Figs. 88 and 89. The iron discs forming the core of the armature are held together by two large washers screwed on the spindle, thus doing away with the bolts used by Edison. The field is formed by one single horse-shoe only, the core being 18 inches wide by 9| inches thick, with rounded corners. Area of core 171 square inches. The armature is 10 inches in diameter, and contains 80 conductors, or 40 complete turns, each conductor consisting of 16 strand •069 wire. A stranded conductor is used in preference to a solid one, on account of the greater facility of bending and laying across the ends. The commutator contains 40 bars of hard-drawn copper insulated with mica. The resistance of the field magnet coils, which are coupled up as a shunt to the armature and external circuit, is 16 ohms, that of the armature is *009 ohms, and at a speed of 800 revolutions a minute the electro-motive force is 110 volts, and the maximum current is 300 amperes. Similar machines, but wound for 250 volts, are used both for generators and receivers at the Bessbrook and Newry Electric Tramway, where the motive power is furnished by turbines. The Thomson- Houston machine seems, on account of its high electro-motive force, particularly suitable for the transmission of energy over long distances. Fig. 91 EDISON-HOPKINSON DTNAMO, EDISON-HOPKINSON DYNAMO 266 ELECTRIC TRANS3IISSI0N OF ENERGY. shows a general view of this remarkable dynamo. The armature contains only three coils, but each of many turns wound over an ellipsoidal core consisting of an inner cast-iron shell and winding of iron wire. Since the copper coils are in planes containing the spindle, as in all drum armatures, the radial depth of the coils is greatest near the axle and least at the equator of the ellipsoid, thus bringing the external surface of the arma- ture when completed up to a true sphere. The field Fig. 90. magnet cores are short cylinders of cast iron, provided at their outer ends with external flanges for connection with Avrought-iron bars forming the yoke (see also Table of types of magnet. Fig. 43), and at their inner ends with pole pieces forming the zones of a sphere within which the armature revolves. The action of the machine will be understood by refe- rence to Fig. 90, which represents diagrammatically the armature, commutator, brushes, and pole-pieces, S N. Since diametrically opposite points of the same coil pass always before poles of opposite sign, the electro-motive Fig. 91. THOM SON - HO LISTON D YN A M O. 268 ELECTRIC TRANSMISSION OF ENERGY. forces created in those points are of opposite direction as regards a fixed point in space, but of the same direction as regards the coil itself. In considering the action of the armature, it will, therefore, suffice if we substitute for each coil one half turn of wire ; the effect will be the same in kind, though, of course, reduced in magnitude. Let A, jB, C represent the three half turns in end view. The coils themselves are wound in the following manner The first half of coil A is wound, the starting end being left near the axis and free. To this is joined the starting end of coil and the first half of it is likewise wound, forming with coil A an angle of 120 degrees. The start-' ing end of coil C is next joined to the two others, and the whole of coil C is then wound, forming with the two others an angle of 120 degrees. Coil B is next com- pleted, and finally coil Ay which finishes the winding. The three free ends of the coils are brought out through the hollow spindle, as shown in Fig. 91, and are attached to three segments of a commutator, each a little less than 120 degrees long, so as to leave an insulating air space between adjacent segments. We assume in Fig. 90 that the lines of force pass straight across from one pole piece to the other. In this case n n will be a neutral line, and no electro-motive force will be created in any of the wires whilst passing it. To- the right of that line the electro-motive force is directed towards the observer — the direction being indicated by a little cross inscribed into the circle representing the wire — and to the left of that line the electro-motive force is directed from the observer, the direction being indicated by a dot placed similarly. On each side of the neutral line there are fixed two brushes forming an angle of about 60 degrees with each other, and being in metallic THOMSON-HOUSTON DYNAMO, 269 connection as shown. The current enters the armature by the brushes on the left, and leaves it by the brushes on the right. Since the commutator segments form an arc of nearly 120 degrees, it will be seen that A is placed in contact with the lower positive brush as soon as it has passed the neutral line, whilst B only leaves the upper positive brush a moment before it reaches the neutral line. Each coil is thus in contact with one or the other set of brushes for nearly one half revolution and two coils are connected in parallel for nearly a sixth part of a revolution, the third coil being during that time in series with them. When B has passed the neutral line it be- comes connected in parallel with (7, and A is in series with them. The next sixth of a revolution brings C and A into parallel and B into series connection, and so on. It might be thought that on account of the small number of coils on the armature the current must be pulsating. This, however, is not the case, and the steadiness of the current is partly due to the fact that each coil, when it is in the position of strongest action, is coupled with the two other coils, which are in the position of weakest action, and partly to the effect of self-induction in the field magnet coils, w-hich are in series with the armature and external circuit. The magnetic inertia of the field opposes a certain passive resistance to any sudden change in the intensity of the current and acts as a steadying agent in the same manner as a heavy fly-wheel on a steam-engine tends to keep the speed uniform. Self- induction plays also an important part in the armature itself, preparing, as it were, each coil for the current which is generated in it as soon as it passes the neutral line, and yet preventing any undue amount of back flow of current through any single coil whilst the same is in a 270 ELECTRIC TRANS3IISSI0N OF ENERGY. weak part of the field. It is evident that in a symmetrical field the electro-motive forces in A and B will be equal at the moment when these wires are equidistant from the neutral line, but not in any other position. When A has advanced into a position where its rate of cutting lines of force is greater, B will have advanced into a position where its rate of cutting lines of force is less than before, and consequently the electro-motive forces in these two coils (which, as mentioned above, are in parallel connec- tion) will no longer be equal. If there were no self- induction in B the excess of electro-motive force in A would simply be used up in urging a local current through the two coils. This current would be quite useless as far as the external circuit is concerned, and the energy thus wasted would, of course, result in a reduc- tion of the available electro-motive force. In reality this is not the case. The coil B^ although of lower electro- motive force than A^ is able by its self-induction to resist for a certain time the current which A tries to force back through it. This resistance can only last a very short time, after which, figuratively speaking, B would be overpowered by A ; but the time during which the two coils are coupled parallel is also exceedingly short. In a machine running at 850 revolutions a minute it would only require the one hundred and seventieth part of a second for the wire B to move from a position where it is equivalent to A into a position where it is already dis- connected from A. Small as this interval of time may appear, it suffices for the creation of some, though not a very large, back current in B. This is an advantage, for when B has passed the neutral line it becomes coupled in parallel with C, and could, therefore, receive a strong local current from the latter coil, which at the time is THOMSON-HOUSTON DYNAMO. 271 near its position of best action. But since B has been provided by A with a downward current before passing the neutral line^ the inertia of this current and the self- induction of the coil B are sufficient to resist for a short time the tendency of C to set up a back current. By the time this resistance could be overcome coil B itself has passed into a strong part of the field and has thus become the seat of a high electro-motive force. A moment later C enters the weak part of the fields and is charged by B with an upward current, preparing it for parallel connec- tion with A on the right of the neutral line, and so on. It cannot, of course, be expected that these inter-actions take place with mathematical precision, and that the forces be balanced to a nicety, and it is, therefore, necessary to make provision by which any want of balance, manifest- ing itself in sparking at the commutator, may be rendered harmless. For this purpose an air blast is fitted to the machine, the jets of air being directed at the two points on the commutator where the forward or leading brushes touch it, and the action of the blast, which is intermittent, is so timed that a puff of wind is produced at each moment when a segment leaves the brush, thus blowing out the spark. The machine is made self-regulating for constant current by an electro-magnetic device (shown on the left in Fig. 91), which causes the angle between each set of brushes to increase when the current becomes too strong, whereby the armature is for shorter or longer periods short-circuited on itself, and its electro-motive force with- drawn from the external circuit. A detailed description of the mechanism employed will be found in an article by the author, published in ^^The Engineer” for August 28, 1885. 272 ELECTRIC TRANSMISSION OF ENERGY, The Immisch Dynamo^ which is used both as generator and as motor, has since its first conception undergone considerable alterations. In his earliest machines Mr. Immisch made an attempt to reduce the magnetic induc- tion between like poles in armature and field, and to in- crease it between unlike poles by recessing a portion of each pole piece. He further attempted to minimize the effect of self-induction in the armature coils whilst the current w^as commutated by a somewhat complicated arrangement of double commutator and winding, but these devices have now been abandoned, and the Immisch dynamo has thus been brought to a form very similar to that of a number of other machines which, although not possessed of any special features, have given very good results in actual practice. Machines for high voltage are made with cylinder, and those for low voltage with drum armatures, the field being in all cases of the double horse-shoe ‘^Manchester” type. In small machines the core of each magnet is a single wrought iron slab, but in large machines each core is composed of a number of slabs, partly to facilitate handling and partly to avoid heating when the machine is at work. This result is due not only to the greater cooling surface obtained by the subdivision of magnet cores, but also to the reduction of Foucault currents in the mass of the cores. The Immisch machines are largely used in collieries for pumping, and the load is in many cases extremely variable, causing rapid fiuct nations in the current. The magnetizing power (the field being series wound) is thus also liable to vary very rapidly, producing considerable and abrupt variations in the strength of the field, which give rise to Foucault currents in the mass of the magnets. By sub- dividing the latter, these currents are brought within safe T Fig. 92. — Immisch Dynamo. 274 ELECTRIC TRANSMISSION OF ENERGY. limits. The illustrations, Figs. 92 and 93, show the type of machine usually employed in colliery pumping work. In external appearance there is hardly any difference be- Fig. 93. tween generator and motor, though the arc embraced by the pole pieces in the latter is somewhat smaller than in the former, and there is less iron in and less exciting wire IMMISCH DYNAMO. 275 on the magnets. The same armature is used in both machines. The author is indebted to Mr. Albion Snell, engineer to Messrs. Immisch & Co., Limited, for the fol- lowing particulars of a plant erected at St. John’s Colliery, Normanton. Generator, 480 revolutions per minute, 690 volts at terminals, average current, 59 amperes. Sectional area of magnet cores on each side, 1 60 square inches ; sec- tional area of armature, one side, 54 square inches. Arma- ture, 24 inches diameter, 16 inches long ; radial depth of discs, 4^ inches. The armature is gramme wound with 760 turns of No. 9 S. W. G. wire, there being two layers on the outer surface. The resistance is *36 ohm. The field winding consists of 6 layers No. 4 S. W. G. wire, there being 984 turns on each limb, the two limbs being in parallel. With the average current of 59 amperes, the existing power is, therefore, about 29,000 ampere turns. The resistance of the field is *25 ohm. The arc embraced by each pole piece is 150^^ in the centre and 140"^ at the ends of the armature, and the lead of the brushes is about 20 The weight of the complete machine is about 5i tons. In the motor, the field magnet cores have a cross sectional area of 115 square inches, and the exciting coils contain only 5 layers of No. 4 S. W. G. wire, making 840 turns per limb, or an exciting power of about 24,000 ampere turns. The arc embraced by each pole piece is 130^ in the centre and 120° at the ends of the armature, and the backward lead of the brushes is about 15°. It will be seen from these figures that the induction through the iron in the armature is in the generator 18*4, and in the motor 16*8 lines per square inch, or in C. G. S. mea- sure 17,100 and 15,600 respectively. For small motors, and especially where saving of weight is important, Mr. Immisch employs an ordinary double 276 ELECTRIC TRANSMISSION OF ENERGY. horse-shoe field of the type shown in Fig. 30. The tram- way motor of Fig. 94 is provided with such a field. It Fig. 94. weighs 800 lbs. and can develop 10 horse-power when running at a speed of 600 revolutions per minute. The Laurence, Paris, and Scott Dynamo. Two types of LAURENCE^ PARIS AND SCOTT DYNAMO, 277 this machine are made, one as shown in Fig. 95 for smaller, and one as shown in Fig. 96 for larger sizes. Fig. 97 illustrates the arran«:ement of field mamets and arma- o O ture in the latter or double horse-shoe type. The magnets are of cast iron and very massive, whilst the armature is a Pacinotti drum with very deep and narrow teeth. The armature discs are provided with hexagonal holes, which 278 ELECTRIC TRANSMISSION OF ENERGY. fit the shaft which is of corresponding section, and thus the armature is securely driven. Messrs. Laurence, P aris. I and Scott, Limited, have also built so called motor gene- rators on these lines. These machines are intended for the conversion of a small high pressure current into CONTINUOUS CURRENT TRANSFORMER, 279 a low pressure current of correspondingly greater strength^ and are provided with an armature having a double winding and two commutators, one at each end. The fine wire winding, when traversed by the high pressure current, causes the armature to rotate, and, in consequence of this movement, a low pressure current is generated in the thick wire winding. The two windings are arranged to occupy alternate grooves in the armature, and thus the self induction in one circuit neutralizes the self induction in the other ; the current collection on both commutators can, at all loads, be effected without any sparking, and the brushes need not be shifted or adjusted for different loads. Since the strains are all internal, that is to say, only between the wires of the two circuits, very little power is lost through me- chanical friction in the bearings, and a fairlj^high efficiency of conversion can be obtained. A machine of this type was shown at the Newcastle Exhibition in 1887, and was there tested by the author with very satisfactory results. The machine at the time attracted a good deal of atten- tion, and it was generally believed that similar machines would speedily come into use for long distance transmis- sion and conversion of electric energy, especially in con- nection with central electric light stations. These hopes have up to the present however remained unfulfilled, although one electric light company in London at least has adopted motor generators as an integral part of their system. This is the Chelsea Company, who intend to increase the output of their storage batteries by adding to the system motor generators worked direct from the dynamos at the distant central station. One of the greatest difficulties which will have to be overcome before these machines can be employed with safety is that of 280 ELECTRIC TRANSMISSION OF ENERGY, insulation between the two armature circuits. It is obvious that with a primary current at 1,000 or 2,000 volts the insulation of the fine wire circuit from the core of the armature and from the thick wire circuit must be ab- solutely perfect, as otherwise there would be danger in handling the low pressure circuit in the houses. Engineers who have had experience with alternate cur- rent transformers know how difficult it is to obtain an ab- solutely perfect insulation between the primary and secondary circuits, but it can be done, thanks to the fact that the coils are not subject to motion and mechanical strains, and that with a proper design a sufficient amount of insulating space can always be provided. In both these respects the rotary transformer is at a great disadvantage. The insulating space is required in a position where it m exceedingly expensive, viz., on the armature ; the coils cannot be concentrated as in an alternate current trans^ former, but must be spread over the whole surface of the armature, and thus the liability to leakage is increased ; and last but not least, the mechanical strains coming upon the wires and insulation tend to destroy the latter. These difficulties are so formidable that it would appear better not to place the two windings upon the same armature core, but employ two distinct machines, either belted together, or, better still, with their shafts coupled in line by means of an insulating coupling. Insulating space outside of a machine costs nothing, and therefore this coupling, although subjected to mechanical strains, can be perfectly insulated ; the whole motor can also be insulated from earth, so that the danger of its being injured by internal leakage is reduced, and, finally, there is perfect safety for the secondary circuit. Returning after this digression to the machines as now LAURENCE, PARIS AND SCOTT MOTOR. 281 made by Messrs. Laurence, Paris, and Scott, Limited, the- author is able, by the courtesy of the firm, to give some particulars of a generator and a motor. The armature of the machine shown in Fig. 96 has a sectional area of 29*7 square inches between bottom of grooves, and 208 active wires connected to a 52-part commutator. The total sectional area of the magnet cores is 98 square inches, and the magnets are compound wound, the main winding con- sisting of 28 turns of sheet copper on each upper limb 4^ inches wide and 25 mils thick. The shunt coils are on the lower limbs and produce an exciting power of 6,468 ampere turns, whilst the main coils at full current produce 3,360 ampere turns, total 9,828. The machine when running at 7 10 revolutions per minute gives a current of 120 amperes at 100 volts. From these figures it will be found that the induction through the armature core, assuming it to take place wholly through the solid part of the discs, is about 23 lines per square inch, or in C. G. S. measure 21,500, a remarkably high figure. It is, however, pro- bable that part of the induction takes place also across the projections, so that the density of lines in the central part of the core will be slightly less than the figure here- given. Fig. 98 shows a motor intended to work on a 100 volt circuit, and to give off H horse-power. The armature core is 3f inches diameter and 4 inches long. It contains. 7*6 square inches of iron, and the winding consists of 720 active wires 42 mils thick. The magnet section is 17*72: square inches cast iron, and the shunt exciting coils con- tain 4,960 turns of 25 mils wire ; resistance 140 ohms warm. To facilitate the starting, a small series coil is placed upon one limb, which can be short circuited by an iron lever placed across the pole pieces on the top, as 282 ELECTRIC TRANSMISSION OF ENERGY. shown. When the magnets are excited this lever is pulled down, but is prevented from moving by a small trigger. After the machine has attained some speed the trigger is released by hand and the lever allowed to descend, when it bridges the terminals of the series-exciting coil, thus cutting it out. The machine then works as a pure shunt motor, and is approximately self-regulating. When stopped, the lever is pulled up by a spring, and again puts the main coil in series with the armature ready for Tig. 98. the next start. The weight of the motor is 154 lbs., and its normal speed 1,300 revolutions per minute. The Manchester Dynamo . — The shape of the field magnets of this machine gives it a very compact appear- ance, and where a large external leakage-field and great weight are not objectionable, the design is very advan- tageous. It will be seen from the illustration. Fig. 99, that the magnetic circuit is of the double horse-shoe pattern, the magnetizing coils being placed over that portion of the MANCHESTER DYNAMO. 283 magnet which in other machines constitutes the yoke. The pole pieces are heavy cast-iron blocks, the lower one being provided with extensions for carrying the bearings of the armature spindle. The magnet cores are wrought- iron cylinders, and their ends are fitted tightly into ex- tensions of the pole pieces. The area of contact between Fig. 99. the cast-iron and wrought-iron portions of the magnetic circuit is about twice as large as the cross-sectional area of the magnet core, in order that the lines of force in passing from the material of greater magnetic conduc- tivity into that of lesser magnetic conductivity may not be throttled, as was the case in the original Edison dynamo. The armature core consists of a series of thin 284 ELECTRIC TRANSMISSION OF ENERGY. wrought-iron discs insulated from each other at their outer periphery, and supported on the spindle by metal arms in a positive mechanical manner. The wire coils are, however, held on the core by friction only, which is increased by the presence of the usual binding hoops* The electrical data of a machine intended for a current of 200 amperes at 110 volts pressure, as given by ^^The Engineer” for Aug. 7, 1885, are as follows: — Magnet cores, inches diameter ; length of magnetizing coils (which are wound on separate metal sleeves), 12^ inches ; armature core, 12 inches diameter and 12 inches long; armature conductor, 203 mils solid wire wound Gramme fashion in 120 convolutions, and connected in the usual way with a 40-part commutator ; resistance of armature, •023 ohm ; field magnets are compound-wound ; re- sistance of shunt coils on magnets, 19*36 ohms ; resistance of main coils on magnets, *012 ohm ; each magnet limb contains 1,680 turns of 65 mils shunt wire, and 42 turns of treble 203 mils main wire ; normal speed, 1,050 revo- lutions a minute. There is no provision made for venti- lating the interior of the armature core. The Elwell-Parker Dynamo . — The machine originally known under this name has during recent years been con- siderably modified. In its earliest form it had a Gramme armature, the core of which was made of iron wire, coiled on gun-metal supporting arms. The field was of the double horse-shoe pattern, in the shape of a square framework, with the armature passing through the centre between cast-iron pole pieces. In the modern machines the arma- ture is generally of the drum type, and the field a single horse-shoe for small and moderate sizes, whilst for larger sizes a system of field magnets having four poles is em- ployed. The armature core consists of iron discs sup- ELWELL-PARKER EYNAMO. 285 Fig. 100. Fig. 101. DIAGRAMS OF SMALL MACHINE. 286 ELECTRIC TRANSMISSION OF ENERGY, ported on the spindle by metal arms in the usual way, and the field magnets are of wrought iron. Figs. 100 and 101 give outline views of the type adopted for small and moderate size machines. The same design is used for motors where there is no necessity to reduce space and weight, but if lightness and small bulk are of importance, machines with a double horse-shoe field are employed. In dynamos of larger size the field magnets are so arranged as to present four poles to the armature, Fig. 102 ; diametrically opposite poles being of the same sign. In FOUR-POLE MACHINE. this design, which was first employed in the Elphinstone- Vincent machine, now long obsolete, four distinct circuits through the armature are obtained, and correspondingly four brushes are used. If equi-potential points of the armature conductor were permanently connected with each other, as has been done in the Elphinstone-Vincent and in the Victoria dynamos, two brushes would suffice. The practical advantage of four poles over two is that double the current can be obtained without increasing the density of current in the armature conductor. On the other hand, there is a slight sacrifice of electro-motive force due to the greater magnetic resistance of air space ELWELL-PAHKER DYNAMO. 287 Fig. 104. ELWELL-PARKER MOTOR. Fig. 105 . elwell-parker motor for To fore parje 2S9. To face j)age CROMPTON DYNAMO. 289 and consequent weakening of the field. In cylinder armatures the area of each pole piece (;^ b of the formula given in Chapter IV.) must evidently be the smaller the more separate pole pieces have to be placed round the armature, and, consequently, the four-pole machine must have considerably more magnetic resistance than a two- pole machine of equal size. The illustration. Fig. 103, represents one of the several dynamos of this type, which are supplying current for the Fig. 106 . Blackpool Electric Tramway. They are each wound for a current of 180 amperes, and at 350 revolutions a minute their electro-motive force is 200 volts, which is found to be sufficient for the working of the line. The field mag- nets are excited by a small separate dynamo. The motors used on the tramcars of this line are re- presented by Fig. 104, whilst Fig. 105 represents a later form of motor, used on tramcars where the current is supplied by storage batteries. The Crompton Dynamos . — The core of the Crompton u 290 ELECTRIC TRANSMISSION OF ENERGY. armature consists of a number of thin wrought-iron discs, about 25 to the inch, smooth on the outside, but provided on the inside with three or more dovetail notches placed equidistantly ; and into the grooves thus formed radial bars are fitted, as will be seen from Fig. 106, which is a cross-section through a 21,000 watt Crompton dynamo. Fig. 107 is a longitudinal section. The inner edges of the radial bars are fitted into grooves slotted out in the Fig. 107. steel spindle, the cross-section of which is triangular, so as to afford sufficient depth of grooves without weakening the central portion of the spindle. Each alternate disc is coated on both sides with insulating paint, and at stated intervals fibre distance pieces are inserted by which the core is subdivided into a number of comparatively narrow rings, the object being to afford passages for air through the body of the core to cool it. These divisions are shown in Fig. 107. The field magnets are of the double horse-shoe pattern, and consist of straight wrought-iron slabs bolted together at the yokes, and attached to a cast- CROMPTON DYNAMO. 291 iron bed plate by gun-metal chairs. In the machine here illustrated, and in fact in all machines of large size, the armature is wound, not with wire, but with square bars of Fig. 108. CROMPTON ARC LIGHT DYNAMO. copper. The electrical data of this machine^ as given by The Engineer/’ are as follows : Core of armature 12 inches diameter, 2i inches deep, and 28 inches long ; air space between core and pole pieces "47 inches ; core of 292 ELECTRIC TRANSMISSION OF ENERGY. field magnets inches thick by 24 inches wide ; con- ductor on armature 300 mils by 180 mils, w^ound over the core in 120 turns, and connected in the usual way to a 60-part commutator; resistance of armature *021 ohm. The machine is intended for a current of 200 amperes, and at 450 revolutions the electro-motive force is 110 volts. Fig. 109 shows the arrangement adopted by Messrs. Crompton for driving these dynamos direct by Willans’ high-speed engines. To get the centre of rota- tion low, a point of importance in ship lighting, the dynamo is placed horizontally as shown in these illustra- tions, but where there is sufficient over-head room a single horse-shoe vertical arrangement is also sometimes used. For low voltage the armatures of Mr. Crompton’s latest machines are drum wound. The belt-driven ma- chines are made vertical, as shown in Fig. 108, and for high voltage the armature is Gramme wound, and of large diameter as compared to the size of the field magnets. The Andrews Dynamo is remarkable for a peculiar method of connecting up the coils. It is a four-pole ma- chine with two brushes only, but diametrically opposite, and therefore equi-potential points of the conductor are not connected in the usual way so as to split up the wind- ing into four parallel circuits from which double the cur- rent can be obtained. In this case the connections are made in such a way as to obtain the same current, but double the electro-motive force. The difference will best be understood by reference to Fig. 110, which shows both systems side by side, that on the left being the coupling for quantity, that on the right the coupling for tension. In the latter system an uneven number of coils must be employed, generally fifty-nine, but for clearness of illus- tration only eleven are shown. One end of each coil is Fig. 109. CROMPTON DYNAMO AND WILLANS ENGINE. To face j>age 292. ANDREWS DYNAMO. 293 connected to its commutator-plate and the other to the wire connecting the opposite coil with its commutator- plate. Thus the front end of 1 is connected to the back end of 2 and to plate 2 of the commutator, the front end of 2 is connected to the back end of 3 and to plate 3, and SO on, the last connection being the front end of 1 1 to the back end of 1 and to plate 1. The current entering the armature at the negative brush, where it touches plate 6, splits into two circuits, one going round coil 6, up on the outside of the armature, the other round coil 5, down on Fig. 110. successively up in 7, 8, and 9, leaving the armature at plate 10 by the positive brush, whilst the latter goes suc- cessively down in 5, 4, 3, 2, 1 and 11, leaving the ar- mature also at plate 10. If there were 59 coils the current would go similarly up in 28, and down in 31 coils, and by following the direction of the current in the diagram it will be seen that it is the same as the electro- motive force induced in each coil — in other words, that the electro-motive force created by one pair of poles is added to that created by the other pair. The Siemens Dynamo . — The machines formerly made 294 ELECTRIC TRANSMISSION OF ENERGY. by Messrs. Siemens had field magnets of the type shown in Fig. 29, and placed either vertically or horizontally. A large number of these machines are still in use. The magnets are formed of several bars arched in the centre to form the polar cavity, and the armature is drum wound. Some three years ago Messrs. Siemens and Fig. 111. SIEMENS DYNAMO. Halske, in Berlin, made a series of experiments with dif- ferent forms of field magnets, and adopted finally the single horse-shoe type, in which the magnets were either cast iron in one with the bed plate which formed the yoke, or of wrought iron bolted to the bed plate as shown in Fig. 111. The Goolden Dynamo . — This machine also has passed through several stages of development. Originally GOOLBEN DYNAMO, 295 Messrs. Goolden made an improved form of Gramme machine^ then followed several forms of single horse-shoe machines with only one exciting coil (one of these designs being still retained for very small motors and generators), and finally they adopted the upright single horse-shoe machine with two exciting coils, both for belt and direct driving. For low voltage machines the armatures are drum wound, and for high voltage machines they are GOOLDEN DYNAMO. Gramme wound. The field magnet cores are ot wrought iron bolted to the bed plate, but the pole pieces are of cast iron held down by stout iron bolts tapped into the ends of the cores. Fig. 112 shows a machine coupled direct to a Brotherhood engine, and designed to give at 430 revs, per minute a current of 300 amperes at 115 volts terminal pressure. The armature is 15^ in. diameter and contains 180 active conductors ; its resistance is *015 ohm. The total induction is 1,430 lines or 8,600,000 C.G.S. lines. The total exciting power is about 20,000 296 ELECTRIC TRANSMISSION OF ENERGY. ampere turns. The interpolar space is *75 in. The re- sistance of the shunt coils is 15*8 and that of the series coils *008 ohm. The motors employed by Messrs. Goolden are of a similar type, and Fig. 113 shows one built for an electric launch. In this machine the magnet cores and pole pieces are both of wrought iron, and cast-iron brackets are fitted Fig. 113. GOOLDEN MOTOR. to the pole pieces by which the motor is attached to the hull of the vessel. One of the brackets is prolonged fore and aft to carry the bearings, thus making the machine entirely self-contained. A motor placed low enough in a boat to work the propeller direct is of course liable to be occasionally immersed, and if of ordinary construction would in such a case become useless. To provide against this contingency the armature is made entirely water- PHCENIX DYNAMO. 297 proof. The tags on the commutator are carried out to the full radial dimension of the armature and form a com- plete disc^ and at the other end a disc of the same dia- meter as the armature is fitted on and the whole surface of the armature is lapped over with several layers of a waterproof material, outside of which the binding hoops are put on. The junction of this waterproof coat with the end discs is also made tight by a serving of binding Fig. ]14. wire. The motor illustrated has an 8 in. armature and w^eighs cwt. It is intended to develop 5 hp. at 500 revs, per minute, with a current of 50 amperes supplied at 96 volts pressure. The total induction is 380 lines, the interpolar space ’625 in., the exciting j30wer about 15,000 ampere turns, and the resistances are, armature *20, and field (shunt wound) 14 ohms. The Phoenix Dynamo . — In the original type of machine known under this name, and made by Messrs. Paterson and Cooper, the field was formed by a double horse-shoe magnet, and the armature was a Pacinotti ring. In later 298 ELECTRIC TRANSMISSION OF ENERGY. machines the core of the armature is however made with a smooth surface, and the field is of the single horse-shoe upright type now so universally employed. Where weight need not be economized the magnets are of cast iron, and the two limbs are cast in one piece with a sole Fig. 115. PHCENIX DYNAMO. plate, A, Fig. 114, by wbicli they are fitted to the yoke proper, which, as usual, forms part of the bed plate. The large area of contact reduces the magnetic resistance of the joint to a minimum. Fig. 115 shows a general view of a machine designed for an output of 18 kilowatts at 1,020 revs, per minute. The armature core is 13 in. dia- meter, 2 in. deep, and 9 in. long. It is wound with 174 PHCENIX DYNAMO. 299' turns of 203 mils wire, and provided with a 58-part com- mutator. The section of the cast-iron magnets is 9 in. by 11 in., and the bore of the polar cavity is 13|- in. The magnet winding consists of 54 turns of double 203 mils- wire in each of the main coils, which are coupled parallel, and 1,360 turns of 65 mils wire in each of the shunt coils, which are coupled in series. At the full output of 160* amperes at 118 volts, the exciting power of the main is about 8,500, and that of the shunt about 11,500 ampere turns, total 20,000 ampere turns. The resistances are armature ’033, main *015, and shunt 29 ohms. In machines intended for very large currents Messrs. Paterson and Cooper employ a stranded cable for the armature conductor, to avoid the Foucault currents which would be generated in the conductor if this were a solid wire or bar. The first machine wound in this manner was made by the firm about six years ago, and since then other makers have also adopted stranded conductors for their armatures if intended to carry heavy currents. Where smallness of weight is desirable the magnets are made of wrought iron, and the bearings are carried in gun- metal brackets attached to the pole pieces, as shown in Fig. 116. Instead of a bed plate proper the machine is- provided with two cast-iron angle brackets by which it is bolted down. The following particulars of a machine of this type have been published in Industries,” of the 29th of July, 1887. Output 100 amperes at 250 volts pressure. Speed 700 revs, per minute. Armature core, 13|- in. external, and 8 in. internal diameter, 12 in. long, wound with 360 turns of 150 mils wire in two layers. Sectional area of armature core, allowing for insulation, 60 sq. in. Magnets 8 by 12, sectional area 96 sq. in. Bore of polar cavity 15 in. The magnets are shunt 300 ELECTRIC TRANSMISSION OF ENERGY. woundj 6ftcli ]ini}) conts^ining 3j540 turns of 62 mils wiro^ and the total shunt resistance of the two limbs, which are coupled in series, is 83 ohms. The exciting power is about 21,000 ampere turns, and the useful induction through the magnets is 11, and through the armature 17 Pig. 116 . PHGENIX DYNAMO. lines 2 ier sq. in. The total weight of the machine is 28 cwt., or ‘85 cwt. per hp. output. The Kapp Drjnamo . — The author’s machine is of the bi-polar upright type with single horse-shoe magnet, the armature being of the drum type for low, and the Gramme type for high voltage. The armature core consists of FHCENIX DYNAMO, 301 thin charcoal iron discs, provided on the inside with notches, by which they are fitted on radial bars cast to- gether with a central hub, which, as well as the bars, ex- tends the whole length of the armature, thus forming a tube and giving additional stiffness to the shaft. To pro- vide internal ventilation of the core, pairs of stouter discs are inserted at intervals, the two discs of each pair being separated by distance pieces to form free internal spaces through which the air can pass. On their outer circum- ference the stout discs are provided with a certain number of projections spaced evenly around, and over each is fitted a piece of fibre to insulate the Avinding from the projections. The winding fills the whole of the interme- diate space, and the openings in the fibre pieces allow the air to pass through. The object of employing stout discs with projecting teeth is thus to make ventilation possible, but they also serve another and very important purpose. It will be readily^seen that the whole of the useful work of a dynamo machine, be it used as generator or motor, is done by the armature conductors, which must be thus subject to a considerable mechanical strain, and it be- comes therefore important to hold the conductors in the most secure possible manner. In small machines the frictional grip of the binding hoops may be sufficient to hold the wires, but with larger machines it is not so, and therefore the author makes use of the projections or driving horns ” above described to securely hold the armature conductors in place. Fig. 117 shows one of the author’s machines as made for belt driving. The magnets are of annealed wrought iron bolted at the bottom to a yoke, which is cast in one with the bed plate, the area of cross-section in the yoke being about 40 per cent, greater than that of the magnet 302 ELECTRIC TRANSMISSION OF ENERGY. <5ores, to compensate for the greater magnetic resistance of cast as compared with wrought iron. The pole pieces being one forging with the cores are also of wrought iron, Tig. 117. KAPP DYNAMO. but at the upper and lower edges there are bolted on strips of cast iron, forming slight extensions of the polar area. The object of this arrangement is two-fold. In -the first place, the greater polar area thus obtained re- KAPF DYNAMO, 303 duces the magnetic resistance of the inter-polar space, and consequently the exciting energy required ; and in the second place, the thin edge of the strips causes a gradual shading off between the very intense field of the pole pieces proper and the neutral spaces, so that the conductors are gradually brought within the influence of the field. In machines where the field commences and ends very abruptly there is danger that the conductors, if solid, become the seat of very strong Foucault currents at the moment of entering and leaving the field, but this danger is greatly reduced, if not entirely avoided, by the use of these cast-iron pole horns, and the author has been able to run large machines with armature bars |-in. square without detecting the slightest heating from this cause. The useful induction through the armature core is from 20 to 21 lines per sq. in., and that through the magnet cores is from 12 to 14 lines per sq. in. For direct-driven machines the author adopts the arrangement shown in Fig. 118, which represents the usual type of steam dynamo made by Messrs. W. H. Allen and Co. For low voltage machines the armature is drum wound, the conductor consisting partly of bars {on the outer surface of the armature), and partly of thin semi-circular strips of copper (across the ends of the arma- ture). Each strip is provided with a tag at either end, and these tags are bent at right angles to the surface of the strips, but to the right at one end and to the left at the other. After being insulated the strips are placed spirally one behind the other over a flanged sleeve, the tags forming circular rows at the ends of the sleeve, and serving for the attachment of the bars, which receive saw-cuts for this purpose. The interior diameter of the sleeve is made sufficiently large to allow the air to enter Fig. 118. COMBINED KArP DYNAMO AND ALLEN ENGINE. BROWN DYNAMO, 305 into the interior of the armature core, and since all the connections are outside it is very easy to replace any single conductor without disturbing the rest, which is not possible with the ordinary form of drum winding. The Brown Dynamo , — The machines made by the Maschinenfabrik Oerlikon in Switzerland to the designs of their electrical engineer, Mr. C. E. L. Brown, have been extensively used for the electric transmission of energy, and as in the last chapter of this book a some- what detailed account of several of Mr. Brown’s installa- tions will be given, only a short general description of his machines is required in this place. They are of various types, according to the requirements of each particular case, but the following may be taken as fairly represen- tative examples. For a small or moderate output the field is of the Manchester ” type, and the armature is Gramme wound with the active conductors outside the core. F or very large currents the armature core is pro- vided with holes near its outer circumference, and the active conductors, in this case stout copper rods, are drawn through these holes, which are fibre-lined for purposes of insulation. The inner or idle conductors are flat bars of copper held in place and insulated by grooved wooden rings. For small direct-driven dynamos Mr. Brown adopts an ironclad” field having four poles, but pro- duced by only two exciting coils and a drum armature with multi-polar series winding. For large machines the field is of the type shown in Fig. 48, and the armature a Gramme, or in the latest machines a multi-polar drum, with an improved form of winding, rendering it not only possible but also perfectly safe to go up to 1,000, or even 2,000 volts, which pressure could not be safely carried by a drum armature wound in the usual way. X ’6 II VICTORIA DYNAMO. VICTORIA DYNAMO. 307 The Victoria Dynamo . — This machine, made by the Brush Company, is frequently used as motor in connec- tion with an ordinary Brush ” dynamo working as gene- rator. The latter machine has already been described in another volume of the present series, and no further description of it is required in this place. As regards the Victoria dynamo, a longitudinal section of which is shown in Fig. 119, its field is of the type represented by Figs. 46 and 47, and its armature consists of a wrought-iron ring, upon which is coiled No. 30 B. W. G, charcoal iron tape, the convolutions being insulated from each other by a tape of equal width of insulating paper coiled together with the iron tape. About one-seventh of the gross area of the core is occupied by this insulation. Although the iron tape is excessively thin, there is still a tendency to heat, which can only be explained by the circumstance, that on the outer periphery, where the lines of force are at right angles to the axis, they pierce the tape on its broad surface, and thus cause it to become hot. To mitigate the evil radial grooves are turned into the core from the outside, thus subdividing the wide tape into a number of narrow strips. The core is supported by five gun-metal arms, each arm consisting of two halves, which are clamped together by screw-bolts, and to make the fastening more secure, slots are cut out of the wrought- iron ring and part of the core into which the extremities of the arms enter. In the dynamo here illustrated the oore contains 7*8 square inches of iron in cross-section, and it is wound with 60 coils of 165 mils wire, each coil consisting of 6 turns. Total number of turns 360. The machine gives a current of 150 amperes at 75 volts pres- sure when running at 800 revolutions a minute. The G'ulcher Dynamo . — This machine is very similar Fig. 120. GULCHER DYNAMO, GULCHER DYNAMO, 309 to the Victoria in general arrangement, but differs from it in the way the core of the armature is constructed. In the original Giilcher,” which was the earlier machine of the two, the core consisted of a malleable iron ring of H- section, provided with external Pacinotti projections, which, for purposes of ventilation, were perforated. Flat iron washers were laid on either side of the central web, between the top and bottom flange of the H-shaped ring, which were kept a small distance apart by insulating pieces, and were also perforated to admit air to the in- terior of the core. In this arrangement those lines of force which enter the core in a direction more or less parallel to the spindle, pierce the iron washers on the broad surface and cause them to heat. To remedy this evil, the original design has been altered by employing a ring of T-section, the head of the T being directed towards the centre, and iron tape being coiled on either side of the middle web. When the winding of the tape is completed, the outer periphery of the core is turned in a lathe to a semicircular section. In this manner only the edges of the iron tape, but not its broad surface, are pre- sented to the lines of force, and thus heating is avoided. In a still more recent design, which is illustrated in Fig. 120, the armature core' is made by coiling rectangular iron wire, thinly covered with cotton, upon a supporting ring. In this manner the subdivision of the core is carried sufficiently far to avoid Foucault currents almost com- pletely. The Reckenzaun Motor , — The latest form of this type of motor, which has been specially designed for electric tramcars and launches, is shown in Fig. 121. The field magnets consist of wrought-iron plates bent hot into such a shape as to form the magnet cores of a Manchester ” 310 ELECTRIC TRANSMISSION OF ENERGY. double horse-shoe magnet, and the edges turned inwards are notched out to receive a series of segmental iron pieces which form the poles, as will be seen from the illustration* The armature is of the Pacinotti type, the core consisting of notched iron discs supported in the usual way. The motor illustrated weighs 560 lbs., and is rated at 8 hp., though capable of working up to 14 hp., as will be seen from the tests given below. The armature is 11 in. in Eig. 121. RECKENZAUN MOTOR. diameter and 10 in. long, the section of its core is 10 in. by If in., and that of the magnet cores is 9 in. by If in. The field magnets are series wound, each limb containing two coils which can be put in parallel or series so as to vary the exciting power and permit of the different rates of speed and power required in tramway work, without overstrain- ing the batteries and without having to introduce resis- tances. The following table shows the results of tests made with this size of motor by the Electric Car Com- pany of America at their works in Philadelphia. RECKENZAUN MOTOR, 311 Rev. Amperes. Volts. Brake H. P. Elec. H. P. Efficiency. 448 35 74 2-46 3*4 •72 576 36 91 3-16 4-4 •72 894 17-5 102 1-78 2-3 •77 466 44 81 3*49 4-79 •73 980 26 135 3*43 4-7 •73 680 53 118 6-12 8-3 •737 864 54 145 7*77 10*5 •74 940 44*5 150 6-58 8-9 •74 1034 37 154 5*68 7*6 •747 892 59*5 154 8*92 12-2 •73 826 64 146 9*3 12-5 •744 912 66 158 11-01 14 •78 914 66 150 10-16 13-2 •77 878 64*5 150 9-65 12-9 •75 CHAPTEE XI. Historical Notes — Fontaine’s Discovery — Figuier’s Explanation — Early Patent of Pinkus — Early Electro- Motors — Page’s Electric Eailway — Ploughing by Electricity at Sermaize — Electric Cranes — Ventilating and Pumping by Electricity — Modern Electric Railways — Different Systems — Comparative Merits of Battery System and Conductor System — The Bessbrook-Newry Electric Railway — The Blackpool Electric Tramway — The Telpher Line at Glynde — Reckenzaun’s Electric Tramcar — Com- parative Estimates for Horse Traction and Electric Traction — Progress of Electric Railways and Tramw'ays in Anaerica. The discovery of the principle that mechanical energy may be transmitted over considerable distances by the employment of two dynamo machines and a conductor, is commonly ascribed to M. Hippoly te F ontaine, who has in a recent pamphlet^ given a detailed account of the way he was led to make the invention. Since the matter is now of historical interest, an abbreviated translation of M. Fontaine’s account is here given. M. Hippoly te Fontaine says: — On the 1st of May, 1873, the International Exhibi- tion in Vienna was formally opened, although the ma- chinery hall, which was as yet incomplete, remained closed until the 3rd of June. I was engaged with the arrangement of a series of exhibits, then shown for the first time in public. There was a Gramme dynamo for electro-plating, capable of delivering 400 amperes at 25 volts, a magneto machine which I intended to work as a ^ “ Transmissions Electriques,” by Hippoly te Fontaine. Baudry & C*% Paris. FONTAINE S DISCOVERY. 313 motor from a primary battery^ or from a Plante accumu- lator, in order to demonstrate that the Gramme dynamo is reversible. There was also a steam-engine of my in- vention arranged for coke-firing, and a small motor of the same type, but arranged for gas-firing ; a centrifugal pump, which was intended to feed an artificial waterfall, and numerous other exhibits. To vary the experiments, I had arranged the pump so that it could receive motion either from the Gramme magneto machine or from my steam-engine. On the 1st of June it was announced that the machinery hall would be formally opened by the Em- peror on the 3rd, at 10 a.m. Nothing was then in readi- ness, but those who have been in similar situations know how much can be got into order in the space of forty-eight hours just before the opening of an exhibition. In every department members of the staff*, with an army of work- men under their orders, were busy clearing away packing- cases and decorating the spaces allotted to the diff*erent nations. The staff visited all the exhibits in order to determine which of them should be selected for the special notice of the Emperor. Roullex Duggage, the French Commissioner, asked me to set in motion all the machinery on my stand, and on the 2nd of June I was so far ready as to get the steam-engines, the plating dynamo, and the centrifugal pump to work. I failed, however, to get the motor into action, either from the primary or from the secondary battery. This was a great disappointment, especially because it prevented my showing the reversibility of the Gramme dynamo. It puzzled me the whole of the even- ing and ensuing night to find a means to accomplish my object, and it was only in the morning of the 3rd of June, a few hours before the exhibition was to be opened, that 314 ELECTRIC TRAN^^MISSION OF ENERGY. the idea struck me to work the small machine by a de- rived circuit from the large machine. Since I had no cable, I applied to the representative of Messrs. Manhes, of Lyons, who was kind enough to lend me a small quan- tity ; and when I saw that the magneto machine when coupled to the plating dynamo was not only set in motion, but developed so much power as to throw the water from the pump beyond the reservoir, I added more cable until the flow of water became normal. The total length of cable in circuit was then over two kilometers, and this great length gave me the idea that by means of two Gramme machines it would be possible to transmit mechanical energy over long distances.” Another version of this discovery, as given by M. Figuier, is that it was purely accidental. He says that at the Vienna Exhibition in 1873 the Gramme Company had two machines exhibited. One machine was in motion and the other was standing still. A workman noticed some cable ends trailing on the ground, and thinking they belonged to the machine which was standing, placed them in its terminals. To the surprise of everybody the machine immediately began to turn of its own accord, and then it was discovered that it was being worked by the current from the other machine. Whichever of these two versions may be the true one, it is certain that the electric transmission of energy was known at least as early as 1873, but there is reason to believe that the idea is even older. Dr. W. Adams, in his paper On the Evolution of the Electric Railway,” ^ states that in 1840 one Henry Pinkus obtained from the United States Patent Office provisional protection for his ^ Read in 1884 before the Society of Civil Engineers, America. EARLY ELECTRIC PROPULSION. 315 invention of an electric railway. The power was to be obtained from an electric motor — placed on the car — and set in motion by the current obtained from huge batteries. Since the latter were supposed to be buried in the ground the current must have been led to the car over some distance^ and Dr. Adams says that the principle of the transmission of the current to the car while in motion for the purpose of effecting its propulsion, was the same as that used nowadays. Thus we see that the earliest attempts at electric trans- mission of energy were made in combination with the problem of electric locomotion, and the following is a brief summary of the various stages the invention has passed through, as given by Dr. Adams. For fuller particulars the reader is referred to the paper already mentioned. The first electric motor for producing rotary motion direct — as distinguished from the earlier electric engines,” which had a reciprocating action — was that in- vented in 1833 by Professor Henry in America. This motor was but a toy, but shortly afterwards Davenport in America, Professor Jacobi in Russia, Davidson in Scot- land, and Little in England constructed motors of con- siderable size. Amongst these the best known is Jacobi’s motor, as applied to the propulsion of a boat on the Neva in 1839. In this instance the motive power was furnished by a primary battery, and the motor developed about two horse-power. In 1845 Professor Page invented a new form of electric engine based on the axial force of electro-magnetism,” and a few years later he proposed the use of this engine for the propulsion of railway trains. The idea gained public favour, and Congress actually placed the sum of £6,000 at the disposal of Professor 316 ELECTRIC TRANSMISSION OF ENERGY. Page for the purpose of practically developing the inven- tion. In 1851 an electric locomotive was built and employed to draw a train of cars between Washington and Bladensburg, a distance of five miles. The speed obtained was 19 miles an hour^ but since the current was furnished by batteries, the working expenses were so great as to preclude the possibility of commercial success. It was only after the discovery of the dynamic principle by Varley, Siemens, and Wheatstone that electric rail- ways and, indeed, any form of electric transmission of energy, became commercially possible. The idea of generating electricity at a fixed point by dynamo- machines and conveying the current through conductors and sliding contacts to the car whilst in motion, was first put into practice by Siemens in 1879, and this system forms to the present day the basis of all electric railways operated direct from the generating dynamo. After this short review of the history of electric loco- motion it will be opportune to cast a rapid glance over the earliest examples of electric transmission between two fixed points. As was already stated, the first of these experiments dates back to 1873. These, however, were experiments only, undertaken to demonstrate the idea at the Vienna Exhibition. In 1879 we find one of the earliest practical applications of the new invention under- taken by MM. Chretien and Felix at the sugar works in Sermaize. The manufacture of beetroot sugar can only be carried on during a small portion of the year, and for the rest of the time the machinery remains idle. It was thought advantageous to utilize the steam-engine at the works during slack time for ploughing the fields round about the factory, and if this should prove success- ful, to extend the system to other kinds of agricultural PLOUGHING BY ELECTRICITY, 317 work. A Gramme dynamo in the factory was set in motion by the steam-engine there, and the current was led by insulated cables to the field to be ploughed, a distance of about half a mile. The ploughing tackle was arranged in similar manner to that in use for steam ploughing, but instead of the two steam-engines at opposite sides of the field, trollies, each provided with two Gramme dynamos and suitable gear, were employed. Each trolly was pro- vided with the usual cable drum, and the plough was drawn backwards and forwards across the field by a steel wire-rope coiled and uncoiled alternately on these drums. Thus it was only necessary to switch the current into one or into the other set of dynamos on the trollies to produce the to-and-fro motion of the plough. After each set of furrows was completed the trollies were advanced by an equal distance until the whole length of the field was ploughed. The forward motion of the trollies was effected by the power of the motors, suitable gear having been provided for that purpose. The speed of the plough was 55 feet a minute, and the work was done at the rate of 200 square feet a minute. This performance is about equal to that which could have been obtained with a 5 to 6 horse-power Fowler steam-tackle. At the same works, M. Felix installed in 1878 an elec- tric chapelet-lift for discharging the beetroot from the vessels, by which means a saving in labour of 40 per cent, was effected. A similar but larger lift has recently been erected at Soissons in France, which is capable of dis- charging 500 tons of beetroot in twenty hours. The early example set by M. Felix has been largely followed in France, where a considerable number of elec- tric cranes and hoists have been erected. To mention only a few of the more important examples, the cannon 318 ELECTRIC TRANSMISSION OF ENERGY. foundry at Bourges was provided in 1882 with an electric foundry crane capable of lifting 20 tons. It is worked by a 12 horse-power Gramme motor, the current being sup- plied by a Gramme dynamo, situated 330 yards from the crane, and requiring 20 horse-power when the maximum load is being raised. A second crane capable of raising 40 tons has since been constructed for the same works. A 30-ton foundry crane in the works of M. Joseph Farcot, which was originally designed for hand labour, has been fitted with electric gear, and the lifting speed is from three to four times as great as was formerly the case when the crane was worked by ten men. To provide against accidents, an automatic apparatus is introduced which interrupts the current when the load exceeds 30 tons. The commercial efficiency of the system — that is, the ratio of the work done in lifting the weight to the work required to set in motion the generating dynamo — is stated by M. Fontaine to be 38 per cent. For ventilating mines and buildings, electric trans- mission of energy has been largely used. Electro-motors are, indeed, specially applicable for working fans, since, on account of the high speed equally required by both, the motor can be coupled direct to the axis of the fan. An early example of this kind of work is the installation at the Blanzy mines made by M. Mathet. There a fan of 2' 7” diameter, and 12" wide, is placed at the bottom of the pit, 540 yards below the surface, and is worked direct by a Gramme machine, the current being supplied by a similar machine on the surface worked by a 10 horse- power portable engine. The cost of the installation, ex- clusive of that of the portable engine and of the fan, was £160 ; and M. Mathet estimates that to do the same work by pneumatic transmission would have cost £580, ELECTRIC VENTILATION. 319 apart from the fact that, on account of the lower efficiency, a much larger portable engine would have been required. Amongst later examples of ventilating by electricity may be mentioned the Hotel de Ville in Paris, where 35 fans, each provided with a small electro-motor, are distributed throughout the building. The current is supplied by two Gramme dynamos, each capable of de- livering 50 amperes at 110 volts pressure. Either one or both of these generators may be used. Their speed is 1,250 revolutions a minute ; and that of the fan motors, which are of different size, varies from 1,450 to 1,750 revolutions a minute. The current is distributed from a central switch-board, so that all the 35 fans can be con- trolled from this point. A similar installation has recently been fitted up in the Ecole Centrale in Paris, but since there the fans are coupled to the motors by means of belts, special apparatus had to be employed to give warning to the engineer in case one of the belts should come off. This is done in the following manner : — In each motor circuit there is in- cluded an electro-magnet, the armature of which can assume three positions — the one home against the core when contact is made, and the normal current passes ; the other right off, when an alarm-bell is put into circuit ; and the third midway between these extreme positions, when no current passes. The armature is held in the middle position by a catch. A spring tends to draw the armature away from the core, but with the normal current the electro-magnet is sufficiently strong to keep the armature on. Should, however, a belt fly off, then the motor will begin to race, whereby the current will be re- duced, and the attractive power of the electro-magnet 320 ELECTRIC TRANSMISSION OF ENERGY. will become weakened so far as to allow the spring to pull the armature off. This breaks the circuit of that parti- cular motor, and rings the alarm-bell, thus calling the attention of the engineer to the fact that one of the venti- lators is out of action. It has been proposed by Professors Ayrton and Perry to establish electric ventilators at different points in the tunnel of the Metropolitan Railway, the current to be generated at some distant point where the nuisance of a stationary steam-engine would be far less objectionable than the fumes at present emanating from the stations and blow-holes along the line. The foul air was to be discharged from the ventilators by pipes passing through water-tanks. Amongst other early applications of electric transmis- sion of energy, may be mentioned the pumping arrange- ments in use since 1883 at a mine in Thallern, in Austria. Previously to the introduction of electricity, a 6 horse- power portable engine, placed at the bottom of the pit, was employed to work a centrifugal pump delivering 68 gallons a minute through 850 yards of tubing to a height of 200 feet. Now the engine has been replaced by an electro-motor, the power being supplied by a dynamo on the surface, which gives a current of 15 amperes at a pressure of 500 volts. This represents an electrical energy of 10*2 horse-power, and allowing 80 per cent, for the commercial efficiency of the generator, the total energy expended comes to 12*8 horse-power. The energy represented in water lifted is 4 horse-power, if we do not count friction in the tube ; including friction, it pro- bably amounts to over 6 horse-power. The commercial efficiency of the installation, including the two dynamos and the centrifugal pump, is therefore about 50 per cent. ELECTRIC TRAMWAYS, 321 Generally speaking, electric propulsion of carriages can be effected in one of two ways. We may either place batteries on to the car, and thus carry the source of energy along with it ; or we may employ a fixed source of energy, and transmit the current to the car whilst in motion by means of a conductor and sliding contact. For the sake of brevity we shall call the former the battery system, and the latter the conductor system. As regards their respective merits, it will be evident that the conductor system has the advantage of a more direct action, since only two conversions are required between the energy developed by the prime mover, and that actually used in propelling the car. In the battery sys- tem the energy of the prime mover must first be con- verted into electrical, then into chemical, energy, which is stored in the battery, and finally it must be recon- verted in the motor into mechanical energy. The inter- position of the battery between dynamo and motor must necessarily reduce the efficiency of the whole system, because we can never recover from the battery all the energy which has been put in. The extra weight which has to be carried is also a disadvantage. On the other hand, the loss of electric pressure occasioned by the re- sistance of the conductor may become very considerable, and the corresponding loss of energy may even exceed the energy which would be wasted in the battery. Thus the average resistance of the conductor at the Portrush Railway is 1 ohm, and when five cars are running distri- buted all over the line, requiring a total current of 200 amperes at 250 volts pressure, the loss of energy amounts to 37 horse-power. The power actually required for five cars is 68 horse-power, and therefore the efficiency of the Y 322 ELECTRIC TRANSMISSION OF ENERGY. 68 conductor, even if its insulation be perfect, is ^ = 65 per cent. If we add to this the loss due to the im- perfect insulation of the line, which is dependent on the state of the weather, we find that in this case the con- ductor system is, after all, not more economical than w’ould have been the battery system. The Portrush line is, however, an exceptional case, as the resistance of the conductor is rather great. In the Blackpool Electric Tramway the resistance of the conductor is only half an ohm, and the loss of pressure with six cars running — on the supposition that each requires an average of 18 amperes — would be about 30 volts out of 200. In this case 15 per cent, of the energy is lost in the conductor. If this line were to be worked on the battery system the loss would probably be 15 per cent, greater. But there is another consideration besides efficiency which must be taken into account when deciding between the two systems. In some cases the application of a fixed conductor along the line and above ground is inadmis- sible on account of the other traffic which may pass over the road. Such a conductor would not only interfere with all other traffic, but being always charged, and being of necessity unprotected by an insulating covering, so as to allow for the sliding contact, it would be a constant source of danger in our crowded streets. Mr. Holroyd Smith and Mr. Lineff have overcome the diffi- culty by placing the conductor underground. Where batteries are used, each car is perfectly independent from all the other cars, and this is a great advantage in working over a complicated net of tramroads. After this rapid comparison between the two systems, we may sum up by saying that the overhead conductor system is ELECTRIC TRAMWAYS, 323 adapted for lines running across country, where an over- head conductor and high electric pressure can be used without diflSculty ; the underground conductor system is adapted for suburban lines ; and the battery system is suitable for tramways within the crowded streets of a town. According to the nature of the conductor, the electric railways can be further classified as follows : — 1. The rails are used as conductors, one conveying the outfiowing and the other the returning current. In this case the rails must be insulated from the ground, and at the joints special connecting pieces must be used. The car wheels are insulated from their axles. An example of this kind is the short railway erected by Mr. Magnus Volk on the beach at Brighton, and the line between Berlin and Lichterfelde. 2. A separate conductor is used for the outfiowing, and both rails are used for the returning current. The rails need not be insulated from the ground, but special con- necting pieces must be used at the joints to insure good conductivity. The conductor may be above ground or under ground. Examples of this kind are the railways at Portrush, Newry, and Blackpool. 3. Separate conductors are used for the outfiowing and returning current. These are carried overhead on poles, and consist either of slotted copper tubes on surface rail- ways, or of angle iron on underground railways in mines. Examples of this kind are the railways at Modling, Berlin, Frankfurt, Zankerode mine, and others. 4. Separate conductors are used for the outflowing and returning current. These are attached to poles, and so arranged as to form one single line, along which sus- pended trucks run. An example of this kind is the Telpher line at Glynde. 324 ELECTRIC TRANSMISSION OF ENERGY. The Bessbrook and Newry Electric Railway wa& opened for traffic in September, 1885. It is three miles in length, and was erected to facilitate the traffic between these two towns, which amounts to about 28,000 tons annually. The generating station is placed at about the middle of the line at Millvale, where ample water-power is available. A turbine capable of working up to 65 horse-power is used to drive two Edison-Hopkinson dynamos (see description on page 262), one of these being sufficient to work the traffic, the other being held in reserve. The pressure employed is 250 volts, and the current is conveyed along a channel iron conductor laid at the same level as the rails, and supported on wooden blocks, which are attached to the cross sleepers in the centre of the track. In a front compartment of each of the two passenger cars at present in use there is an Edison-Hopkinson dynamo acting as motor, and there i& a collector with contact sliding on the centre rail, both in front and rear of the car, in order to span the breaks at farm crossings and sidings, where the current is continued by means of an underground cable. At one point the line touches the public road, and since there the con- ductor on the surface would be objectionable, it is inter- rupted for a distance of 50 yards, and the gap is spanned by two overhead wires, supported on poles 15 feet from the ground, and a collector with sliding contact is fixed to the roof of the cars for the purpose of bringing the current to the motor whilst the car is on this part of the line. The passenger car, which performs at the same time the function of an electric locomotive, weighs 8 tons,^ and on the level attains a speed of 15 miles an hour. It is capable of accommodating 34 passengers, and of haul- ing at the same time a train of loaded waggons up an BLACKPOOL ELECTRIC TRAMWAY, 325 incline of 1 in 85, at a speed of 7 miles an hour. The gross weight of the whole train, including locomotive and passengers, is 26 tons, and the motor develops about 25 horse-power. The steepest gradient is 1 in 50, and the sharpest curve has 150 feet radius, but at the two termini there is a pear-shaped loop with a minimum radius of 56 feet 6 inches. This arrangement obviates the difficulty of having to turn the cars on a turn-table at the end of each journey. The cars are 35 feet long, and run on double bogies, having a gauge of 3 feet, and the ordinary flanged wheels. The goods waggons have flangeless wheels, 3 feet 4^ inches gauge, and run on two flat rails placed outside of the car rails, and J inch below them. The car rails form thus a guide for the wheels of the goods waggons, and the latter can by reason of their broad flangeless wheels be at either terminus drawn off the track, and over the ordinary country roads. In the flrst four months after the opening of the line, a total of 25,000 passengers and 1,600 tons of goods was carried, and the total mileage was 5,200. In the Blackpool Electric Tramway, which is two miles long, the conductor is placed under ground and consists of two semicircular channels of copper (Fig 122), sup- ported by, but insulated from cast-iron chairs. The con- ductor is split up into two parts in order that any dirt or otherforeign matterwhich mightfall through the slot in the roadway should also fall through the space between the two halves of the conductor instead of lodging on it, as would be the case if a single conductor were placed directly under the slot. The collector consists of a steel frame narrow enough to pass through the slot and of contacts sliding along the underground conductor. The contacts are insulated from the steel frame, and are in electrical 326 ELECTRIC TRANSMISSION OF ENERGY. communication with a clip terminal on the car by means of an insulated cable. Light leather straps serve to draw the collector along in the slot. Should an obstruction occur in the slot or in the conductor of so serious a nature that it cannot be brushed away by the passage of Fig. 118 . HOLROYD SMITH UNDERGROUND CONDUCTOR. the collector, the latter is arrested and the leather straps break. The strain next comes on to the insulated cable, which is thereby drawn out of the clip terminal, and thus the current is interrupted and the car comes to rest. In this manner the attention of the driver is called to the UNCERTAINTY OF SLIDING CONTACTS, 327 obstruction^ which can then be removed by hand. From the terminal at the under-side of the car the current is led through a variable resistance and a reversing switch to the motor, and returns through the wheels to the rails, and along them back to the generating station. At first the motors were shunt-wound, so as to avoid racing when the car was lightly loaded and running on a level part of the line, or heavily loaded and running down an incline. It has been explained on page 153 that the speed of a shunt motor, when running light, can never exceed a certain limit, whereas a series motor may, under the same condi- tion, assume a dangerously high speed. On purely theoretical grounds shunt motors are, therefore, more suitable for tramway work. But a serious practical diffi- culty was soon encountered. It arose from the uncer- tainty of electrical 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, perhaps only a fraction of a second, but the interval was sufficient to cause the field of the motor to lose its magnetism. The conse- quence of this was, that when contact was restored and the current began again to flow, the armature was not able to offer any counter-electro-motive force, and an abnormal rush of current took place before the field magnets had had time to again become excited. It will be noticed that the injurious eflFect here described will be the greater the lower the resistance of the armature — that is to say, the more efficient the motor, the more will it suffer from 328 ELECTRIC TRANSMISSION OF ENERGY. an occasional interruption of current. Since it was impossible to absolutely 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-electro-motive 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 re- sistances 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. The use of artificial resistances — in this case a necessary adjunct of the system — entails, of course, some waste of energy, and in this respect Mr. Recken- zaun’s method of varying the power by a combination of motors is preferable. The motors — one to each of the six cars now in use — are 6 horse-power nominal, but may for a short time be worked up to 10 horse-power, the speed being 1,000 revolutions a minute. Each motor weighs 9 cwt. It is worked at an average speed of 800 revolutions a minute, and requires an average current of 18 amperes at 200 volts pressure, or about 5 electrical horse-power, equal to 4 brake horse-power, to propel a car with 45 passengers on a level road. The direction of motion is reversed electrically by reversing the direction of the current through the armature, but not through the field magnets. In doing this the brushes are not shifted, and the diameter at commutation remains always at right angles to the magnetic axis of the field. The brushes consist of small solid blocks of copper, pressed by springs very tightly against the commutator. All screws are of steel, and provided wdth lock-nuts to stand the vibration of the car without becoming loose. The TELPHER LINE AT GLYNDE. 329 armatures are 10 in. in diameter, and wound with a single layer of 63 mils wire, insulated with pure silk. The generators, of which there are two, placed in a generating station at about the middle of the line, are of the type described on page 289 and illustrated in Fig. 103. Each of these dynamos weighs 4 tons, and is capable of deliver- ing a current of 180 amperes at 300 volts pressure, when worked at a speed of 500 revolutions a minute. But since it was found that a pressure of 200 volts is sufficient to work the present traffic, the speed has been reduced to 350 revolutions a minute. The armature is 16 inches in diameter, and the field magnets, which are of the four- pole type, are separately excited by small dynamos, for the purpose of being able conveniently to alter the electro- motive force within certain limits, according to the requirements of the service. The Telpher Line at Glynde is an electric railway about a mile long, acting automatically, without assis- tance of guard or driver, and intended for the conveyance of a continuous stream of light vehicles, suspended from and rolling on a single line of rails, which at the same time form the electric conductors. In the illustration (Fig. 123), M is the telpher locomotive, consisting of an electro-motor, chain gear, and driving wheels with india- rubber treads, and also provided with two governors. One of these breaks the current when the speed attains a certain limit, and the other puts a brake on if from any cause the speed of the train should still further in- crease. On either side of the locomotive there are placed 5 skeps, each weighing 101 lbs. and capable of carrying 250 to 300 lbs. of clay, and these skeps are kept the right distance apart by connecting rods. The total length of the train, consisting in all of 11 vehicles, is 330 ELECTRIC TRANSMISSION OF ENERGY. exactly equal to the distance between two poles, and since the sections of the rail attached to the poles form alternately the out-and-home conductor, it follows that the first and last skep are at all times in contact with raik of opposite polarity. The current is collected at the two ends of the train, and conveyed along wires (not shown in Fig. 123. the illustration) to the middle, where it works the motor, M. The arrangement of the circuit is shown diagram- rnatically in Fig. 124, where is the generating dynamo, and X, 1\ Zi, two trains, one up and the other down the line. The sections, Ai, B^, Ar,, and so on, are connected together by cross-connections shown in dotted lines, and are also connected with the positive terminal of BLOCK SYSTE3L 3311 the dynamo, whilst the alternate sections, ^ 3 , ^ 4 , and so on, are similarly connected, and are also connected with the negative terminal of the dynamo. The con- ductor is formed of steel rods, | inch in diameter, 66 feet Fig. 124. 4 - ^ ( 5 ) ^ ^ \/ ^3 \i ^2 \i A A ; i \ ( \ / \ " + ^ 4 . ^3 @ ^ O/i// ^ ~ + long between the poles, and placed 8 feet apart. The motor is regulated to run at a speed of 1,600 to 1,700 revolutions a minute, the speed of the train being 4 to 5 miles an hour. One train running backwards and for- w’ards will deliver 150 tons of clay per week, but as many as 20 trains can be run on the double line at the same time. To avoid in this case the risk of collision, the late- Professor Fleeming Jenkins, in conjunction with Pro- fessors Ayrton and Perry, invented an automatic elec- trical block system, which is shown diagrammatically in Fig. 125. At certain parts of the line the regular cross-over system is modified by inserting idle sections^ Fig. 125. o' -h ^ &S ^ As, Bs, tlirough which the driving current is only trans- mitted if the switches, K, K^, are closed. If the switch, K„ is open, a train arriving on As will stop for lack of current, and similarly, if K is open, a train arriving on Bs will stop for lack of current. On closing the 332 ELECTRIC TRANSMISSION OF ENERGY. switches, the trains wall start again. These switches are worked, like the contact of an ordinary relay, by a small signal current sent back by a separate circuit, and operated automatically by the preceding train. One signal opens the switch, thus blocking the line ; the next signal closes the switch, thus again restoring communica- tion, and allowing the following train to come on. Reclienzaun^s Electric Tramccir, Very soon after the invention of the secondary battery attempts were made both in France and in this country to utilize its capacity to store energy for the propulsion of vehicles. These early attempts, however, failed, for two reasons. In the first place, the earlier forms of accumulators were very heavy in comparison to the amount of energy which could be obtained from them, thus necessitating the carrying of an enormous dead weight, ’which left very little margin for the paying load. They were also unreliable, gave but a poor efficiency, and were in many points mechani- cally defective. In the second place, the gear employed to reduce the high speed of the motor to the compara- tively slow speed of the car-wheels was uncertain in its action, and liable to derangements. At first belt-gear w^as tried, but that failed when used on common roads, as might be expected of belts which have to work wet and dry, and always in a more or less muddy condition. Mr. Magnus Volk, of Brighton, has adopted belt-gear in his little electric railway laid along the beach in Brighton, and is quite satisfied with it. He employs leather link belts in duplicate, and each belt can be tightened by a pulley supported on a lever which is under the control of the driver. Ordinary leather belts were tried at first, but were found quite unsuitable, even on this line, wffiich is exceptionally clean. There is no experience as to how RECKENZAUJSP S ELECTRIC TRAMCAR. 333 leather link belts will stand on a dirty road where ordinary belts fail. Next spur-gear and bevel-gear were tried. If the distances and relative positions between the centres of the geared wheels could be kept rigidly constant, such gear would probably answer very well ; but in a tram- car there is of necessity a certain amount of play between the axle-boxes and the body of the car, and the height of the car-frame above the centre of the axle is variable. There is about an inch difference when the car is loaded and when it is empty ; and along our ordinary tramroads, the vertical oscillation of the car may considerably increase this difference. It is difficult to arrange spur-gear to be sufficiently flexible to accommodate itself to these changes, but this problem has been solved by Mr. Sprague and Professor Elihu Thomson, who adopt a spring suspension for the motor, a system now largely in use on American roads. Chain-gear has also been tried, but not with such success as to become popular. Mr. Keckenzaun employs worm-gearing, as will be seen from Figs. 126 and 127, which represent his electric tramcar in elevation and plan. Two motors are used, each supported on, and forming part of a four-wheel bogie, which is in itself an electric locomotive, and quite independent of the body of the car. The weight of the latter is thus distributed over eight wheels, rendering it possible to run the electric car, notwithstanding its in- creased weight, over the ordinary tramroads. The bat- teries are placed on trays under the seats, and, when ex- hausted, can be withdrawn and replaced by a set newly charged in about the same time as it takes to change a pair of horses. To facilitate the operation, rollers are provided on which the trays run, and the latter are hauled in and out by means of a winch mounted on the trolly bX) -5 RECKENZATJN S ELECTRIC TRAMCAR. 336 ELECTRIC TRANSMISSION OF ENERGY, whicli brings the batteries up to the car. The object of using two motors to each car is partly to distribute the driving power to two axles without the necessity of rigid mechanical connections — which, in the case of bogie cars intended for roads where there are sharp curves would be very difficult to arrange — and partly to obtain variation in speed without the wasteful device of introducing idle resistance into the electric circuit. It will be readily seen that by coupling the motors in series the electro-motive force available for each motor will be half the total electro-motive force of the battery, whereas if we employ only one motor, or if we place both motors into parallel connection, the total electro-motive force will be avail- able for each motor, and consequently the speed will be about double what it was in the former case. A com- pound switch is provided which enables the driver to make these variations in the coupling of the motors (viz., two in series, one only or two parallel) by means of a single handle. By means of another handle the direction of motion is reversed. The car is provided, in addition to the usual hand-brake, with a very powerful magnetic brake, and with an automatic arrangement which puts this brake on if the speed exceeds a certain limit. A car built on this principle four years ago has been supplied to the Berlin Tramway Company, and formed the subject of an interesting paper by Herr Zacharias, read in January, 1886, before the Elektrotechnischer Verein,^ in Berlin, to which the reader is referred for full particulars. Each motor weighs 420 lbs., or, both to- gether, inclusive of the gear, about half a ton. The accumulators, with their trays and accessories, weigh li ^ ‘‘ Elektrotechnische Zeitschrift,” Jxin. 1886 . COMPARATIVE ESTIMATES. 337 ton. They have to be changed every two to four hours. The total weights are as follows : — Car^ with motors, gear, and accumulators . 3*75 tons. 46 passengers, conductor, and guard . . 2*25 tons» Total . . . 6*00 tons.- The tractive force required on a level average road is- 30 lbs. per ton, and at a speed of seven miles an hour thi& represents about 3^ horse-power work done. Herr Zacharias makes the following comparative esti- mate as regards the cost of horse traction and electric traction. He assumes that each car is actually in use from five a.m. until one a.m. — that is, for a period of twenty hours per day — and that it requires a change of horses every four hours. This gives five pairs of horses per day per car. A line worked by sixty cars would, therefore, require 600 horses actually in service, and say ten per cent, more in reserve, or 660 horses in all. To work the same line on the battery system would require steam power up to 750 horse-power, and a pro- portionate amount of electrical plant as given below. The capital outlay becomes — I. For Horse Traction : — Horses ...... £28,512 Harness and other gear .... 2,750 Total . . . £31,262 z 338 ELECTlilC TRANSMISSION OF ENERGY, II. For Electric Traction: — Steam-engines ..... £7,500 Boilers ...... 4,000 Dynamos ...... 2,800 140 sets of batteries .... 12,600 Cables and electric fittings . . 1,100 Motors and gear ..... 6,000 Total .... £34,000 Thus the first capital outlay is for electric traction only slightly greater than for horse traction, and if we consider that the buildings necessary to accommodate steam and dynamo machinery of a total power of 750 horse- power are not so extensive, and do not cover as much land as the buildings required to accommodate 660 horses, the balance in the first outlay may probably be in favour of electric traction. The working expenses are certainly much lower for electric traction. Herr Zacharias esti- mates as follows : — I. Working Ex]3enses with Horse Traction : — Depreciation per horse per day . 0*4840 shillings. Fodder „ „ . 1*5720 „ Shoeing and attendance, per horse per day ..... 0*1613 „ Total . . 2-2173 Total for 660 horses and 365 days . . £26,707 Renewal and repair of harness . . . 723 Total . . £27,430 COST OF ELECTRIC TRACTION. 339 II. Working Expenses with Electric Traction : — Annual expenditure of energy, 6,570,000 horse-pow'er hours. Coal £6,570 Depreciation of batteries, 20 2,520 Depreciation of motors, 20 "^ 1,200 Depreciation of boilers, steam-engines, and dynamos, 10 1,430 Repairs, oil, acid, wages .... 1,180 £12,900 According to these estimates the annual working ex- penses of electric traction on the Keckenzaun system would only be about half as great as with horse traction. The progress made wdth electric tramways and railways in America during the last few years is very remarkable. The roads are mostly worked on the overhead conductor system, and the following table, compiled by Mr. G. W. Mansfield, and given in his paper read on the 8th August, 1889, before the National Electric Light Association, shows clearly this development. Electric Street Railroads in the United States. 1885. 1886. 1887. 1 1883. Jan. 1 — July 1, 1889. Total. Operat- ing. Build- ing. Total. Number of elec- tric railways . Number of miles 3 5 7 33 19 42 61 109 of road . . . 7-5 28 29 130-5 113 267 380 575 Number of cars . 13 39 81 265 1 174 364 538 936 CHAPTER XII. General character of work done in England and abroad — Development of electric transmission of energy in Switzerland — List of installations — The Kriegstetten-Solothurn plant — Official tests of same — The Aichberg plant — Latest type of Brown’s dynamo. At the present time there are so many electric trans- mission plants in actual use, that the task of choosing one or two for more detailed description becomes a some- what invidious one. Such descriptions of work actually carried out are of great importance to the practical en- gineer, and might, therefore, appropriately fill the con- cluding chapter of a book intended not for professors, whose business it is to teach the theory of electrical work, but for practical men who might be called upon to design and erect such work, or to approve and pass it if designed and erected by others. Admitting, therefore, the ne- cessity of giving in this place, by way of example, an account of some of the work done in electric transmission of energy, the question arises as to what examples should be chosen. As the present book is primarily intended for English readers, the author might naturally be ex- pected to select for detailed description work carried out by English firms. But here arise some difficulties. In the first place, although a considerable amount of work of this kind has been carried out by English firms, the magnitude of each individual plant is far below the average of continental or American installations, a fact HISTORICAL NOTES. 341 primarily due to the scarcity of water power and the abundance of cheap fuel in England. Since then the opportunities of utilizing large powers, which otherwise would be running to waste, are few and far between in this country : the application of motors has become rather a question of the convenient distribution of small powers, and the plants required in such cases, as pumping, ven- tilating, driving lathes, and other tools, hoisting or pro- pelling launches and tramcars, are necessarily of limited size. In the second place, English makers have hitherto not considered it necessary or expedient to submit the work they have carried out to the test of independent authori- ties, and in recording the results achieved the author would have to simply repeat the maker’s statements. Now, whilst he is far from suggesting that such state- ments should, as a general rule, be received with incredu- lity, it will be admitted on all hands that the results obtained by independent investigators must, on the whole, be preferred to those obtained by interested parties, and he proposes, therefore, to select for detailed notice the work done by the Maschinenfabrik Oerlikon, Switzer- land, which firm has not only carried out the largest plants of electric transmission, but has also had some of them investigated by independent authorities, and thus given the scientific world the benefit of their experience. Before entering into the technical details of one or two of these plants, it may be interesting to state briefly the reasons which have led Mr. C. E. L. Brown, the elec- trical engineer to the Oerlikon works, to take up the electric transmission of energy as a promising branch of engineering, and to trace the history of the development during the last three years. Up to 1886 the Oerlikon 342 ELECTRIC TRANSMISSION OF ENERGY, firm had only made electric light dynamos, but the high efficiency which could be obtained with these, as with any other w^ell designed machines, and the fact that they could equally well be used as motors, led Mr. Brown to the conclusion that their application to the transmission of power was a question which could be solved by careful design and good workmanship. The necessity to transmit power over longer or shorter distances was, as far as Switzerland was concerned, abundantly proved by the fact that the various tele-dynamic transmission plants had, notwithstanding their recognized defects, been re- tained for many years, and are still retained. The distance over which wire rope can be used is compara- tively small, the installation is costly, the efficiency when the distance is such as to require many intermediate stations is low, and the wear of the cables is great, en- tailing not only a heavy charge for renewal of cable, but also occasional interruptions in the working when it becomes necessary to splice an old or put on a new cable. In spite of all these imperfections tele-dynamic transmission had been retained for want of something better. Here was, then, an excellent field for the introduction of electric transmission, if a convincing and practical proof could be given of its superiority. This proof has been given, and at the present time a slow but sure process of replacement of the wire rope by the electric conductor is going on in Switzerland and neighbouring countries where water power is abundant. The first installation carried out by Mr. Brown was one of 50 horse-power, between Kriegstetten and Solothurn, a distance of five miles, and gave a commercial efficiency of about 74 per cent. Considering the great distance, it was obvious that HISTORICAL NOTES, 343 no tele-dynamic or any other system of transmission could have been employed in this case, even had a very much lower efficiency been tolerated. The installation was, scientifically, a complete success from the first, and what was more important, it worked from the first day to the entire satisfaction of its owner, Herr Miiller-Haiber, but, notwithstanding this result, other mill owners were slow to avail themselves of the new system of transmission. They did not doubt that the results obtained by the com- mission called together to investigate the performance of the plant were quite reliable ; this was guaranteed by the names of the men who formed the commission, but the whole thing was at first looked upon merely as an interesting experiment, and it was feared that sooner or later a break down in what was then regarded as delicate electrical machinery would cripple the plant. The hy- draulic and other engineering works necessary in con- nection with such a system of power transmission entail a heavy expenditure, which might become so much capital thrown away in case the electrical part of the installation could not be kept permanently in good working order, and therefore, those who might at once have used electric transmission, thought it prudent to defer its adoption until prolonged experience with the Kriegstetten-Solothurn plant had proved that the elec- trical part could be as much relied on as the purely mechanical part. This point having been fully demon- strated, in course of time the conversion of mill owners in Switzerland and elsewhere to the new system began to take place, and is now, as already mentioned, in full swing. Even engineers, who make a speciality of tele- dynamic transmission, are rather pleased than otherwise that the growing use of electric transmission relieves them 344 ELECTRIC TRANS3IISSI0N OF ENERGY, of having to do work Avhich, even if executed with the greatest care, gives frequent rise to complaints. Even the rope transmission of Schaffhausen, once the pride of Swiss engineers, is about to be superseded by the new rival system. A little below the present turbine house, from whence about 760 horse-power are now being trans- mitted across the stream by wdre rope, there will be erected a new turbine house, accommodating five 300 horse-power turbines to drive dynamos of corresponding output, and a large spinning mill on the other side of the river has already contracted to take 520 horse-power, supplied by four motors, two of 60 and two of 200 horse- power. Not only in Switzerland, but also in other countries, is there a steady growth of electric trans- mission, as will be seen from the following list of installa- tions already completed by the Oerlikon works, or in hand. In this list installations of less than 50 horse- power, available at the generating station, are omitted as being of subordinate interest. Installations in connection with electric railways are also omitted. The last item on this list is an interesting example of the application of electric transmission of power to rail- way tunnelling, the plant, which is being made to the order of Messrs. J. E. M. Clark, London, for the Argen- tine and Chili Railway, being intended for working the air-compressors which will supply air for the rock drills required in the construction of the tunnels. Since, in this case, the generators and motors will have to be carried to their respective places of destination on the backs of mules, they will be so constructed that in the larger machines no piece will weigh more than 10 cwt., and in the smaller more than 3 cwt. The machines will be six-pole drum dynamos, the limbs of the field magnets STATISTICS OF ELECTRIC TRANSMISSION PLANTS. 345 being separately detachable and mounted on a steel frame- work, which can also be taken to pieces. The type of dynamo which Mr. Brown employed in his Locality. Dynamos. Number. Kilwatts. Transmission. H. P. avail- able. Distance in Meters. Solotburn Lucerne Derend ingen Diessbacli Lugano Wald . . . , Piovene (Italv) Pordenone (Italy) Schio (Italy) Gazzaniga (Italy) . Cuorgne (Italy) . Aichberg Innsbriick Kennelbach . Podolnitcbaia (Russia) Scbaffhausen Argentine Republic and Cbili . 10 10 8 8 20 80 68 94 80 39 bO 84 20 65 167 40 98 72 67 50 81 58 63 59 67 33 65 49 23 17 200 310 50 54 49 27 24 50 120 280 120 60 125 250 60 300 100 120 96 100 50 200 70 I 600 I !,120 I 8,000 3.000 1,300 600 8.000 725 450 1,000 6,000 800 800 S25 600 450 2,500 300 600 3.000 and 7.000 hrst installations, and which is still retained in all cases where a moderate amount of power has to be transmitted, is shown in Fig. 128. ^ The plant erected at Kriegstetten ^ Tbe author is indebted for tins illustration, and fur Fig. 129, to tbe editor of “ Industries.” 346 ELECTRIC TRANSMISSION OF ENERGY. and Solothurn consists of two generators and two motors, the four machines being coupled in series on the three- wire sytem. The reasons for this arrangement were the following : 1st, by subdividing the work between two machines into pairs, the chance of a complete breakdown through the failure of one of the machines was avoided ; 2nd, the machines were designed with an ample margin of output, and the plant could therefore transmit even if partially disabled, considerably more than half the power ; Fig. 128 . 3rd, the great distance rendered the employment of a high pressure desirable (in this case 2,000 to 2,500 volts), but as at that time the adoption of so high a pressure in one single machine was looked upon as a somewhat hazardous experiment, Mr. Brown thought it prudent to minimize the risk by building the machines for half the pressure, and coupling them in series ; 5th, in case that only one set of machines is at work, the idle outer circuit wire can be joined at both ends to the balancing wire, and thus the RESULTS OF TESTS. 347 line resistance reduced. The power to be transmitted varies between 30 and 50 liorse-powei% according to the water level in the river, and the speed of both generators and motors has been fixed at 700 revolutions per minute.. The transmission of the power from the turbine to the generators and from the motors to the mill shafting is by belts. The machines are pure series-wound, and have been designed so as to make their characteristics fit each other in order to obtain a perfectly self-regulating system. The line consists of three overhead wires of bare copper 235 mils diameter, and carried on Johnson and Phillips’ oil insulators, the average distance between the poles, which are of wood, being 120 feet ; only in one place, w^here the line crosses the river Aare, is the span in- creased to 400 feet, and silicon bronze is used instead of copper. It may here be remarked that the insulation of the line was found to be practically perfect, as will be seen from the results of tests quoted below. The first series of tests was carried out at the works of the makers in November, 1886, the machines being con- nected by an artificial line. To ascertain the mechanical power supplied to each generator, and that given off by each motor, the machines were suspended in cradles hung on steel knife edges in line with the centre of the arma- ture shaft. It will be seen that with this arrangement the torque w^hich the armature exerts upon its field must cause the whole machine to assume an inclined position, the deviation from the vertical being the greater the greater is the torque applied to or exerted by the arma- ture. Now the relation between this angle and the corre- sponding torque can be ascertained by direct measure- ment when the machine is at rest ; and when the machine is at work, the product of the torque (as given by the 348 ELECTRIC TRANSMISSION OF ENERGY. angle of deflection) and the speed is a measure of the power supplied or absorbed. Certain corrections are naturally required to eliminate errors which are insepa- rable from this as from any other method of measure- ment, but it would be beyond the scope of the present book to enter into the details of the corrections adopted. Suffice it to say that after elimination of these errors the commercial efficiency of the whole plant, including the artificial line of about 10 ohms resistance, exceeded 70 and in some cases even 75 per cent. A detailed account of these experiments will be found in the Schweizerische Bauzeitung,” 1887. Encouraging as were the results obtained in these expe- riments, there was still the doubt whether the plant, after having been installed and at work for some time, would show an equally satisfactory performance. It is one thing to test a machine in one' s own shop, where everything can be ar- ranged in the most favourable manner, but it is quite another thing to test the same machine after it has been sub- jected to the strain of everyday use for some time, and is certainly workingunderlessfavourable conditions. In order to leave no room for doubt in this respect, the Oerlikon firm instructed a second commission of experts to investigate the transmission plant in situ after it had been at work for nearly a year, and a full report of these new experi- ments, which took place in October, 1887, will be found in the first and second number of 1888 of the journal already mentioned. The determination of the power by means of the cradle suspension method, which had pre- viously been employed, could obviously not be adopted in this case without seriously interfering with the work of the mill, and another method had therefore to be substi- tuted. The power given off by the motors was measured KRIEGSTETTES-SOLOTHURN PLANT. 349 by a brake, and that supplied by the turbine to the gene- rators was ascertained in the following manner : One of the generators was put on one side, and a brake set up in its place. The turbine was then started and the power absorbed by the brake was determined for different levels of water, different speeds, and different numbers of gates opened in the guide-wheel. In this manner a table was prepared giving the power corresponding to all the pos- sible conditions under which the turbine might work when the generator was replaced. The electrical measure- ments were made with instruments specially constructed for the purpose, the commission having found that the employment of commercial ampere and voltmeters might lead to an error of not less than 5 per cent, in the deter- mination of the efficiency. All the measurements were made as far as possible simultaneously at both ends of the line, and those readings about which there was any doubt of having been taken at the same moment were rejected. The resistance of the line with the balancing wire cut out was found to vary between 9*04 and 9 ’228, according to the temperature of the air. The average resistance of the four machines in series was 14*311 ohms, viz., 7*251 the generators, and 7*060 the motors. The power received at Solothurn varied from 17 to 23 horse-power. It will thus be seen that as far as the amount of power transmitted was concerned, the conditions of this test were not the most favourable which might occur, for the total efficiency of the plant would probably be greater at full power than at less than half-power, but the difference can only be small ; and as the smallness of the difference of efficiency between full and light load is one of the advantages which electric transmission has over other systems, we must accept the figures found under light load as fairly repre- 350 ELECTRIC TRANSMISSION OF ENERGY. renting the average efficiency within the limits of power for which the plant is intended. The trials took place on the 1 1th and 12th October, 1887, under the direction of Professors Veith and Weber, and Messrs. Amsler, Keller, and Hagen- bach, and the following is a summary of the results : — Horse- Power. Commercial Efficiencies. Eate. Remarks. Generators. Motors. Generators. Motors. Whole Plant. Oct. 11th j 26-17 17-85 •869 •888 •682 i One Generator ^ and \ One Motor. 24-56 16-74 •871 •868 •682 ■Oct. 12th J 30*85 23-21 •887 •903 •752 i Two Generators > and i 30-85 23*05 •888 •881 •747 j Two Motors. The insulation of the line was ascertained by simul- taneous current measurements at both ends. It would, of course, have been possible to take the insulation resis- tance by means of a bridge, but the value so obtained would scarcely have been reliable. A circuit may show a,pparently perfect insulation when tested by a bridge, with even as many as 50 or 100 cells, and yet leak badly when the full pressure of 2,000 volts is thrown upon it. The higher pressure itself may develop faults which no ordinary bridge test can find out. The line was therefore tested under pressure and the following results were obtained. i i Current at Date. Generator Motor Eemarks. Station. Station, October 11th. | 14-20 14-18 ) During a heavy 13-24 13-29 1 rain. October 12th. | 11-47 9-78 11-42 9-78 1 No rain. If there had been any appreciable leak in the line the current measured at the motor station would, in all cases. KRIEGSTETTEN-SOLOTHURN EXPERIMENTS. 351 have been smaller than that measured at the generator station^ but it will be seen from the above table that the difference is as often positive as negative, showing that it is merely due to errors of observation which in them- selves are exceedingly small. The insulation of the line may therefore be considered as absolutely perfect. A larger plant, but with only one generator and one motor has recently been established near Aichberg, Tyrol. In this case a water power of about 100 effective hp. in Gschroeff is utilized in the paper mill Steirermuehle, 1 kilometer distant. Both machines are of the type repre- sented by Fig. 128, and the working pressure is 1,000 volts, current 67 amperes. Tests carried out by a com- mittee of experts on behalf of the owner have shown that the commercial efficiency of the whole plant when work- ing at full load is about 80 per cent. The method adopted in making these tests was substantially the same as that just described. The line consists of 315 mils bare copper wire, also supported overhead on Johnson and Phillips’ oil insulators, and the machines are series wound and designed so that the characteristics may fit each other to obtain self-regulation. The following are the prin- ciple data of these machines. Generator : Armature Gramme wound with 504 turns of conductor consistino* of two 120 mils wires laid bare upon each other and cotton covered together. The commutator has 126 sec- tions. The diameter of the armature core is 27iin., and its length 20 in., the radial dej^th being in. The mag- net cores are 15|-in. diameter, and contain each 371 turns of exciting wire of 350 mils diameter. Motor : the arma- ture core of the motor is of the same size, but is wound with only 456 conductors joined with a 114 part commu- tator. The magnet cores are 15 in. diameter, and their 352 ELECTRIC TRANSMISSION OF ENERGY. exciting coils are of the same wire as those of the gene- rator and contain the same number of turns. During the tests referred to the speed of the generator varied between 570 and 577 revs, per minute, and that of the motor be- tween 630 and 635 revs, per minute. For larger powers Mr. Brown employs a multipolar type of dynamo, as shown in Fig. 129,^ Avhich represents, a 250 hp. machine having a four-pole field and Gramme wound armature. Similar machines were shown at work in the recent Paris Exhibition, the motor driving the shafting to which the different machines exhibited in the Swiss section were belted. In the latest machine of this type Mr. Brown employs drum armatures with mul- tipolar series winding, and Fig. 130 illustrates a dynama of this type designed to give an output of 200 amperes at 400 volts pressure when driven at 450 revolutions per minute. The dimensions are given in millimeters.^ Where weight need not be economised the field magnets are of cast iron, but if a specially light machine is required, the magnets are separate wrought-iron forgings held to- gether by bolts placed within the exciting coils, and are supported on a steel frame instead of a cast-iron bedplate. With the armature winding adopted only two sets of brushes are required for any number of poles, and the brushes must be placed with an angular distance of 180 degrees, if the field contains an odd number of pairs of poles. If the number is even, then the angular distance between the brushes is equal to that benveen the poles. Thus in a 6, 10 or 14 pole machine, the brushes must be placed diametrically opposite, whilst in a 4, 8 or 12 pole machine they must be placed respectively 90, 45 or 30 ^ See frontispiece. . ISO. A A brown’s bynamo ; 200 umpires, 400 volts, 450 revolutions. 354 ELECTRIC TRANSMISSION OF ENERGY. degrees apart. The induction in these new machines is much lower than in former types being only from' 5^000 to 7^000 lines per square centimeter^ or say from 64 - to ? 4 - lines per square inch^ both in armature and magnet cores. THE END. INDEX. Absolute system, the, of electro- magnetic measurements, 22. Aerial lines, 215 ; estimate for, 225 ; material for, 221. Aichberg plant, the, 351. Allen engine and Kapp dynamo combined, 304. Alteneck armature, the, 69. American Sectional Underground Company, its system, 237. Ampere-Turns, 103. Andrews dynamo, the, 292. Argentine and Chili Eailway tunnelling, 344. Armature, electro-motive force in, 80 ; the Gramme, 73, 77 ; the Hefner-Alteneck, 69 ; the Paci- notti, 73, 77, (310 ; Keaction, 129, 132 ; Siemens’ shuttle- wound, 61 ; torque exerted by, 89 ; types of the, 101, 261. Ayrton and Perry on electro- motors and their government, 165, 173 ; on ventilation, 320. Barlow’s wheel, 49. Battery, the secondary, 8 ; the battery and conductor systems of electrical railways, 324. Beringer’s, Herr, investigations, 245. Bessbrook-N ewry electric railway, the, 324. Blackpool electric tramway, the, 325. Block system, the automatic, 331. Brighton, electric railway at, 332. Brooks’ underground conduit, the, 238. Brown dynamo, the, 305, 341, 345; latest type of, 352. Cables, lead-covered, 239. Capacity of conductors, 47. Capital outlay and waste of energy, relation between, 201. Centimeter, the, 22. Chain of magnetized molecules, 15. Characteristics and characteristic curves, 114, 124, 127; pre-deter- mination of, 115 ; speed, 130. Chretien and Felix MM., their system of ploughing by electri- city, 316. Circuit, magnetic, 102 ; for elec- tric distribution and transmis- sion, 214. Classification of systems, 160 ; of dynamo electric machines, 260. Commercial efficiency, 66, 184, 248, 342 ; maximum, 147. Compound machine, the, as gene- rator, 154. Conductors, area of, 202 ; attach- ment of, to insulators, 218 ; capacity of, 47 ; comparison of, 247 ; heating of, 211; most eco- nomical size of, 202 ; and battery railway systems, 321. Constant pressure, transmission at, 162 ; speed, self -regulation for, 192. Contact, the sliding, uncertainty of, 321. Continental Underground Cable Company, 239. 3o6 INDEX, Continuous ciuTent transformer, 279. Conversion, efficiency of, 97. Cost and efficiency, tables of, 254. Cost of plant, 246 ; of power, the, 245. Counter-electro-motive force, 37, 141. Couplings, 220. Cranes, electric, 317. Crompton dynamos, the, 289, 291. Current and mechanical force, 33. Current, unit, 28. Current, variable, correction for, 209 ; loss of, by leakage, 178. Curves, characteristic, 114; horse- power, 135. Cylinder, the, 100, 261. Deprez, M. Marcel, his exj^eri- ments, 91, 173. Disc, the, 100, 261. Distribution, electric, circuits for, 215 ; at constant current, 172 ; at constant pressure, 163. Dobrowolsky, Herr von, 132. Dobrowolsky triangle, 133. Drum, the, 100, 261. Dynamic, static, and counter- electro-motive force, 126, 140. Dynamo, the Andrews, 292 ; the Brown, 305, 341 ; the compound, 155 ; used as motor, 171 ; the Crompton, 289 ; the Edison- Hopkinson, 262 ; the Elwell- Parker, 284 ; field of the, 74 ; Forbes’ non-polar, 51 ; the Goolden, 294 ; the Giilcher, 307 ; ideal alternating current, 54 ; ideal continuous current, 59 ; the Immisch, 272 ; the Kapp, 300 ; machines, classification of, 260; the Laurence, Paris, and Scott, 276 ; reversibility of, 84 ; the Manchester, 282 ; the PhoG- nix, 296 ; series- wound, 156 ; the Siemens, 134, 293 ; charac- teristics of, 135 ; the Thomson- Houston, 263 ; the uni-polar and non-polar, 53 ; the Victoria, 307. Dynamos, small, difficulty in, 113. Dynamos and motors, different conditions in, 85 ; formulas for, 98. Dyne, the, 22. Ecole Centrale, Paris, ventilation at, 319. Edison electric tubes, 237. Edison-Hopkinson dynamo, the, 262. Edison mains, 230. Efficiency, 43, 97, 183 ; maximum theoretical, 147 ; and cost, tables of, 254. See also Commercial efficiency. Electric machines. See Dynamo, &c. “ Electric Railway, On the Evo- lution of the,” Dr. Adams’s paper, 314. Electric street railroads in the United States, 339. Electric traction and horse trac- tion, comparative estimates, 337. Electric transmission of energy, possible applications of, 241. Electrical efficiency, 183. Electrical potential, the, 31. Electrical and mechanical energy, relations between, 11. Electro- dynamic paradox, an, 159. Electro-magnetic measurements, the absolute system of, 22. Electro-motive force, in armature, 80; counter, 37, 141 ; maximum, 56 ; mean, 56 ; the unit of, 31. See also Force. Electro-motor, first, 49. Electro-motors, early, 315 ; ex- periments with, 66 ; govern- ment of, 163, 173. Elwell-Parker dynamo, the, 284. Elwell-Parker motors, the, 287, 288. Energy given out, 93 ; lost, 45 ; mechanical and electrical, 11 ; transmission of, and ideal motor, 35. Equivalent magnetic sheet, the, 28 Erg," the, 22. INDEX. 357 Esson’s experiments, 129. Estimates, comparative, horse and electric traction, 337. Exciting power, 102. Experiments on self-induction, 65 ; with electro-motors, 156. External characteristic, 114. External energy, maximum, 146. Fans and pumps, electric, 318, 320. Feeders, 236. Field of dynamo, 74. Field, the magnetic, 19 ; strength of, 107. Field magnets, types of, 100. Eiguier’s explanation, 314. First electro-motor, 49, 315. Fontaine, M. Hippolyte, his dis- covery, 312. Forbes’, Professor, dynamo, 51. Force, current and mechanical, 33 ; lines of, 11 ; static, dyna- mic, and counter-electro-motive, 141. See also Electro-motive force, &c. Formulas, for dynamos and mo- tors, 98 ; for maximum, 210 ; for mean current, 209 ; for strength of field, 108. Four-pole machine, 286. Friction, mechanical and mag- netic, 95. Fundamental units, 21. General principles, 10. Generator, the compound machine as, 154. Glynde, the Telpher line at, 329. Goolden dynamo, the, 294 ; motor, 296. Governors, 164. Gram, the, 22. Gramme armature, the, 73, 77. Graphic treatment of problems, 145. Griefs’, Herr, wire table, 224. Griscom, motors, 65. Giilcher dynamo, the, 307. Heating of conductor, 211. Hefner-Alteneck armature, 69. Historical notes, 312. Hopkinson, Edison-, dynamo, the, 262. Horse-power given out, 94 ; curves, 135. Horse traction and electric trac- tion, comparative estimates, 337. Hotel de Ville, Paris, ventilation at, 319. Houston, Thomson-, dynamo, the, 263. Hughes’, Professor, theory of magnetism. 16. Hydraulic transmission, 249. Ideal alternating current dynamo, 54 ; continuous current dynamo, 59. Ideal motor and transmission of energy, 35. Immisch motor, the, 272. Induction. See Self-induction. Installations, list of, 345. Insulation, relative importance of, 215. Insulators, 217. Internal characteristic, 114; losses,, 97. Joints of wire, 219. Junction safety catch -box, the,. 236. Kapp dynamo, the, 300. Kriegstetten-Solothurn installa- tion, the, 342 ; experiments, 348. Lead-covered cables, 239. Leakage, loss of current by, 178 ; and economical speed, IHl. Lescuyer’s, Gerard-, electro-dyna- mic paradox, 159. Lightning, protection from, 227. Line, the, 201. Lines, aerial and other, 215, 229. Lines of force, 11 ; unit lines, 19. Long distance transmission, 243. Loss of current by leakage, 178. 358 INDEX, Losses due to mechanical and magnetic friction, 95 ; internal, 97. Machines, electric, see Dynamo, &c. ; two series, 193. Magnets, saturated, 104. Magnetic circuit, 101 ; field, the, 19 ; lines of force, 12 ; moment, the, 28 ; permeability, 108 ; re- sistance, 103; specific magnetic resistance of air, 108. Magnetic and mechanical friction, losses due to, 95. Magnetic shell, the equivalent, 28. Magnetism, modern theory of, 16. Magnets, field, 100 ; single and double, 112. Mains, Edison, 230. Manchester dynamo, the, 282. Maximum current, formula for, 209. Mean current, formula for, 210. Mean electro -motive force, 56. Measurements, the absolute sys- tem of electro-magnetic, 22. Mechanical and electrical energy, relations between, 11. Mechanical and magnetic friction, losses due to, 95. Moment, the magnetic, 28. Motor, characteristic, 126 ; com- l^ound dynamo used as, 171 ; for constant current made self- regulating, 167 ; the Elwell- Parker, 287 ; ideal, 35 ; the Goolden, 296 ; the Immisch, 272 ; the Reckenzaun, 310 ; self- regulating constant current, 175; series, characteristics of, 137 ; speed, characteristics of, 139 ; series- wound, relation be- tween speed and current in, 141. Motors, horse-power of, 94 ; shunt, 153; self -regulating, 166; theory of, 85. Motors and dynamos, different conditions in, 85 ; formulas for, 98. Natural sources of power, 1. Non-polar dynamo, Eorbes’, 53. Oerlikon Works, 341. Ohm, the, 32. Ohm’s law, 115. Pacinotti armature, the, 73, 77, 130. Page’s electric railway, 315. Parker, Elwell-, dynamo, the, 284 ; motor, the, 287. Permeability, magnetic, 108. Periodic governor, the, 164. Phoenix dynamo, the, 130, 297. Pinkus, the early patent of, 314. Plant, cost of, 246. Ploughing by electricity at Ser- maize, 316. Pneumatic transmission, 251. Pole, unit, 21. Potential, 31. Power, cost of, 245. Practical conclusions, 258 ; diffi- culty, 158 ; examples, 150, 191 ; units, 47. Principles, general, 10. Problems, graphic treatment of, 145. Pumping by electricity, 320. Railway, Page’s electric, 315. Railway tunnelling, Argentine and Chili, 344. Railways, electric. Dr. Adams on, 314 ; modern, 321. Reckenzaun motor, the, 144, 309. Reckenzaun’s electric tramcar, 143, 328. Regulation, self, difficulty of, 173. Resistance, magnetic, 103. Reversibility of dynamo machines, 84. Second, the, 22. Self-induction, 64, 143 ; experi- ments on, 65. Self-regulation, difficulty of, 173. I Sermaize, ploughing by electricity I at, 316. INDEX, 359 Shell, the equivalent magnetic, 28. Shunt motor, 153 ; experiment with, 156 ; variation of speed in, 153. Shuttle-wound armature, Sie- mens’, 61. Siemens’ shuttle- wound armature, 61. Silicon bronze wires, 221 ; copper wires, 221. Sliding contact, uncertainty of the, 327. Smith’s, Holroyd, underground conductor, 326. Solenoid, 104. Spasmodic governor, the, 164. Speed, characteristics, 138 ; de- termination of best, for maxi- mum commercial efficiency, 147 ; regulation of, 142 ; torque in- dependent of, 91 ; variation of in shunt motors, 153. Starting power, 41. Static, dynamic, and counter- electro-motive force, 126, 141. Street railroads in the United States, 339. Strength of field, formulas for, 107. Switzerland, development of elec- tric transmission in, 341. Systems, classification of, accord- ing to source of electricity, 160 ; of electric railways, 321 ; of electric transmission of energv, 161, 241. Transmission of energy the fun- damental problem of mechani- cal engineering, 1. Tables’ of efficiency and cost, 237, 254, 311 ; of temperature, 213 ; for finding the most economical size of conductor, 207 ; of results of Kriegstetten-Solothurn plant, 350. Telpher line, the, at Glynde, 329. Telpherage, 329. Temperature, table for rise of, 213. Theory of motors, 85. Thomson-Houston dynamo, the, 263 ; lightning protector, the, 228. Thomson, Sir William, his law, 203. Thompson’s Silvanus, problem, 145. Three-wire system, the, 233. Torque, the, 41 ; exerted by armature, 89 ; independent of speed, 91. Traction, horse and electric, comparative estimates, 337. Tramcar, Keckenzaun’s electric, 332. Tramcars, electric, application to, 143. Transformer, continuous current, 279. Transmission of energy, electric, possible applications of, 241 ; the fundamental problem of mechanical engineering, 1. Transmission, between two dis- tant points, 160 ; circuits for, 215 ; at constant current, 163 ; at constant pressure, 163 ; at constant speed, 153 ; and ideal motor, 35 ; hydraulic, 249 ; ideal system of, 41 ; over large areas, 161 ; long distance, 245, 342 ; pneumatic, 251 ; systems of, 242 ; tunnelling, 344 ; wire- rope, 253. Trotter, Goolden-, dynamo, the, 294. Tubes, Edison’s electric, 237. Tunnelling, railway, 344. Two series machines, 193. Underground conduits, various systems of, 237. Underground lines, 229. Uniform magnetic field, a, 23. Uni-polar dynamos, 53. Unit of electro -motive force, the, 31. Unit lines, 19 ; current, 28, 29. Unit pole, 21. Units, fundamental, 21 ; practical, 47. ^60 INDEX. Variable current/ correction for, 209. Ventilating by electricity, 318. Victoria dynamo, the, 306. Volk’s, Magnus, electric railway at Brighton, 332. Warning apparatus, 319. Waste of energy and capital out- lay, relation between, 201. Wire-rope transmission, 253. Wires, 221 ; joints of, 219 ; carry- ing capacity and heating of, 213. Work done in England and abroad, character of, 340. Zacharias’, Herr, comparative estimates for horse and electric traction, 337. CHISWICK PRESS I— C. WHITTINGHAM AND CO., TOOKS COURT, CHANCERY LANE. 2 , White Hart Street, Paternoster Square, E.C. WHITTAKER & CO.'S LIST OF Classical, Ctmcational, anS Cetlinlcal Morhs. CONTENTS. Atlases 8 Arithmetic and Euclid .... 24 Barrere’s French Course ... 15 Bibliotheca Classica 5 Cambridge Greek and Latin Texts 7 Cambridge Texts with Notes . . 8 Delille’s French Course .... 15 French 15 Series 16 Classics 17 Modern Authors .... 16 German 1 1 Classics 14 Modern German Authors . 12 Grammar School Classics ... 6 Greek Classics 5-9 Historical and English Class Books 21 Italian 19 Lange’s German Course ... .11 Classics 5-10 Modern Languages it Miniature Reference Library . . 30 Miscellaneous Educational Books 25 Ollendorff’s Methods 20 Pinnock’s Catechisms .... 22 Science 26 Shakespeare 25 Spanish 19 Specialist’s Series 27 Technological Works 29 Theological 32 Whittaker’s Library of Arts. etc. . 28 October^ 1890. 2 Whittaker and Co! s List of RECENT & FORTHCOMING PUBLICATIONS. A SPANISH-ENGLISH TECHNOLOGICAL DICTIONARY, containing terms employed in the Applied Sciences, Industrial Arts, Agriculture, etc., etc. By Ponce de Leon. [/w the press, A MANUAL OF WOOD-CARVING: By C. G. Leland, M.A. Revised by John J. Holtzapffel, Assoc. M. Inst. C.E. With numerous illustrations. 5 ^. METAL TURNING. By the author of ^‘Iron- Founding, Pattern-Making,” etc., etc. \ln the press. URSULE MIROUET. By Balzac. An edition specially prepared for use in Schools. With Notes, by J. Boielle, Assistant- Master, Dulwich College. 35. EKKEHARD. By J. V. von Scheffel. (Abridged Edition.) Edited with Notes, etc., by Herman Hager, Ph.D. [Immediately. ADVANCED GERMAN READER. Pieces from the best writers in Prose and Verse, selected by Dr. F. Lange, of R. M. A., Woolwich; with Notes and a Vocabulary by J. F. Davis, D.Lit. London, ^s. [Just published. Tnsi Published^ price 3 ^. nett. AN INTERNATIONAL IDIOM. A Manual of the Oregon Trade Language, or “Chinook Jargon.” By Horatio Hale, M.A., F.R.S.C., Member of the American Philosophical Society ; of the Anthropological Institute of Great Britain and Ireland ; of the Anthropological Societieso” Vienna and Washington ; of the Canadian Institute, etc. Author of “ Ethnography and Philology of the U.S. Exploring Expedition,” “ The Iroquois Book of Rites,” etc, Classical and Educational Works. 3 HOW TO PATENT AN INVENTION. By T. Eustace Smith, Barrister-at-Law. Third edition^ revised and enlarged. 2 s. 6 d. nett. THE WORKING AND MANAGEMENT OF AN ENGLISH RAILWAY. By George Findlay, General Manager of the London and North-Western Railway. With numerous Illustrations. Second Edition^ revised. Crown 8vo. ^s. 6 d. Popular Edition, 2 s. 6 d. MOROCCO. Journeys to the Kingdom of Fez and to the Court of Mulai Hassan. With Itineraries constructed by the Author, and a Bibliography of Morocco from 1844 to 18B7. By H. M. P. de la Martiniere, F.R.G. S., Member of the Geographical Society of France, and the Topographical Society of France. With a Preface by Lieutenant-Colonel Trotter, 93rd Highlanders. Demy 8vo, with numerous Maps, Plans, etc. 143-. THE CHEMICAL ANALYSIS OF IRON. A Complete Account of all the Best Known Methods for the Analysis of Iron, Steel, Ores, etc. By A. A. Blair, Chief Chemist, U.S. Geological Survey, etc. Royal 8vo. 143-. I voL demy \to. with 25 Double and 40 Single Plates, £2 I 03 -. TECHNICAL SCHOOL AND COLLEGE BUILDING. Being a Treatise on the Design and Construc- tion of Applied Science and Art Buildings, and their suitable Fittings and Sanitation. With a chapter on Technical Educa- tion. By Edward^Cook worthy Robins, F.S.A. Outline of Contents. — Introduction — English and Foreign Technical Education — Analysis of the Second Report of the Royal Commissioners on Technical Education — Buildings for Applied Science and Art Instruction, with examples of Foreign and English Buildings— Analysis of the Fittings necessary for these Buildings — British and Foreign Examples of the Details of the Fittings — Heat- ing and Ventilation generally — Heating and Ventilation necessary for Applied Science and Instruction Buildings — The Planning of Buildings for Middle Class Education — Sanitary Science — Appendix. Full prospectus post free on application. “It will prove an indispensable work of reference to architects, builders, and managers of technical —Spectator. “ A most valuable contribution to architectural literature .” — British Architect. 4 Whittaker and Co.’s List of PRACTICAL EDUCATION. A work on Pre- paring the Memory, Developing Quickness of Perception, and Training the Constructive Faculties. By Charles G. Leland, Author of Tv^elve Manuals of Art Work,” “The Album of Repousse Work,” “Hints on Self-Education,” etc. Third Edition, crown 8vo. 6 s. RECORDS OF THE MANOR, PARISH, AND BOROUGH OF HAMPSTEAD, in the County of London, to December 31st, 1889. With Maps and Illustra- tions. By F. E. Baines, C.B. 8vo, 24.S. nett. *** A few copies only remain. MACKAY (C.) SELECTED POEMS AND SONGS OF CHARLES MACKAY, LL.D. With a Commendatory and Critical Introduction by Eminent Writers. Wide foolscap 8vo, half cloth boards, is. 6 d. Sewed, is. A DICTIONARY OF LOWLAND SCOTCH. By the late Charles Mackay, LL.D. With an Introductory Chapter on the Poetry, Humour, and Literary History of the Scottish Language, and an Appendix of Scottish Proverbs. Large post 8vo. cloth, *]s. 6 d. half bound, Ss. 6 d. SOBRIQUETS AND NICKNAMES. By Albert R. Frey. Large post 8vo, cloth, ys. 6 d, half bound, 9^. ARGOT AND SLANG. A new French and English Dictionary of the Cant Words, Quaint Expressions, Slang Terms, and Flash Phrases used in the high and low life of old and new Paris. By Albert Barrere, Officier de ITnstruction Publique. New and Revised Edition^ large post 8vo. loj. 6 d. First and only Complete Edition.^ limited to 675 copies.^ which will he sold to Subscribers only. SLANG, JARGON, AND CANT. A Dictionary of Unconventional Phraseology, comprising English, American, Colonial, Tinkers’, Yiddish, Pidgin and Anglo-Indian Slang ; with Philological Notes and Illustrative Quotations. Compiled and edited by Prof. Albert Barrere, R.M.A. Woolwich, and Charles G. Leland, M.A., Hon. F.R.S.L., Author of “The English Gypsies and their Language,” etc., “ Hans Breitmann,” etc. In co-operation with Dr. Charles Mackay, LL.D., Prof. Douglas B. W. Sladen, Mr. J. Coleman, and other authorities. A few copies of the above remain, for which applications will be received. Price, 2 vols. Roxburgh, qj. ; vellum, y^s. Classical and Educational Works. 5 EDUCATIONAL WORKS. IBifiliotbeca (!Dlas0ica. A Series of Greek and Latin Authors^ with English Notes ^ edited by eminent Scholars. Svo. iESCHYLUS. By F. A. Paley, M.A. CICERO’S ORATIONS. By G. Long, M.A. 4 vols. 12S. od. DEMOSTHENES. By R. Whiston, M.A. 2 vols. Ss. each. EURIPIDES. By F. A. Paley, M.A. 3 vols. HERODOTUS. By Rev. J. W. Blakesley,B. D. 2 vols. 1 2 j*. HESIOD. By F. A. Paley, M.A. 55*. HOMER. By F. A. Paley, M.A. 2 vols. 14^. HORACE. By Rev. A. J. Macleane, M.A. Ss. JUVENAL AND PERSIUS. By Rev. A. J. Macleane, M. A. 6s. LUCAN. The Pharsalia. By C. E. Haskins, M.A., and W. E. Heitland, M.A. 14s. PLATO. By W. H. Thompson, D.D. 2 vols. 10s. SOPHOCLES. Vol. I. By Rev. F. H. Blaydes, M.A. 8^. Vol. II. Philoctetes — Electra — Ajax and Tra- chiniae. By F. A. Paley, M.A. 6s. TACITUS : The Annals. By the Rev. P. Frost. TERENCE. By E. St. J. Parry, M.A. VIRGIL. By J. Conington, M.A. 3 vols. loi'. 6 d. each. *** In some cases the volumes cannot be sold separately. The few copies that remain are reserved for complete sets, which may be obtained, at present, for 9/. 6 Whittaker and Co.’s List of (Grammar ^cf)ooI Cla0sic0, A Series of Greek and Latin Authors^ with English Notes, Fcap, %vo. Ci^SAR: DE BELLO GALLICO. By George Long, M.A. 4s. Books I. -III., for Junior Classes, u. 6d. ; Books IV. and V. in i vok, is. 6d. ; Books VI. and VII. in I vok, IJ. 6d. CATULLUS, TIBULLUS, AND PROPER- TIUS. Selected Poems. With Life. By Rev. A. H. Wratislaw. 2s. 6d. CICERO : DE SENECTUTE, DE AMICITIA, and SELECT EPISTLES. By George Long, M.A. 3^. CORNELIUS NEPOS. By Rev. J. F. Mac- michael. 2s. HOMER: ILIAD. Books I. -XII. By F. A. Paley, M.A. 4s. 6ct. Books I.-VL, 2j. 6d. ; Books VII. -XII., 2j. 6d. HORACE. With Life. By .A. J. Macleane, M.A. 3^. 6d. In 2 Parts : Odes, 2s. ; Satires and Epistles, 2s. JUVENAL: SIXTEEN SATIRES. By H. Prior, M. A. ^s. 6ct. MARTIAL: SELECT EPIGRAMS. With Life. By F. A. Paley, M.A. 4^-. 6d. OVID : The FASTI. By F. A. Paley, M.A. si*. 6d. Books I. and II. in i vok, is. 6d. ; Books III. and IV. in i vok, is. 6d . ; Books V. and VI. in i vok, is. 6d. SALLUST: CATILINA and JUGURTHA. With Life. By G. Long, M.A., andj. G. Frazer. 3^. 6d. Catilina, 2s. Jugurtha, 2s. TACITUS: GERMANIA and AGRICOLA. By Rev. P. Frost. 2s. 6d. VIRGIL : BUCOLICS, GEORGICS, and ^NEID, Books I. -IV. Abridged from Professor Conington’s edition. 4s. 6d. /ENEID, Books V.-XII., 4^. Also in separate volumes.^ is. 6d. each. Classical and Educational Works, ; XENOPHON: The ANABASIS. With Life. By Rev. J. F. Macmichael. ^s. 6d. Also in 4 separate volumes^ u. each. The CYROP^DIA. By G. M. Gorham, M.A. 3^. (id. Books I. and II. in i vol., is. 6d. ; Books V. and VI. in i vol., IS. 6d. MEMORABILIA. By Percival Frost, M.A. 35. A GRAMMAR-SCHOOL ATLAS OF CLAS- SICAL GEOGRAPHY, containing Ten selected Maps. Imperial 8vo. 3^. Uniform with the Series. THE NEW TESTAMENT, in Greek. With English Notes, &c. By Rev. J. F. Macmichael. 4s. 6d. Separate parts, St. Matthew, St. Mark, St. Luke, St. John, Acts, 6d. each, sewed. Camtititige anD Latin Cert0. These Texts, which are clearly printed at the Cambridge Univer- sity Press, on good paper, and bound in a handy form, have been reduced in price, and will now meet the requirements of masters who wish to use Text and Notes separately. iESCHYLUS. By F. A. Paley, M.A. 2s. CiESAR: DE BELLO GALLICO. By G. Long, M.A. ij. 6 d. CICERO: DE SENECTUTE et DE AMICITIA, et EPISTOL.E SELECTiE. By G. Long, M.A. IS. 6d. CICERONIS ORATIONES. Vol. I. (in Verrem.) By G. Long, M.A. 2S. 6d. EURIPIDES. By F. A. Paley, M.A. 3 vols., each 2 s. Vol. I. Rhesus — Medea — Hippolytus— Alcestis — Heraclidae — Supplices — Troades — Index. Vol. II. Ion — Helena — An- dromache — Electra — Bacchse — Hecuba — Index. Vol. III. Hercules Furens — Phocnissae — Orestes— Iphigenia in Tauris — Iphigenia in Aulide — Cyclops — Index. HERODOTUS. By J. G. Blakes- ley, B.D. 2 vols., each 2^. 6 d. HOMERI ILIAS. I. -XII. By F. A. Paley, M.A. is. 6 d. HORATIUS. By A. J. Macleane, M.A. IS. 6d. JUVENAL ET PERSIUS. By A. J. Macleane, M.A. u. 6d. LUCRETIUS. By H.A.J.Munro M.A. 2j. SALLUSTI CRISPI CATI- LINA ET JUGURTHA. By G. Long, M.A. u. 6 d. SOPHOCLES. By F. A. Paley M.A. 2s. 6 d. TERENTI COMCEDIiE. ByW. Wagner, Ph.D. 2s. THUCYDIDES. By J. G. Donald- son, D.D. 2 vols., each 2s. VERGILIUS. By J. Conington, M.A. 2S. XENOPHONTIS EXPEDITIO CYRI. By J. F. Macmichael, B. A. IS. 6d. NOVUM TESTAMENTUM GR.^CE. By F. H. Scrivener, M.A. 4^. 6d. An edition with wide margin for notes, half bound, 12s. 8 Whittaker and Co.’s List oj CambriDge Cert 0 toitb H3ote0. Being select portions of the Greek and Latin Authors, edited by eminent scholars, with Introduction and Notes for School Use. The Texts are those of the “ Bibliotheca Classica ” and the “ Grammar School Classics,” and the Notes are printed at the end. Price i^. 6 d. each, with exceptions. AESCHYLUS. 6 Vols. PROME- THEUS VINCTUS-SEPTEM CONTRA THEBAS — AGA- MEMNON - PERSAE — EU- MENIDES— CHOEPHOROE. By F. A. Paley, M.A., LL.D., Late Classical Examiner to the University of London. EURIPIDES. 13 vols. ALCESTIS — MEDEA — HIRPOLYTUS — HECUBA — BACCHAE — ION ( 2 ^.)- 0 RESTES-PH 0 E- NISSAE— TROADES — HER- CULES FURENS— ANDRO- MACHE — IPHIGENIA IN TAURIS — SUPPLICES. By F. A. Paley, M.A., LL.D. HOMER. ILIAD, Book I. (i^.) By F. A. Paley, M.A , LL.D. SOPHOCLES. 5 vols. OEDIPUS TYRANNUS— OEDIPUS CO- LONEUS — ANTIGONE — ELECTRA— AJAX. By F. A. Paley, M.A., LL.D. THUCYDIDES. Book VI. By F. A. Paley, M.A., LL.D. XENpPHON’S ANABASIS. With Life, Itinerary, Index, and Three Maps. Edited by the late J. F. Macraichael. New Edition, revised by J. E. Melhuish, M.A., Assistant Master at St. Paul’s School. In 4 vols. \_Inthe press. Book I. (with Life, Introduction, Itinerary, and 3 Maps) ; Books II. and III.; Books IV. and V.; Books VI. and VII. XENOPHON’S HELLENI -'A. Book II. By L. D. Dowdall, M.A., B.D. CICERO’S DE AMICITIA— DE SENECTUTE- EPISTOLAE SELECT AE. By George Long, M.A. OVID. Selections from the AMORES, TRISTIA, HERO- IDES, and METAMORPHO- SES. By A. J. Macleane, M.A. OVID’S FASTI. By F. A. Paley, M.A., LL.D. Third Edition, revised. In 3 vols. 2j. each. Books I. and II. ; Books III. and IV. ; Books V. and VI. TERENCE. 4 vols. ANDRIA— HAUTON TIMORUMENOS — PHORMIO — ADELPHOE. By Prof. Wagner. VIRGIL’S WORKS. Abridged from Prof. Conington’s Edition by Professors Nettleship and Wagner and Rev. J. G. Shep- pard. In 9 vols. BUCOLICS— GEORGICS, I. and II.— GEORGICS, III. and IV.— AENEID, I. and II.— AENEID, III. and IV. — AENEID,V.andVL— AENEID, VII. and VIIL— AENEID, IX. andX.— AENEID, XL and XII. 3tla0e0. LONG’S ATLAS OF CLASSICAL GEOGRAPHY. Containing Twenty-four Maps. Constructed by William Hughes, P\R.G.S., and Edited by George Long, M.A. New edition, with Coloured Outlines, and an Index of Places. Royal 8vo. 6s. LONG’S GRAMMAR SCHOOL ATLAS OF CLAS- SICAL GEOGRAPHY. Containing Ten Maps, selected from the larger Atlas. Constructed by W. Hughes, F.R.G.S., and edited by George Long, M.A. New editio7t, with Coloured Outlines. Royal 8vo. ^s. Classical and Educational Works. 9 ®reefe Ce0tament0, etc. NOVUM TESTAMENTUM GR^CE. Textus Stephanici, 1550. Accedunt variae Lectiones editionum Bezae, Elzeviri, Lachmanni, Tischendorfii, Tregellesii, curante F. H. Scrivener, M.A. 4s. 6d. An Edition with wide margin for MS. Notes, 4to. half-bound morocco, 12s. Editio Major containing, in addition to the matter in the other Edition, the Capitula {majora et minorei) and the Eusehian Canons, the various Readings of Westcott and Hort, and those adopted by the Revisers ; also a revised and much-enlarged series of References, Small post 8vo. cloth, pp. xvi.-598, *]s. 6d. VALPY’S. For the use of Schools. i2mo. 55. Macmichaers. See Grammar School Classics, p. 7. Crown 2 >vo. cloth. THE GOSPEL OF S. MATTHEW. The Greek Text, with Critical, Grammatical, and Explanatory Notes, &c., by the late Rev. W. Trollope, M.A., re-eclited by the Rev. W. H. Rowlandson, M.A. 5 j. GOSPEL OF S. MARK. The Greek Text, with Critical, Grammatical, and Explanatory Notes, Prolegomena, &c., by Rev. W. H. Rowlandson, M.A. 4s. 6 d. GOSPEL OF S. LUKE. The Greek Text, with Critical, Grammatical, and Explanatory Notes, &c. , by the late Rev. W. Trollope, M.A., revised and re-edited by the Rev. W. H. Rowlandson, M.A. 5^-. ACTS OF THE APOSTLES. The Greek Text, with Critical, Grammatical, and Explanatory Notes, and Exami- nation Questions, by Rev. W. Trollope, M. A., re-edited and revised by the Rev. G. F. Browne, M.A. 5^. DAWSON^S GREEK-ENGLISH LEXICON TO THE NEW TESTAMENT. New Edition, by Dr. Tylor. 8 VO, cloth, gj. A 2 lo Whittaker and Co's List of ILatin anD ®reefe €la00 TBoofe0. PENROSE’S LATIN ELEGIACS. Easy Exercises in Latin Elegiac Verse. By the Rev. John Penrose. New edition. i 2 mo. cloth, 2s. Key to ditto, for Tutors only, 3^*. bd. BEATSON^S GREEK IAMBICS. Progressive Exercises on the Composition of Greek Iambic Verse. By Rev. B. W. Beatson. i 2 mo. cloth, 3 ^. GRADUS AD PARNASSUM; sive novus synony- momm, epithetorum, versuum, ac phrasium poeticarum thesaurus. Ne^v edition. By G. Pyper. i 2 mo. cloth, BY VALPY. Whittaker's Improved edition. Latin and English. New edition. Royal i 2 mo. *]s. 6d. STODDART’S NEW DELECTUS; or, Easy Steps to Latin Construing. For the use of Pupils commencing the Language. Adapted to the best Latin Grammars, with a Dictionary attached. New edition. l 2 mo. ^s. BEDFORD'S PROPRIA QU^ MARIBUS; or, Short Rules for the Genders of Latin Nouns, and a Latin Prosody. i 2 mo. is. BOSSUT’S LATIN WORD BOOK; or, First Step to the Latin Language. l 8 mo. is. LATIN PHRASE BOOK. i8mo. is. OVID AND TIBULLUS. A Selection of Elegiac Extracts (Florilegium Poeticum). With English Notes by the Rev. P. Frost, M.A. 2s. Classical and Educational Works. 1 1 MODERN LANGUAGES. A COMPLETE GERMAN COURSE FOR USE IN PUBLIC SCHOOLS. Edited by F. Lange, Ph.D., Professor, R. M.A., Woolwich, Examiner in German to the College of Preceptors, etc., etc. GERMAN GRAMMAR. With especial reference to Phono- logy, Comparative Philology, English and German Corre- spondences, and Idioms. In three Parts. Part L— ELEMENTARY ACCIDENCE AND SYN- TAX. 2J. Part II. — ACCIDENCE SUPPLEMENTED AND SYNTAX CONTINUED. 2.r. Part III.— A CONCISE BUT COMPLETE SURVEY OF GERMAN GRAMMAR. 3^-. 6d. PROGRESSIVE GERMAN EXAMINATION COURSE. In Three Parts. Comprising the Elements of German Grammar, an Historical Sketch of the Teutonic Languages, English and German Corre- spondences, Materials for Translation, Dictation, Extempore, Conversation and complete Vocabularies 1. ELEMENTARY COURSE. Cloth, 2 l 2. INTERMEDIATE COURSE. Cloth, 2J. 3. ADVANCED COURSE. Second revised edition. Cloth, \s. 6d. ELEMENTARY GERMAN READER. A Graduated Col- lection of Readings in Prose and Poetry. With English Notes and a Vocabulary, is. 6d. ADVANCED GERMAN READER. A Graduated Collection of Readings in Prose and Poetry. With English Notes and a Vocabulary, By F. Lange, Ph.D. and J. F. Davis, M.A., D.Lit. 3J. 12 Whittaker and Co! s List of ^ede0 of ^oDern (German autl)or0. With Introduction and Notes. Edited by F. Lange., Ph.D.., Professor^ Royal Military Academy., Woolwich. The attention of the heads of Colleges and Schools is respectfully directed to this new Series of ‘‘ Modern German Authors ” which is intended to supply the much-felt want of suitable Reading Books for English Students of German who have passed through the preliminary stages of fables and anecdotes. To those who wish to extend their linguistic and grammatical knowledge, these volumes will afford, in one respect, a great advantage over those of an earlier period, presenting, as they do, the compositions of the best living, or only recently deceased authors. The Notes, besides etymological and other explanations, will contain many useful idiomatic expressions suggested by the text, and worth committing to memory. FIRST SERIES. For Beginners. Edited, with a Grammatical Introduction, Notes, and a Vocabulary, by F. Lange, Ph. D. , Professor, R. M. A. Woolwich, Examiner in German to the College of Preceptors, and H. Hager, Ph.D., Examiner in German to the London University. KEY’S FABELN FUR KINDER. Printed in roman cha- racters. Illustrated by O. Speckter. Edited, with an Intro- duction, Grammatical Summary, Words, and a complete Vocabulary. By F. Lange, Ph.D., Professor, is. 6d. The Same, with a Phonetic Introduction, Phonetic Transcription of the Text. By F. Lange, Professor, Ph.D. 2.s. SECOND SERIES. For Intermediate Students. Edited, with a Biogra- phical Introduction, Notes, and a complete vocabulary, by F. Lange, Ph.D., Professor, and H. Hager, Ph.D. DOKTOR WESPE. Lustspiel in fiinf Aufzligen von JULIUS RODERICK BENEDIX. Edited by F. Lange, Ph.D., Professor. 2J. (id. Classical and Educational Works. 13 THIRD SERIES. P'OR Advanced Students. Edited, witli a Literary In- troduction and Notes, by F. Lange, Ph.D., Professor, R. M.A. Woolwich, in co-operation with F. Storr, JhA. ; A. A. Mac- donell, M.A. ; H. Hager, Ph.D. ; C. Nenhaus, Ph.D. and others. MEISTER MARTIN, der Kiifner. Erziihlung von E. T. A. Hoff- man. Edited by F. Lange, Ph.D., Professor, Royal Military Academy, Woolwich, is. 6d. HANS LANGE. Schauspiel von Paul Heyse. Edited by A. A. Macdonell, M.A., Ph.D., Taylorian Teacher, University, Oxford. 2s. AUF WACHE. Novelle von Berthold Auerbach. DER GEFRO- RENE KUSS. Novelle von Otto Roquette. Edited by A. A. Macdonell, M.A. 2s. DER BIBLIOTHEKAR. Lustspiel von G. von Moser. Edited by F. Lange, Ph.D. Second revised Edition. 2s. EINE FRAGE. Idyll von George Ebers. Edited by F. Storr, B. A., Chief Master of Modern Subjects in Merchant Taylor’s School. 2 s. DIE JOURNALISTEN. Lustspiel von Gustav Freytag. Edited by Professor F. Lange, Ph.D. Second revised Edition. 2s. 6d. ZOPF UND SCHWERT. Lustspiel von Karl Gutzkow. Edited by Professor F. Lange, Ph.D. 2s. 6d. GERMAN EPIC TALES IN PROSE. I. Die Nibelungen, von A. F. C. Vilmar. — II. Walther und Hildegund, von Albert Richter. Edited by Karl Neuhaus, Ph.D., the International College, Isleworth. 2s. 6d. EKKEHARD. By J. V. von Scheffel. (Abridged Edition.) Edited by Herman Hager, Ph.D., Lecturer on the German Language and Literature in the Owen’s College, Manchester. ^^Immediately. 14 Whittaker and Co.’s List of GERMAN CLASSICS, WITH ENGLISH NOTES. Fcap. 8vo. GERMAN BALLADS. From Uhland, Goethe, and Schiller. With Introductions to each Poem, copious Notes, and Biographical Notices. By C. Bielefeld, u. GOETHE’S HERMANN AND DOROTHEA. With Short Introduction, Argument, and Notes Critical and Explanatory. By Ernest Bell and E. Wolfel. ij*. 6^. SCHILLER’S MAID OF ORLEANS. With Intro- duction, and Notes. By Dr. Wagner, u. 6 d. MARIA STUART. With Introduction and Notes By V. Kastner, M.A. is. 6 d. WALLENSTEIN. Complete Text. Neiu edition. With Notes, Arguments, and an Historical and Critical Intro- duction. By C. A. Buchheim, Ph.D., 5^. Or separately — Part I.— THE LAGER AND DIE PICCOLOMINI. 2 l 6^. Part II.— WALLENSTEIN’S TOD. 2,f. 6^. MISCELLANEOUS GERMAN CLASS BOOKS. OLLENDORFF’S METHOD OF LEARNING GERMAN. Crown 8vo. yj. Key, 8vo. ^s. {See p. 20.) GRENFELL’S ELEMENTARY GERMAN EX- ERCISES. Part I. Adapted to the Rugby School German Accidence. i2mo. u. (id. SHELDON (E.S.) A SHORT GERMAN GRAM- MAR FOR HIGH SCHOOLS AND COLLEGES. Crown 8vo. 3L CRABB’S EXTRACTS FROM THE BEST GER- MAN AUTHORS. Edited by Emanuel Hausmann, Ph.D. 1 2mo. ^s. WENDEBORN’S GERMAN GRAMMAR. Re- vised by A. Heimann, Ph.D. i2mo. 6 s. FLUGEL’S COMPLETE DICTIONARY OF THE GERMAN AND ENGLISH LANGUAGES. German- English, and English-German. By C. A. Felling, A. Heimann, and J. Oxenford. Revised edition. 2 vols. 8vo. i/. il ABRIDGED EDITION. Carefully compiled from the larger Dictionary. By C. A. Felling and J. Oxenford. New edition. Royal i8mo. 6 s. Classical and Educational Works. 15 JFrencb. FRENCH COURSE. BY PROFESSOR A. BARRERE, R. M.A., Woolwich. JUNIOR GRADUATED FRENCH COURSE, affording materials for Translation, Grammar, and Conversation. Being an introduction to the Graduated French Course. Cloth, IJ-. 6d. ELEMENTS OF FRENCH GRAMMAR AND FIRST STEPS IN IDIOMS. With numerous Exercises and a Vocabulary, being an Introduction to the Precis of Compara- tive French Grammar. Crown 8vo. cloth, 2s. PRECIS OF COMPARATIVE FRENCH GRAMMAR AND IDIOMS, and Guide to Examinations. Cloth. Second edition^ revised^ 3 .?. 6d. RECITS MILITAIRES. Selections from modern French authors, with short biographical introductions in Frencli, and English notes for the use of army students and others. Crown 8 VO. 3«f. DELILLE’S FRENCH COURSE. BEGINNER’S OWN FRENCH BOOK. By C. J. Delille. Being a Practical and Easy Method of Learning the Elements of the French Language. i2mo. cloth, 2s. Key, 2 s^ FRENCH GRAMMAR. In two Parts. I. — Accidence. IT. — Syntax, written in French, with Exercises conducive to the speaking of the French Language, &c. Cheai)er edition. i2mo. 3 J. (id. Key, y. EASY FRENCH POETRY FOR BEGINNERS; or, Short Selections in Verse on a Graduated Plan for the Memory. With English Notes. i2mo. 2 s. MODELES DE POESIE FRANCAIS. With Treatise on French Versification. New edition. i2mo. 3 j*. 6^. REPERTOIRE DES PROSATEURS FRANCAIS. With Biographical Sketches, &c. New edition. i2mo. '^s. 6d. MANUEL ETYMOLOGIQUE ; or, an Interpretative Index of the most recurrent Words in the French Language, i2mo. 2S. (id. i6 Whittaker and Co! s List of 3Hf)ittafecr’s Jfrenc^ ^erie0. F or the use of Schools and Private Students. Edited by A. Barrere, Prof. R.M.A., Woolwich, F.E. A. Gasc. , &c., and others. Each number with a literary Introduction and Arguments in English, foot-notes explaining the more difficult passages, and translations of the idiomatic expressions into the corresponding English idioms. Fcap. 8vo, each number, sewed, 6^/. ; cloth, ^d. SCRIBE. LE VERRE D’EAU. Barrere. MOLIERE. LE BOURGEOIS GENTILHOMME. Gasc. MOLIERE. L’AVARE. Gasc. SOUVESTRE. SOUS LA TONNELLE, I. Desages. GALLAND. ALI BABA. Clare. LAMARTINE. JEANNE D’ARC. Barrere. PIRON. LA METROMANIE. Delbos. RACINE. ANDROMAQUE. Gasc. RACINE. IPHIGENIE. Gasc. MOLIERE. LES PRECIEUSES RIDICULES. Gasc. RACINE. ATHALIE. Gasc. RACINE. LES PLAIDEURS. Gasc. MOLIERE. LE MISANTHROPE. Gasc. CORNEILLE. LE CID. Gasc. RACINE. ESTHER. Gasc. ajQbittafeeP0 j^onern jfrcncl) autf)or0. With Introduction and Notes by J. Boielle, Senior French Master at Dulwich College. For Beginners. LA BELLE NIVERNAISE. Histoire d’un vieux bateau et de son equipage. By Alphonse Daudet. With 6 illustrations, 'is. 6d. For Advajtced Students. BUG JARGAL. By Victor Hugo. 3^-. URSULE MIROUET. By Honorfe de Balzac. 3^*. Others to follow. Classical and Educational Works. 17 jFrencf) Cla00ic0, toitl) €ngli0ft J 13 ote 0 . Fcap, Zvo. cloth, AVENTURES DE TELEMAQUE. Par Fene- Ion. New edition. Edited and revised by C J. Delille. 2s. 6d, HISTOIRE DE CHARLES XII. Par Voltaire. New edition. Edited and revised by L. Direy. \s. 6d. PICCIOLA. Par X. B. Saintine. New edition. Edited and revised by Dr. Dubuc. is. 6d. SELECT FABLES OF LA FONTAINE. New edition. Edited by F. Case, M.A. is. 6d. ^i0cellaneou0 jFrenci) Cla 00 TBoofe0. A PUBLIC EXAMINATION FRENCH READER. With a Vocabulary to every extract, suitable for all Students who are preparing for a French Examination. By A. M. Bower, F.R.G.S., late Master in University College School, &c. Cloth, ^s. 6d. DES CARRIERES^ FRENCH IDIOMATICAL PHRASES AND FAMILIAR DIALOGUES. 35 -. 6d. HISTOIRE DE FRANCE, DEPUIS LETAB- LISSEMENT DE LA MONARCHIE. Continuee jusqu’au retablissement de I’Empire sous Napoleon III., par C. J. Delille. i 2 mo. Js. DUVERGER’S COMPARISON BETWEEN THE IDIOMS, GENIUS, AND PHRASEOLOGY OF THE FRENCH AND ENGLISH LANGUAGES. New edition. i 2 mo. 4 -s. 6d. ABREGE DE L’HISTOIRE DE GIL BLAS de Santillane, de Le Sage. By Nicholas Wanostroct. Revised by J. C. Tarver. i 2 mo. 5 ^-. i8 Whittaker and Co.'s List of HAMEL’S NEW UNIVERSAL FRENCH GRAM- MAR. Nnv Edition. i2mo. 4^. GRAMMATICAL EXERCISES UPON THE FRENCH LANGUAGE. Nei.v edition. i2mo. 4s. Key, ^s, FRENCH GRAMMAR AND EXERCISES. New edition. i2nio. 6d. Key, 4s. LEVIZAC’S DICTIONARY OF THE FRENCH AND ENGLISH LANGUAGES. Edited by N. Lambert i2mo. 6s. 6d. NUGENT^S POCKET DICTIONARY OF THE FRENCH AND ENGLISH LANGUAGES. New edition, revised by J. C. J. Tarver. Pearl edition, 4s. 6d. WANOSTROCHT’S FRENCH GRAMMAR. Revised and enlarged by J. C. Tarver. i2mo. 4s. RECUEIL CHOISI. Edited by C. J. Delille. i2mo. 3^-. PERRIN’S FRENCH FABLES. i2mo. 2.5*. MALAHER. THE IDIOMS ; or, Exercises on the Difficulties of the French Language. By L. Malaher, B.A. i2mo. 3J. ; with Key, ^s. OLLENDORFF’S METHOD OF LEARNING FRENCH. {Seep. 20.) i2mo. 6s. 6d. Key, 8vo. *]s. BOSSUT’S FRENCH WORD BOOK. i 8 mo. i.r. FRENCH PHRASE BOOK. i 8 mo. li*. CHAMBAUD’S FRENCH FABLES. For the use of Children. i2mo. 2s. CAMPAN. CONVERSATIONS D’UNE MERE AVEC SA FILLE, ETC., A L’USAGE DES DEMOI- SELLES ANGLAISES. Par Mme. D. Cloth, 3,1-. M. Classical and Educational Works. 19 3[talian. BARETTFS DICTIONARY OF THE ENGLISH AND ITALIAN LANGUAGES. To which is prefixed an Italian and English Grammar. New Edition. By G. Comelati and J. Davenport. 2 vols. 8vo. i/. it. GRAGLIA’S POCKET DICTIONARY OF THE ITALIAN AND ENGLISH LANGUAGES. With con- siderable Additions, and an Elementary Italian Grammar. i8mo. 4.S. 6d. OLLENDORFF’S METHOD OF LEARNING ITALIAN. (See p. 20.) i2mo. 7 t. Key, 8vo. yj*. SOAVE’S NOVELLE MORALI. i2mo. 4s. VENERONI’S COMPLETE ITALIAN GRAM- MAR. By P. Rosteri. l2mo. 6s. VERGANI AND PIRANESI’S ITALIAN AND ENGLISH GRAMMAR. With Exercises, &c. By J. Guichet. edition, by Signor A. Tommasi. i2mo. 5^. Key, 3.f. ^panisL NEUMAN AND BARETTFS SPANISH AND ENGLISH, AND ENGLISH AND SPANISH DIC- TIONARY. Revised and enlarged by M. Seoane, M.D. 2 vols. 8 VO. i/. 8 t. POCKET DICTIONARY. Spanish and Eng- lish, and English and Spanish. Compiled from the larger work. i8mo. 5'^* OLLENDORFF’S (DR. H. G.) METHOD OF LEARNING SPANISH. New edition. 8vo. I2t. Key, 8vo. 7 t. PONCE DE LEON’S ENGLISH - SPANISH TECHNOLOGICAL DICTIONARY. 8vo. £\ i6t. 20 Whittaker and Co.^s List of practical mercantile Correspontience. A Collection of Commercial Letters and Forms, with Notes, Explanatory and Grammatical, and a Vocabulary of Commercial Terms, edited by L. Simon, Chr. Vogel, Ph.D., II. P. Skelton, W. C. Wrankmore, Leland Mason, and others. Intended as Class Books for Schools and for Self-Instruction. Now Ready ^ crown 8z^o, cloth : ENGLISH, with German Notes, 3i'. GERMAN, with English Notes, 3 j. ENGLISH, with French Notes, 4s. 6d. FRENCH, with English Notes, 4^-. 6d. This new Collection of Model Letters and Epistolary Forms embraces the whole sphere of Commercial Transactions. Each example is provided with such remarks and explanations, that any one with a fair grammatical knowledge of the particular language will find it an easy matter to prepare a well-expressed letter. DR. H. G. OLLENDORFF^S WORKS. A NEW METHOD OF LEARNING TO READ, WRITE, AND SPEAK A LANGUAGE IN SIX MONTHS. Adapted to the French, German, Italian, and Spanish. The utility of the following well-known books has been abundantly proved by an extensive sale for many years past — which still con- tinues. They are specially intended for the use of schools and private teachers ; but as each volume has a Key they are no less adapted for the use of students who have not the advantage of a teacher. OLLENDORFF’S METHOD, adapted to the French. New edition, carefully revised and augmented, with a complete treatise on the gender of French substantives and an additional treatise on the French verbs. i2mo. 6^. 6d. Key, ys. Adapted to the German. New edition, revised and cor- rected. Crown 8vo. ys. Key, ys. Adapted to the Italian. Seventh edition, carefully revised and augmented. i2mo. ys. Key, ys. Adapted to the .Spanish. Sixth edition. 8vo. 12^. Key, 71. Classical and Educational Works, 21 lj)i 0 toncal atiD Cnglisb Class T5oo6s. ANGLO-SAXON CHRONICLES, 800-1001 a.d. Edited, with Introduction, Notes, and a Complete Glossary, by J. F. Davis, D.Lit., M.A. Crown 8vo. 4^’. (id. KEIGHTLEY’S MYTHOLOGY OF ANCIENT GREECE AND ITALY, for the use of Schools. Illustrated. i8mo. will and cheaper edition, 2 s, ELEMENTARY HISTORY OF ROME. i8mo. loth edition, is. 6 d, ELEMENTARY HISTORY OF GREECE. i8mo. ^th edition, u. (id. ELEMENTARY HISTORY OF ENGLAND. i2mo. 2j. HISTORY OF ROME. i2mo. c^th edition. 3^. (d, HISTORY OF THE ROMAN EMPIRE. i2ino. 3^. (id, HISTORY OF GREECE. i2mo. Zth edition, y, (d. HISTORY OF ENGLAND. 2 vols., i2mo. ^s. CHEPMELL’S (REV. DR.) SHORT COURSE OF GRECIAN, ROMAN, AND ENGLISH HISTORY. New edition. i2mo. 5^. Questions on, i2mo. is. SHUMWAY(E.S.) A DAY IN ANCIENT ROME. With numerous Illustrations. By Edgar S. Shumway, Pro- fessor, Rutger’s College, New Brunswick. Small 4to. cloth, 5^. PINNOCK^S HISTORY OF ENGLAND. From the Invasion of Julius Caesar. With a Biographical and His- torical Dictionary. Questions for Examination, Genealogical Tables, Progress of Literature and the Constitution, &c. Illus- ted. Continued by the Rev. W. H. Pinnock, LL.D. New edition. i2mo. 6 s. HISTORY OF GREECE. With an Introduc- tion on the Natural and Political Geography of Greece, Dic- tionary of Difficult Terms, Questions for Examination, etc. Illustrated. Edited by Dr. W. C. Taylor. i2mo. 5^-. 6 d. HISTORY OF ROME. With an Introduc- tion, the Geography of the Roman Empire, Notices of the Roman Manners, Questions for Examination, Chronological Index. &c Illustrated. Edited by Dr. W. C. Taylor. i2mo. 5^. 6 d. 22 Whittaker and Co! s List of SCHOOL AND UNIVERSITY ANALYSES. By the Rev. Dr. Pinnock. AN ANALYSIS OF SCRIPTURE HIS- TORY ; Intended for Readers of Old Testament History, and the University Examinations; with Maps, Copious Index, and Examination Questions. i8mo. cloth, ^s. 6d. AN ANALYSIS OF NEW TESTAMENT HISTORY ; Embracing the Criticism and Interpretation of the original Text; with Questions for Examination. i8mo. cloth, 4^^. AN ANALYSIS OF ECCLESIASTICAL HISTORY ; From the Birth of Christ, to the Council of Nice, A.D. 325. With Examination Questions. i8mo. cloth, 3 ^. 6d. ANALYSIS OF ENGLISH CHURCH HIS- TORY ; comprising the Reformation period, and subsequent events ; with Questions of Examination, especially intended for the Universities and Divinity Students in general. i8mo. cloth, 4 ^“. 6d. A SHORT ANALYSIS OF OLD TESTA- MENT HISTORY. With Questions for Schools. i8mo. cloth, IS. 6d. A SHORT ANALYSIS OF NEW TESTA- MENT HISTORY. With Questions for Schools. i8mo. cloth, IS. 6d. PINNOCK’S CATECHISMS OF THE ARTS, SCIENCES, AND LITERATURE Whittaker’s Improved Editions. Illustrated with Maps, Plates, and Woodcuts, care- fully re-edited. i8mo. price 9^. each. HISTORY. — Modern — Ancient — Universal — Bible and Gos- pel — Scripture — Chronology — England — Scotland — France — America — Rome — Greece — ews. GEOGRAPHY. — Ancient — Modern, Improved Edition — Modem, Original Edition — Sacred — England and Wales — Use of the Globes. GRAMMAR — English — French — German — Italian — Latin — Spanish — Greek : Part I. Accidence. Part II. Syntax and Prosody — Hebrew. Classical and Educational Works. 23 PINNOCK’S CATECHISMS {Continued) : MATHEMATICS, &c. — Algebra (two Parts) — Arithmetic — Geometry — N avigation — Land Surveying RELIGION. — Religion — Natural Theology — Scripture His- tory — Bible and Gospel History. FINE ARTS, &c. — Architecture — Drawing — Perspective — Music — Singing. LITERATURE. — Mythology — Rhetoric — Logic — British Biography — Classical Biography. MISCELLANEOUS. — First Catechism — General Know- ledge — Intellectual Philosophy — Agriculture — English Law — Heraldry — Medicine — Moral and Social Duties — Trade and Commence. WATTON’S ORIGINAL AIDS TO EDUCATION. Hand-series of Tablets, Leading Events of General His- tory. Chief Events of Old Testament History. Chief Events of New Testament History. Prophecies and other Scripture Subjects. Chief Events of Grecian History. Chief Events of Roman History. Chief Events of Eastern Empire. Chief Events of German History. Chief Events of English History. Chief English Battles and Results. Chief Events of Scottish History. Chief Events of French History. Chief Events of Prussian History. in Stiff Covers, '^d. each. Chief Events of Russian History. Eminent Men of Modern Times. Chief Events of Church History. Natural System of Botany. The Linnaean System of Botany. Natural History — Zoology. Natural Philosophy. Principles of Grammatical Analy- sis, with Examples. Guide to English Parsing, with Examples. Abstract of Heathen Mythology. Word Formation — Saxon, La- tin, and Greek Prefixes, with Examples. Chief Grecian and Roman Battles and Results. WATTON’S SKELETON EXERCISE BOOKS For History, Geography, Biography, Analysis, Parsing, and Chro- nology, with Script Headings and Specimen Page. Price ij-. and 2 s. each. Biographical Exercise Book, 2 Series, each \s. Charts sys- tematically arranged with date words, 60 pages, cloth, \s. Selected Descriptive Poetry, is. Object Lessons, Nos. i, 2, 3, and 4, 32 pp., in stiff covers, 2 d. each. Large Type Series of Tablets (20 by 23 inches), embracing His- torical, Geographical, and odier Subjects, 4d. each, for suspension. WATT’S SCRIPTURE HISTORY. From the Creation to the time of Christ. 121110. 3J. 6 d. 24 Whittaker and Co.’s List of aritljmetic anD CucliD. PIN NOCK'S ARITHMETICAL TABLES OF MONEY, WEIGHTS, AND MEASURES. With Questions for Examination, and Explanatory Notes, &c. i8mo. '^d. FIRST CIPHERING BOOK. Containing Easy Exercises in the First Rules of Arithmetic. 4to. sewed, u. RYAN’S CIVIL SERVICE ARITHMETICAL EX- AMINATION PAPERS, ByL.J.Ryan. Cloth, 2J. Key, it. 6^/. SONNENSCHEIN AND NESBITT’S ARITHMETIC. The Science and Art of Arithmetic for the use of Schools. Post 8vo. $s. 6d. Or separately, Part I. — Integral. 2J. (yd. Parts II. and III. — Fractional and Approxi- mate Calculations, ^s. 6d, Answers to the Exercises, it. 6d. Exercises separately. Part I. it. Parts II. and III. it. ^ d . A B C OF ARITHMETIC. Teacher’s Book, Nos. I and 2, each it. Exercise Book, Nos. i and 2, each ^d. WALKINGAME’S TUTOR ASSISTANT (FRASER’S). Being a Compendium of Arithmetic. i2mo. 2T. Key, 3T. SAUL’S ARITHMETIC. Revised by S. Maynard. 25 “. EUCLID, THE FIRST BOOK. With an Introduc- tion and Collection of Problems for the use of Schools. By J. M. Wilson, M.A. ind edition. 4to. 2t. THE FIRST SIX BOOKS, together with the ELEVENTH and TWELFTPI. From the Text of Dr. Simson. New editiony revised by S. Maynard. i8mo. 4 t. PLAYFAIR’S ELEMENTS OF GEOMETRY, comprising the first six books of Euclid, with the elements of Plane and Spherical Trigonometry. 14//^ edition. By Rev. P. Kelland, F.R.S. 3 t. (yd. Classical and Educational Works. 25 ^i0cellaneous oBDucational TBoofeg. SHAKESPEARE'S PLAYS, with Text and Intro- duction in English and German. Edited by C. Sachs, Prof. Ph D 8vo. cloth, each Play or Number, lod. Now Ready : 10. King Richard II. 11. King Henry IV. I. 12. ,, „ II. 13. King Henry V. 14. King Richard HI. 15. Cymbeline. 16. Coriolanus. 17. Antony and Cleopatra. 18. Merchant of Venice. 19. Much Ado about Nothing. Julius Caesar. Romeo and Juliet. King Henry VIII. King Lear. Othello. 6. Hamlet. 7. A Midsummer Night’s Dream. 8. Macbeth. 9. King J ohn. Others to follow, “This edition will be quite a godsend to grown-up students of either language, for the ordinary class reading books are too childish to arrest their attention. The parallel paging saves the labour of using a dictionary, and the series is so low in price as to place it within the reach of all.” Saturday Review. SHAKESPEARE REPRINTS. i. King Lear. Parallel Texts of Quarto i and Folio i. Edited by Dr. W. Vietor, of Marburg. Square i6mo. cloth, 3 j-. (id. The texts of the first quarto and folio ^ with collations from the later quartos and folios, are here printed in a compact and con- venient volume, intended as a class-book in the University. SEIDEL (ROBT.) INDUSTRIAL EDUCATION : A Pedagogic and Social Necessity. Crown 8vo. 4^. WOODWARD (C.M.) THE MANUAL TRAIN- ING SCHOOL, ITS AIMS, METHODS, AND RESULTS. With Figured Drawings of Shop Exercises in Woods and Metals. 8vo. lOJ-. BIBLIOGRAPHY OF EDUCATION. Hints toward a Seleet and Descriptive Bibliography of Education. Arranged by Topics, and Indexed by Authors. By G. Stanley Hall and J. M. Mansfield. Post 8vo. cloth, qs. 6d. DRAWING AND DESIGNING: in a series of lessons. By Charles G. Leland, M.A., F. R. S. Fcap. 4to, is. sewed, is. 6d. cloth. THE THEORY OF MUSIC. Applied to the Pianoforte. By Louisa Gibson. In three Books. Price is. sewed, is. 6d. cloth, each part. 26 Whittaker and Co! s List of MURRAY’S ENGLISH GRAMMAR. Abridged. 138M edition, is, BEDFORD’S QUESTIONS FOR JUNIOR CLASSES. Compiled by the Rev. J. G. Bedford. i2mo. 2s, 6d, BLAIR’S UNIVERSAL PRECEPTOR; or, General Grammar of Arts, Sciences, and useful knowledge. i8mo. 5^. COLTON (B. P.) ELEMENTARY COURSE OF PRACTICAL ZOOLOGY. By B. P. Colton, A. M. , Instructor in Biology, Ottawa High School. Crown 8vo. cloth. 4^. 6d. CORNWELL’S SCHOOL GEOGRAPHY. 35-. 6 d, With Thirty Maps on Steel, 5^. 6d, GEOGRAPHY FOR BEGINNERS. i.y. With Questions, is. 4^^. JOYCE (P. W.) A HANDBOOK OF SCHOOL MANAGEMENT AND METHODS OF TEACHING. By P. W. Joyce, LL.D., &c. iit/i edition, revised. Cloth, y, 6d, SPELLING BOOKS: PINNOCK’S First Spelling Book. i8mo. is. False Spelling, is. 6d. WILLIAMS’ Syllabic Spelling. i2mo. 2s. 6d. CARPENTER’S Spelling Assistant. i2mo. is. DUNCAN’S English Expositor. i2ino. is. 6d. Scientific C®orfe0. PLANTE (G.) THE STORAGE OF ELECTRI- CAL ENERGY, and Researches in the Effects created by Cur- rents combining Quantity with High Tension. Translated from the French by Paul Bedford El well. With Portrait, and 89 Illustrations. 8vo. pp. vii.-268, cloth, 12s. GASEOUS FUEL, including Water Gas, its produc- tion and application. By B. II. Thwaite, C.E. Crown 8vo. NITROGENOUS MANURE: Increase in the Increase in the produce of the soil through the rational use of. By Prof. Paul Wagner, Darmstadt. Translated by G, G. Hen- derson, M.A, 8vo, sewed. Classical and Educational Works. 27 Cfte ^peciali0f0 ^eric0. A Series of Handbooks for Students and Practical Engineers, Crown 8vo. GAS ENGINES. Their Theory and Management. By William Macgregor. With 7 Plates. Crown 8vo. pp. 245, Zs, 6 d. BALLOONING : A Concise Sketch of its History and Principles. From the best sources, Continental and English. By G. May. With Illustrations. Crown 8vo. pp. vi.-py, 2 s, 6 d. ELECTRIC TRANSMISSION OF ENERGY, audits Transformation, Subdivision* and Distribution. A Prac- tical Handbook by Gisbert Kapp, C. E., Associate Member of the Institution of Civil Engineers, &c. With 119 Illustrations. Cr. 8vo. pp. xi.-33i. Third edition^ revised and enlarged, *]s. 6 d, ON THE CONVERSION OF HEAT INTO WORK. A Practical Handbook on Heat-Engines. By William Anderson, M. Inst. C.E. With 64 Illustrations. Pp. viii.-254. Second Edition. Cr. 8vo. 6 j“. SEWAGE TREATMENT, PURIFICATION AND UTILIZATION ; A Practical Manual for the Use of Corporations, Local Boards, Officers of Health, Inspectors of Nuisances, Chemists, Manufacturers, Riparian Owners, Engi- neers and Ratepayers. By J. W. Slater, F. E.S., Editor of “ Jo^i' 113.1 of Science.” Crown 8vo. cloth, price THE TELEPHONE. ByW. H. Preece,F.R.S.,and J. Maier, Ph.D. With numerous illustrations. Cr. 8vo. 12s. 6 d. MANURES, OR THE PHILOSOPHY OF MANURING. By Dr. A. B. Griffiths, F.R.S.Ed., F.C.S., Principal and Lecturer on Chemistry in the School of Science, Lincoln, &c., &c. Cr. 8vo. Js. 6 d, HYDRAULIC MOTORS: TURBINES AND PRESSURE ENGINES. By George R. Bodmer, Assoc. M.Inst.C.E. I4J-. ALTERNATING CURRENTS OF ELEC- TRICITY. By Thomas H. Blakesley, M.A., M.Inst.C.E. Second edition. 4^-. 6^/. COLOUR IN WOVEN DESIGN. By Professor Roberts Beaumont, Textile Department, The Yorkshire College, Leeds. With numerous Coloured Plates. 2\s, Others in preparation. 28 Whittaker and Co's List of Small crown Svo. cloth. With many Illustrations. 2 i 0 t)ittafeer's ILitirarp of 3 rt 0 , ^cicnce 0 , 9 @anufacture 0 anD 31nlm0trie0. ELECTRIC LIGHT INSTALLATIONS AND THE MANAGEMENT OF ACCUMULA- TORS. A Practical Handbook by Sir David Salomons, Bart., M.A. 5th Edition, Revised and Enlarged, with 100 Illus- trations. 5^. “ To say that this book is the best of its kind would be a poor compliment, as it is practically the only work on accumulators that has been written.” — Electri§al Review. ELECTRICAL INSTRUMENT - MAKING FOR AMATEURS. A Practical Handbook. By S. R. Bottone, Author of “ The Dynamo,” &c. With 60 Illustrations. Third edition. 3^". ELECTRIC BELLS AND ALL ABOUT THEM. A Practical Book for Practical Men. By S. R. Bottone. With more than 100 illustrations. Second edition, revised, 33-. ELECTRIC MOTORS : how Made and how Used. By S. R. Bottone. 33-. PRACTICAL IRON FOUNDING. By the Author of “ Pattern Making,” &c., &c. Illustrated with over one hundred engravings. 4s. ELECTRICITY IN OUR HOMES AND WORKSHOPS. A Practical Handbook. By Sydney F. Walker, M. Inst. C.E., M. I. E.E. Second edition. 5^. ELECTRICAL INFLUENCE MACHINES: Containing a full account of their historical development, their modern Forms, and their Practical Construction. By J. Gray. B. Sc. 4^. (id. THE PROTECTION OFBUILDINGS FROM LIGHTNING. A Treatise on the Theory of Lightning Conductors from a Modern Point of View. Being the substance of two lectures delivered before the Society of Arts in March, 1888. By Oliver J. Lodge, LL.D., D.Sc., F.R.S., Professor of Phy- sics in University College, Liverpool. METAL TURNING. By the Author of ‘‘ Practical Iron Founding.” hnmediately. Classical and Educational Works. 29 Cecftnological 2jaiorfe0* WERSHOVEN (F. J.) TECHNOLOGICAL DICTIONARY OF THE PHYSICAL, MECHANICAL, AND CHEMICAL SCIENCES. English and German. 2 vols. cloth, 5J-. PONCE DE LEON. ENGLISH - SPANISH TECHNOLOGICAL DICTIONARY. Containing Terms employed in the Applied Sciences, Industrial Arts, Mechanics, Fine Arts, Metallurgy, Machinery, Commerce, Ship-building and Navigation, Civil and Military Engineering, Agriculture, Railway Construction, Electro-technics, &c. 8vo. £\ i6j. Vol. II. — Spanish-English. In preparation. HOBLYN’S DICTIONARY OF TERMS USED IN MEDICINE AND COLLATERAL SCIENCES, wth edition. Revised throughout, with numerous Additions. By John A P. Price, B.A., M.D. Oxon., Assistant-Surgeon to the Royal Berkshire Hospital. Post 8vo. 814 pp. ioj“. 6 d. FODEN’S MECHANICAL TABLES, Fifth Edition. Cr. 8vo. cloth. \s. 6 d. CULLYER’S TABLES FOR MEASURING AND MANURING LAND, ETC. i6mo. cloth. 2J-. 6 d. HUTTON’S MATHEMATICAL TABLES. Con- taining Common, Hyperbolic, and Logistic Logarithms, also the Values and Logs, of Trigonometrical Ratios, with other Tables. By Chnrles Hutton, LL. D., F.R. S., with additional Tables by Dr. O. Gregory. New edition. Medium 8vo. cloth. 12 ^. 30 Whittakei^ and Co.'s List of ^isccllaneou0 publication0. THE FAMILIAR QUOTATION SERIES. Price \s. cloth limp ; \s. 6d. half-bound. Familiar Latin Quotations and Proverbs. *2. Familiar French Quotations and Proverbs. ^3. Familiar English Quotations. ^4. A Handy Classical Diction- ary, denuded of indecencies, for Family and School use. 5. Bible Truths with Shakespe- rian Parallels, showing the indebtedness of Shakespeare to the Bible. *6. A Handy Book of Synonyrns, containing about 33,000 English words. ^7. The Secretary’s Assistant and Correspondent ’sGuide, show- ing the correct mode of addres- sing, by letter or otherwise, persons of every degree of rank. 8. The Tourist’s French Pro- nouncing Handbook, giving the pronunciation for the use of travellers who have little or no knowledge of the French lan- guage. Reasons why we Believe the Bible ; the historical evidence of the DTvine origin and inspira- tion of the Scriptures. By the Rev. James Copland, M.A. 10. Tales from Shakespeare. By Charles Lamb. A Dictionary of Daily Blun- ders, correcting mistakes con- stantly made in speaking and writing. 12. Rasselas. By Dr. Johnson. 13. Elizabeth, or the Exiles of Siberia. 14. Rejected Addresses. By Horace and James Smith. 15. A Dictionary of English Proverbs. ^16. Plutarch’s Lives of Illus- trious Men, denuded of inde- cencies, for Family and School use. 17. De Quincey’s Opium Eater. 18. The Story of the New Testa- ment. By the Rev. A. Carter, M.A. This volume gives much informa- tion regarding the early and later history of the Testament, and the bases, codices, etc., on which its truth is founded, and other in- teresting particulars. It forms a very useful companion volume to the Revised Version. *^ig. A Handy Dictionary of My- thology, for every-day readers and for school use, denuded of indecencies. 20. Moore’s Irish Melodies. 21. Moore’s Lalla Rookh. Duplicate Volume, price 2^., comprising Tourist’s French PronouMcing Handbook, with Phrases, etc. * These nine volumes are also issued as the MINIATURE REFERENCE LIBRARY, neatly arranged in a box, price lo^. 6 d. A selection from the series may also be had in five triplicate volumes neatly bound with coloured edges, 3^. each. Classical and Educational Works. 31 DOD’S PEERAGE, BARONETAGE, AND KNIGHTAGE of Great Britain and Ireland for 1890. Including all Titled Classes. Fiftieth Year. Crown 8vo. cloth gilt. iot. (id. This is without doubt the handiest, and within the necessary limits, the most complete work of its class. It is not only a complete Peerage, Baronetage, and Knightage, but it also in- cludes Privy Councillors and four separate lists of all who bear courtesy titles, and of the inferior titles of living peers. A chapter on precedence, an explanation of titular distinctions and terms relating to ranks, and instructions as to modes of address are amongst the special features of this compendious volume. PARLIAMENTARY COMPANION for 1890. Fifty- eighth Year, and Sixty-third Issue. 32mo. gilt edges. 4^. 6 d. In use in the House of Commons, the Government Offices, and the principal London Clubs. This volume contains, in small compass, a complete list with short descriptions of the House of Peers, an explanation of Par- liamentary Terms and Proceedings ; a list of the Pollings in all constituencies at the last election ; a list with detailed descrip- tions of members of the House of Commons ; lists of the Principal Officers of both Houses, and other information frequently required by the politician or the student of Parliamentary debates. It is corrected every year to the latest possible date before the opening of Parliament. A HISTORY OF THE YEOMEN OF THE GUARD, with a concise account of the Tower Warders. By Thomas Preston, F.R.H.S. i2mo. NIMROD’S REMARKS ON THE CONDITION OF HUNTERS, the Choice of Horses, and their Management. By C. Tongue. Crown 8vo. cloth. THE DURHAM UNIVERSITY CALENDAR AND ALMANACK. Annually, u. 6 d. SURTEES SOCIETY’S PUBLICATIONS. 32 Whittaker and Co.’s List. THEOLOGICAL AND OTHER WORKS PUBLISHED BV THOMAS WHITTAKER, NEW YORK. BARRETT (R. S.) THOUGHT SEED FOR HOLY SEASONS. Crown 8vo. 4^. 6d. BOSS (N. R.)THE prayer BOOK REASON WHY. i2mo. w. ed. BUEL’S (Prof. S.) DOGMATIC THEOLOGY. 2 vols. 8vo. 28^. CLARK (R. W.) THE CHURCH IN THY HOUSE— Daily Family Prayers. Crown 8vo. 2,s. 6d. DELITZSCH (FRANZ) BEHOLD THE MAN ! Crown 8vo. 2^. FAIRBAIRN (R. B.) OF THE DOCTRINE OF MORALITY, in its relation to the Grace of Redemption. Crown 8vo. js. 6d. FOLEY (G. C.) AN OUTLINE HARMONY OF THE FOUR GOSPELS. i2mo. 6d. FORRESTER (H.) CHRISTIAN UNITY AND THE HISTORIC EPISCOPATE. Royal 8vo. 55. HARRIS (S. S.) THE RELATION OF CHRISTIANITY TO CIVIL SOCIETY. Crown 8vo. 4.?. 6d. “IN THE PRESENCE ’’ and other Verses. Small 4to. ss. JACOB AND JAPHETH : Bible Growth and Religion. Crown 8vo. 6^. LITTLE GRANDPA. By M. A. C. Crown 8vo. 4^. MANN (CAMERON) ON FUTURE PUNISHMENT. Crown 8vo. 3^. 6d. MUHLENBERG fW. A.) LIFE AND WORK OF. By Anne Ayres. 8vo. los. SNIVELY (W. A.) PARISH LECTURES ON THE PRAYER BOOK. Crown 8 VO. TESTIMONIES TO THE SUPERNATURAL. Crown 8vo. 6^. SNYDER (A. W.) THE CHIEF THINGS. Crown 8vo. C/itrap edition. Sewed. 2^. 6