UNIVERSITY OF CALIFORNIA 
 LOS ANGELES 
 
 ,
 
 -^^^ 
 
 TRANSFORMERS. 
 
 THEIR THEORY, CONSTRUCTION AND 
 APPLICATION, SIMPLIFIED. 
 
 BY CARYL D. HASK1NS. 
 
 ASSOCIATE MEMBER AMERICAN INSTITUTE OF 
 ELECTRICAL ENGINEERS. 
 
 ILLUSTRATED. 
 
 1892. 
 BUBIER PUBLISHING COMPANY, 
 
 LYNN, MASS.
 
 COPYRIGHTED BY 
 BUBIER PUBLISHING COMPANY, 
 
 LYNN, MASS. 
 1892. 
 
 G. H. & W. A. NICHOLS,
 
 Ttf 
 
 DEDICATION. 
 
 TO MY EARLIEST, MOST THOROUGH, AND MOST VALUED 
 IXSTICTOR IN PRACTICAL ENGINEERING, MY FATHER, 
 JOHN F. HASKINS, M. E., M. I. M. E., M. S. A., ETC., THIS 
 LITTLE VOLUME IS RESPECTFULLY DEDICATED. 
 
 379385
 
 CONTENTS. 
 
 DEDICATION. 
 
 Page 
 
 CHAPTER 1 9 
 
 INDUCTION AND DISTRIBUTION BY ALTER- 
 NATING CURRENT. 
 
 CHAPTER II 31 
 
 THEORETIC CONSIDERATIONS OF THE TRANS- 
 FORMER. NON-REGULATION. SELF-INDUC- 
 TION. MUTUAL INDUCTION. LOSSES, FOU- 
 
 CAULT CURRENTS. HYSTERESIS, LEAKAGE. 
 
 CHAPTER III 55 
 
 THE THEORY OF THE TRANSFORMER MATH- 
 EMATICALLY CONSIDERED. 
 
 CHAPTER IV 63 
 
 EVOLUTION OF THE ELECTRICAL TRANS- 
 FORMER. 
 
 CHAPTER V 73 
 
 TRANSFORMER CONSTRUCTION. 
 
 CHAPTER VI 96 
 
 THE TRANSFORMER IN SERVICE. 
 
 CHAPTER VII 110 
 
 COMMERCIAL TRANSFORMERS. 
 
 APPENDICES 128 
 
 GLOSSARY 143 
 
 INDEX. 145
 
 PREFACE. 
 
 It has often been observed by almost every member 
 of the electrical fraternity, that induction, and its out- 
 come, the transformer, is to the popular mind, the 
 greatest mystery of the whole lighting system with 
 which they come in contact. 
 
 There is something tangible about the dynamo. Its 
 movement, and the applied power are apparent. The 
 lamp glows, and its action is appreciable, but the trans- 
 former remains to them an uncanny mystery. 
 
 So too the average electrician whose training has 
 long accustomed him to the management and applica- 
 tion of the electric current, finds in the transformer as 
 a rule, more points regarding which his mind is hazy 
 and uncertain, than in any other one piece of appara- 
 tus with which he has to deal. 
 
 The greater part of the printed matter dealing 
 with the transformer which has from time to time 
 appeared, has been either strictly technical, or entirely 
 popular. 
 
 I have endeavored in the following pages to treat of 
 the transformer and its action in such a manner as to 
 render the work of especial value to the central sta- 
 tion electrician, the student, and the investigator,
 
 8 PREFACE. 
 
 while the greatest care has been exercised to render 
 the matter so clear, simple and interesting, that it may 
 come within the scope of the general public, and meet 
 the demand for a semi-technical, and yet semi-popular 
 treatise on the Electrical Transformer which has not 
 heretofore been obtainable. 
 
 CARYL D. RASKINS. 
 
 Boston, August 1st, 1892.
 
 TRANSFORMERS. 
 
 CHAPTER I. 
 
 INDUCTION AND DISTRIBUTION BY ALTERNATING 
 CURRENT. 
 
 BEFORE taking up and considering in detail the 
 calculation, construction and use, of the piece of 
 apparatus which forms the subject matter of this 
 treatise, it is necessary that the natural laws and 
 commercial necessities which lead up to its manu- 
 facture and application should be briefly con- 
 sidered. It has been thought best therefore, to 
 devote a preliminary chapter to the natural phe- 
 nomenon know as "Electrical Induction," and* to 
 the general character, functions, and application 
 of alternating currents. The object of this pre- 
 liminary chapter, of necessity renders it of an 
 elementary nature, but it can scarcely be con- 
 sidered superfluous, as it deals with the funda- 
 mental principle which underlies the whole subject. 
 Since certain statements must in the consideration 
 of all subjects be taken for established certainties,
 
 10 TJRANSFOKMEKS. 
 
 as bases for future argument, let us consider it an 
 axiom that all electrical currents have a direction 
 of flow, just as have all streams of water or cur- 
 rents of air. To render the idea of electricity 
 more thoroughly consistent with our mode of 
 thought, it will be found preferable to cease to 
 consider it as an imperceptible existence, and to 
 treat it as a tangible body ; it will therefore be 
 dealt with and regarded as a subtle fluid through- 
 out the following pages. 
 
 It is a recognized fact that electricity as an 
 imaginary fluid may under certain conditions be 
 confined, and may exist in inertia; but at such 
 times it is generally known as difference of 
 potential, only when some form of motion results 
 however, does electricity become manifest to 
 human perception, arid then only within certain 
 ranges and under certain conditions. 
 
 The particular function of electricity with which 
 we are chiefly to deal in this treatise is a very 
 remarkable natural law known as Electrical In- 
 duction. Unlike many other electrical terms, the 
 word induction is here very aptly applied. Induc- 
 tion, the noun, from the verb induce, which Webster 
 defines as " to lead or influence by persuasion, to 
 actuate, to impel, to urge;" "electrical induc- 
 tion" in fact, is that force which "persuades," 
 "actuates" or "impels" an electrical current in 
 one body by the influence of current in another.
 
 INDUCTION AND DISTRIBUTION. 11 
 
 To state this briefly it may be said, that whenever 
 an electrical current comes into existence it induces 
 or creates a current in all surrounding conductive 
 masses, in a direction opposite to its own. During 
 the period of its flow in uniform quantity, it has 
 no inductive influence ; the period of flow of the 
 induced current extending only over a time equal 
 to that between the commencement and attain- 
 ment of full quantity of the inducing current. 
 (This statement will be modified later, but for the 
 present it may be safely accepted.) When a 
 current ceases it induces a current in surrounding 
 masses, in a direction similar to its own. 
 
 As has been already stated, currents are induced 
 in " all surrounding conductive masses." This is 
 strictly true, however remote these masses may be. 
 The " making " or " breaking " of ever so insignifi- 
 cant a current circuit on the earth, undoubtedly 
 induces currents in conductive masses on the 
 Moon, but since the inductive influence decreases 
 as the square of the distance, the effect, even 
 through comparatively small intervals of space, 
 becomes inappreciable. The interval of time, 
 between the commencement of the current flow in 
 the inducing circuit, or, as it is generally called, 
 the " primary," and the commencement of flow in 
 the induced circuit, or secondary, is known as the 
 "lag" between primary and secondary. 
 
 Having briefly stated the general character of
 
 12 TRANSFORMERS. 
 
 inductive law, it will be well to consider the cause 
 of this effect. This can be more clearly exempli- 
 fied by resorting to simile. We do not claim the 
 similes as original, but doubtless none more plain 
 or conclusive could be found. 
 
 All active electrical conductors (for convenience 
 we will limit our consideration of the matter to 
 wires,) are considered to be surrounded by a field 
 of force, little ripples or rings of energy, concentric 
 with the conductor, extending equally in all direc- 
 tions, and becoming less and less intense as they be- 
 come more and more remote from their source. 
 By referring to diagram No. 1, this may be- made 
 more clear. Let A represent a section through the 
 wire or conductor, and B, B', B" the lines of force 
 surrounding it ; becoming less intense as they be- 
 come more remote. Now the theory on which all 
 inductive electric law is based is : That whenever 
 a conductor cuts through " lines of force," or vice 
 versa, when " lines of force " cut through a con- 
 ductor, a current is set up in that conductor, pro- 
 vided always, that the conductor be part of a 
 closed or complete circuit. Current effects may 
 in reality be created in what would, generally 
 speaking, not be considered a complete circuit. 
 These effects are known as " Eddy " or " Fou- 
 cault " currents, the former term exactly defining 
 their character. Further consideration will be 
 given to these effects in a later chapter.
 
 INDUCTION AND DISTRIBUTION. 13 
 
 ii. a.
 
 14 TRANSFORMERS. 
 
 In the dynamo, current is generated by moving 
 conductors, in such a way that they will cut 
 through lines of force. In the transformer, current 
 may be said to be generated by moving lines of 
 force, in such a way that they will cut through 
 conductors. Thus, in one sense, the dynamo and 
 the transformer may be said to be opposites, al- 
 though in reality almost identical in theory. 
 
 We will now endeavor to describe this whole 
 matter more graphically. It should be borne in 
 mind that in this, as in all similar descriptions 
 throughout the volume, due allowance must be 
 made for the liberty which is taken in treating of 
 electricity, lines of force, etc., as actual palpable 
 existences. In the case of the latter, at least, we 
 may say quite frankly that they are simply influ- 
 ences which have been universally assumed to 
 plausibly account for effects, the true cause of 
 which we can only surmise. As has already been 
 hinted, it is only the effects that are caused by 
 electricity that are appreciable to us, the exact 
 nature of the subtle influence itself is no more 
 understood today than it was when the ancient 
 Greeks discovered the effect of friction upon 
 amber, and christened it "Electron." 
 
 Turning to Fig. 2, let us consider A and B 
 to be electrical conductors, each one a portion of 
 two separate complete circuits. A current springs 
 into existence in A, the primary (no matter, for
 
 INDUCTION AND DISTRIBUTION. .15 
 
 the present, what created it). Lines of force are 
 immediately projected into the surrounding space, 
 just as ripples are set in motion on the surface of 
 water by a falling stone. These lines, in course 
 of projection, must of necessity cut through B, the 
 secondary. This induces current in B. Let us 
 suppose that the current in A is flowing towards 
 us ; then the current in B is flowing away from us. 
 As soon as the current in A becomes fixed in 
 quantity, and the lines become stationary, the cur- 
 rent in B ceases. Now let us suppose that the 
 current in A ceases. The lines of force immedi- 
 ately collapse upon the centre B, In doing this 
 they must necessarily cut through B again, but in 
 an opposite direction. This sets up a second cur- 
 rent in B, also in an opposite direction to the former 
 one, so that the current in B is now coming 
 towards us ; the same direction that the current in 
 A held until it ceased. Thus it will be seen, that, 
 as has already been stated, when an electric cur- 
 rent comes into existence it induces a current in 
 neighboring conductors, in a direction opposite to 
 its own. When a current ceases it induces a 
 current in surrounding conductors, in a direction 
 similar to its own. 
 
 Having made the fundamental principle of in- 
 duction somewhat clear, we may now give some 
 brief consideration to the system of electrical 
 distribution, which this remarkable law has enabled
 
 16 TRANSFORMERS. 
 
 man to formulate. Generally speaking, it may 
 quite safely be said that the distribution of elec- 
 tricity through long distances is not economical, 
 and for a very simple reason. The loss due to 
 electrical resistance of the conductors must be 
 considerable, since C 2 R = watts lost in transmis- 
 sion. Where C = Current in Amperes, R==Resist- 
 ance of Conductor in Ohms. Thus it is evident 
 that either the resistance of the circuit must be 
 kept at a minimum, or the current strength must 
 not be great. 
 
 The first of these two alternatives would at once 
 necessitate the use of large masses of copper for 
 conductors, but this is generally precluded by the 
 high price of that valuable commodity. The other 
 alternative, then, which in any case is the most 
 desirable, is the only one worthy of consideration 
 from a commercial standpoint. 
 
 To reduce the current strength or volume, and 
 yet retain for transmission the same amount of 
 force or energy, it of course becomes necessary to 
 increase the voltage or pressure. In this respect 
 electricity is quite analogous to water the same 
 amount of power may be delivered by a small 
 stream at a high pressure, as by a large stream 
 with small head. Thus far, then, the second al- 
 ternative presents no difficulty, as by sufficiently 
 increasing the pressure and reducing the current, 
 high powers may be transmitted through long dis-
 
 INDUCTION AND DISTRIBUTION. 17 
 
 tances without the use of large quantities of cop- 
 per and without excessive loss. 
 
 But a new difficulty now presents itself, inas- 
 much as currents of high pressure are extremely 
 difficult to control and use for general purposes ; 
 besides which fact, it must be borne in mind that 
 contact with high voltages is dangerous and some- 
 times fatal to human life. Incandescent lamps have 
 not as yet been so constructed as to operate success- 
 fully on voltages in excess of 200 volts, which may 
 be considered low pressure ; whilst even if the lamps 
 would bear higher voltage, such careful and perfect 
 insulation would be necessary as to render their 
 use commercially impracticable. It is plain, there- 
 fore, that the very pressure which renders it possi- 
 ble to carry electrical energy economically through 
 long distances, would render it practically value- 
 less, provided there were no way to reduce it at 
 points of application without serious loss. 
 
 Induction provides us at once with a means of 
 accomplishing this ; as by proportioning the amount 
 of wire in inducing and induced circuits, the 
 energy delivered at a high pressure on the primary, 
 may be transformed to a reduced pressure and 
 greater volume in the secondary, or vice versa. 
 This can of course only be accomplished by the 
 use of intermittent, pulsating or alternating cur- 
 rents ; any or all of which will produce the pulsa- 
 tions of the lines of force, which forms the
 
 18 TRANSFORMERS. 
 
 fundamental principle of induction. The electric 
 current which was first applied to incandescent 
 lighting, was the direct, flowing constantly and 
 uninterruptedly in one direction. This class of 
 current does not of course permit of the use of 
 induction for transformation, and has therefore to 
 be distributed at useful or low pressures, which as 
 we have seen renders it wasteful for long distance 
 work. 
 
 As has just been stated, distribution by pulsat- 
 ing or alternating current, renders it possible to 
 adjust the potential at any given point to any 
 voltage, without serious loss. In practice how- 
 ever, only one of these classes of current is 
 practicable, and this for reasons which will be 
 presently stated, is the alternating.* 
 
 It m&y be well however, before going further, to 
 make plain the difference between the pulsating 
 and the alternating current. To state this quite 
 plainly, it may be said that a pulsating current is 
 one which starts into existence in direction "a" 
 Fig. 3, dies down, more or less rapidly, then 
 comes into existence again in the same direction.^ 
 We have endeavored to make this a trifle more 
 explicit, by the use of the diagram, the arrows 
 
 *This statement is made in a general way; there is at least one 
 marked exception to the general rule. . 
 
 . t" Pulsating," " intermittent," " vibratory," " interrupted," etc., are 
 all names in common use for various modifications of the same class 
 of current.
 
 INDUCTION AND DISTRIBUTION. 19 
 
 showing the direction of current flow during 
 different periods. 
 
 In the technical consideration of alternating and 
 intermittent currents, a system of "curve" draw- 
 ing is used, whereby every rise and fall of current, 
 every break and every alternation is accurately 
 set forth. This system, which is almost invaluable 
 
 in all work connected with alternating or 
 fluctuating currents, we will only briefly describe, 
 as it is in general use, and almost all are familiar 
 with it. Let us suppose that we have a rectangle, 
 the surface of which is divided into very minute 
 squares by vertical and horizontal lines. In draw- 
 ing a curve showing the fluctuations of an alter- 
 nating current, the distance between the vertical
 
 TRANSFORMERS. 
 
 lines, or, as they are technically termed, "Ordi- 
 nates," should be considered as fractions of time 
 (generally taken in decimals of a second or min- 
 ute), whilst the " Abscissae," or distance between 
 the horizontal lines represent amperes (or fractions 
 of current.) Two such curves are here represented. 
 The "cross sectioning" is not shown, it being 
 really unnecessary for purposes of description. 
 
 Fig. 4 represents an intermittent current. 
 The line A B is the neutral line, and where the 
 curve line touches this, no current is flowing. The 
 portion of the curve above A B indicates that the 
 current is flowing in, say a -J- direction, whilst 
 when the curve passes below the line, the current is 
 indicated as flowing in a direction opposite to that 
 which is held when the curve was above the line 
 or is flowing in, say a direction. A B is con- 
 sidered as 0, the amperes being numbered each 
 way from it, those above being in one direction, 
 those below in another. 
 
 It will be easy to imagine by casting the eye 
 upon the upper curve, ( the pulsating current ) 
 Fig. 4, how the lines of force will alternately be 
 projected from and collapse upon the primary 
 conductor, cutting surrounding conductors and 
 inducing currents in them. The distance between 
 the points where the curve touches the line A 
 B, shows us the length of duration of one period, 
 or the speed with which the current is alternately
 
 INDUCTION AND DISTRIBUTION. 
 
 21
 
 22 TRANSFOKMEKS. 
 
 " made " and broken, whilst the distance between 
 the line and the maximum elevation of the curve, 
 show us the maximum current strength. These 
 curves are more often drawn to show the pressure 
 waves, in which case of course, the abscissae 
 indicate volts ; or to show the power curve, the 
 product of pressure and current (Cx V= watts) in 
 which case the abscissae become watts. 
 
 The same general discription applies equally 
 well to the alternating curve shown in the lower 
 position, Fig. 5. It is only necessary to bear in 
 mind the fact that where the curve line crosses the 
 line, first a cessation and then a reversal of 
 direction of current flow is indicated. It may be 
 well to state that these lines follow very closely 
 the rule of " sine " curves. We will make no 
 attempt to discuss the theory of the sine curve as it 
 has been thought best to confine the mathematics 
 in this treatise exclusively to the principle and 
 theory of the transformer.* 
 
 We may now return to our more immediate sub- 
 ject, and point out why the alternating, as com- 
 pared with the pulsating current, is the only 
 commercially practicable means of distributing 
 large quantities of energy by the induction system. 
 At first this seems rather incongruous, since we 
 have already seen that the pulsating and the alter- 
 
 * For a concise, simple and clear elucidation of the " sine curve " we 
 would refer the reader to Messrs. Slingo & Brooker's " Electrical Engi- 
 neering," an excellent work.
 
 INDUCTION AND DISTRIBUTION. 23 
 
 nating current will "induce" equally well and 
 efficiently, the former, in fact, probably more effi- 
 ciently under certain conditions. 
 
 First and most important in considering this 
 question, we must remember that the natural out- 
 come of the dynamo-electric machine is an alter- 
 nating current. It is the commutator, quite a 
 distinct portion of the dynamo, which serves to 
 " straighten " the current, and the commutator 
 might quite as well be a distant machine, quite 
 separate from the dynamo, so far as theory is con- 
 cerned. It is only mechanical considerations and 
 convenience which have caused it to be incorpo- 
 rated in that machine. The function of the 
 commutator, in fact, is simply to reverse the con- 
 nections at the same instant that the direction of 
 current flow reverses, that is all. We may illustrate 
 this : Suppose there were two tanks full of water, 
 as illustrated in Fig. 6-1, A, B, A being higher than 
 B, and that these two tanks were connected by a 
 long rubber tube, easily removable from spickets, 
 and that you held the end of the tube at X in the 
 left hand, and the end Yin the right; the water 
 would be running from J., through the tube in the 
 direction of the arrow. Now suppose B were 
 suddenly lifted higher than A, but at the same 
 instant you pulled the end of the tube X off of A 
 spicket, and thrust it on to B, and vice versa with 
 the end y, then the direction of flow would be
 
 24 
 
 TRANSFORMERS.
 
 INDUCTION AND DISTRIBUTION. 25 
 
 maintained the same in the tube. It is just the. 
 function of your hands here that the commutator 
 performs for the dynamo. 
 
 If the commutator was omitted, and collector 
 rings substituted, we should have an alternating 
 current output, easier to get than the direct, but 
 not at first used because supposed to be less man- 
 ageable. 
 
 On the other hand, to get a pulsating current 
 we must either have a dynamo of special construc- 
 tion, or we must have an interrupter or circuit 
 breaker which will vibrate very rapidly, alternate- 
 ly opening and closing the circuit. This latter 
 device is obviously only practicable with low 
 voltages, for with high potentials constant arcing, 
 and speedy destruction, would inevitably result. 
 
 Much attention has been given of late by a 
 number of the world's best electricians, to the 
 special dynamos already mentioned as being de- 
 signed to generate pulsating currents, and consid- 
 erable advance has been made in this direction. 
 The system would present at least one marked 
 advantage, provided it could be reduced to a basis 
 equally efficient with the alternating. It would 
 permit of the distribution of power to electric 
 motors, as well as for lighting purposes, a use to 
 which alternating current has not as yet been 
 applied with any marked degree of success, except 
 in very small powers ; in fact, the use of the
 
 26 TRANSFORMERS. 
 
 alternating current is practically confined today 
 to the distribution of light, and to a few special 
 purposes, such as electric welding, etc. 
 
 It is easy to convert mechanical energy into elec- 
 trical energy in the form of alternating current, but 
 it is most difficult to convert alternating current 
 energy back again into mechanical energy or motion 
 with any degree of efficiency. 
 
 Pulsating or interrupted currents are largely 
 used in connection with batteries and induction 
 coils, where but small amounts of energy are to be 
 dealt with, but not in connection with lighting 
 work. Inductive coils and transformers, are 
 practically the same thing, the former term is 
 used in relation to coils for raising low potentials 
 to high, especially where the powers dealt with are 
 small, almost always in conjunction with batteries. 
 Where large powers are raised to higher potentials, 
 the coils are generally spoken of as Step-Up Trans- 
 formers, which are treated of quite exhaustively 
 in an appendix. The term Transformer, or as it is 
 frequently called (especially in Europe) Converter, 
 is used to specify coils intended for reducing high 
 potentials to lower. Induction coils are exten- 
 sively used in conjunction with batteries and an 
 interrupter, for experimental work, medical pur- 
 poses, electric gas lighting, the explosion of 
 blasts, etc. They will be treated of in the following 
 pages only in the most cursory manner, they are
 
 INDUCTION AXD DISTRIBUTION. 27 
 
 most ably described in a number of existing 
 volumes, and scarcely come within our subject 
 matter.* 
 
 Distribution by alternating current and trans- 
 former is commonly accomplished by the multiple 
 or parallel system. Fig. 7 shows quite clearly the 
 relative positions of dynamo and transformer when 
 the latter are placed in multiple ; with this arrange- 
 ment the primary circuit, (and each secondary 
 also,) has a constant potential, the amount of cur- 
 rent varying with the load. Transformers have 
 also at times been arranged in series, that is one 
 after another, as for example in the old Jabloch- 
 koff system of arc lighting, but this arrangement 
 is not common, and has only been used in connec- 
 tion with lighting service for street and arc lights. 
 The series arrangement is clearly shown in Fig. 8. 
 With the series system the potential is variable 
 with the load, and the current constant. 
 
 The series and multiple arrangements of trans- 
 formers may be clearly defined as follows : When 
 transformers are placed in series the same current, 
 and whole current, passes through the primary of 
 each transformer one after another, and then back 
 to the dynamo. 
 
 When transformers are arranged in multiple, a 
 certain proportion of the entire current passes 
 
 * For information relating to induction coils, the author refers the 
 reader to "Electricity and its Recent Applications," by Edward 
 Trevert.
 
 28 
 
 TRANSFORMERS.
 
 INDUCTION AND DISTRIBUTION.
 
 TRANSFORMERS. 
 
 from the mains through each transformer primary 
 individually, and thence back to the dynamo. If 
 a single primary burn out or break with the series 
 system, all of the lights go out, for the main cir- 
 cuit is open. (Special devices are however made 
 to provide for this contingency.) If a primary 
 burns out with the multiple arrangement only the 
 lights on that particular transformer go out, or if 
 the main circuit is broken anywhere along the line, 
 then only those transformers which are beyond the 
 break from the dynamo cease to operate. The 
 lights which are between the dynamo and the 
 break still continue to burn. 
 
 The multiple system is the only one in common 
 use, and is the one with which we shall chiefly 
 deal. The pressures, or voltages (primary), which 
 are in common use are 1000, 2000 and 2500 volts, 
 and in a few cases 5000 volts. Yet higher voltages 
 than these are in use in Europe, but with these 
 very high pressures proper insulation becomes a 
 very difficult problem.
 
 THEORETIC CONSIDERATIONS. 31 
 
 . CHAPTER II. 
 
 THEORETIC CONSIDERATIONS OF THE TRANS- 
 FORMER. IRON. REGULATION. SELF-INDUC- 
 TION. MUTUAL INDUCTION. LOSSES. FOU- 
 CAULT CURRENTS. HYSTERESIS. LEAKAGE. 
 
 IN the preceding chapter we have reviewed the 
 principle and characteristics of induction, "and the 
 necessities and advantages which have led up to 
 the use of the alternating current and the trans- 
 former, as an economic means for the distribution 
 of light. We will now endeavor to treat some- 
 what in detail those various influences which serve 
 to modify or enhance the phenomenon which we 
 have already described. 
 
 As we have already seen, the creation or stop- 
 page of a current in a conductor will induce a 
 current in surrounding conductors, and it is known 
 that this induction is subject to definite and known 
 laws. 
 
 It is obvious that to bring long lengths of wire 
 into one another's useful inductive influence, or 
 "field " some other means must be adopted than 
 that of stretching them side by side through a long 
 distance, for this would be thoroughly impracti-
 
 32 TRANSFORMERS. 
 
 cable, neither would this serve the purpose, even 
 if convenient, for experiment has shown that the 
 resistance of the two circuits being consistent, the 
 voltage in the primary and secondary is almost 
 exactly proportional to the respective lengths of 
 the two circuits within one another's influence. 
 Thus we may say, to state this arithmetically, that: 
 
 As influencing length primary : Influenced 
 length secondary : : Voltage primary : Voltage 
 secondary. 
 
 Now to bring, say ten feet of secondary, into the 
 equal influence of one hundred feet of primary, 
 both stretched in a straight line, is obviously im- 
 possible. It has been found necessary and most 
 desirable in practice, therefore, to make the two 
 wires into coils, placed one next to, or one over 
 the other (see Fig. 9). In this way it is perfectly 
 easy to bring a long wire into equal influence 
 throughout its length, upon a short wire, A A 
 being the short wire, the secondary (generally 
 speaking,) and B B the primary, or long wire. 
 But here is met a new contingency, that of self- 
 induction. Let us consider this same coil (Fig. 9) 
 in section Fig. 10. It is quite obvious that the 
 lines of force projected from any turn of the pri- 
 mary B, must cut and influence, not only the 
 neighboring turns of the secondary A A, but also 
 the neighboring turns of itself, thereby setting up 
 an influence in itself, in direct opposition to, and
 
 THEORETIC CONSIDERATIONS.
 
 34 TRANSFORMERS. 
 
 tending to nullify, the primary influence. Sup- 
 posing for an instant that the secondary coil were 
 absent, leaving the primary to act solely upon 
 itself, turn upon turn, and supposing that nothing 
 is lost in induction, or in other words, that the 
 efficiency be 100 per cent, then the induced in- 
 fluence, the counter or opposing pressure (the 
 "counter electro-motive force," as it is called), 
 should exactly equal the initial primary force 
 one hundred units of force pushing against one 
 hundred, and nothing would flow in the primary. 
 A coil of this kind is used for some purposes, being 
 known as an inductive resistance. At-the first 
 glance this would seem to entirely destroy the 
 value of the transformer, but in reality it plays a 
 most useful and important part when modified by 
 the influence of mutual induction, as we shall 
 presently see when considering the question of 
 regulation. 
 
 Referring again to Fig. 10, we will now consider 
 the part which the secondary coil plays. We have 
 already seen how the primary induces current in 
 the secondary, and opposing force by self induc- 
 tion within itself. It is obvious that since the 
 primary coil induces current in the secondary, by 
 reason of the lines of force projected from and 
 retracted to itself, the secondary coil must in its 
 turn react by induction upon the primary as soon 
 as it becomes active, and this reactive induction of
 
 THEORETIC CONSIDERATIONS. 35 
 
 the secondary upon the primary is, of necessity, 
 favorable to and in the same direction as the prime 
 current of the primary coil and in direct opposi- 
 tion to, and tending to nullify the effect of the 
 self-induction in the primary. Meantime however, 
 the secondary has set up within itself, by its self- 
 induction, a force in opposition to its prime or 
 useful current, which is again practically nullified 
 by the induction from the primary. Thus, the two 
 coils act and react upon themselves and one 
 another, the reaction of a coil upon itself, being 
 termed as we have already said, self-induction, and 
 that of one coil upon the other, which we have 
 just discribed, mutual induction. The two terms 
 are sufficiently indicative, being, in fact, almost 
 descriptive of their relative action. 
 
 On these two factors of self and mutual induc- 
 tion, chiefly depends the successful operation of 
 the transformer, for upon them alone rests the self- 
 regulation of the transformer ; without regulation 
 it would be practically useless, because wasteful 
 and inefficient. 
 
 Transformer regulation is simple and readily 
 understood, and its presence amounts to a ne- 
 cessity. 
 
 In Fig. 11, we show a transformer which we 
 will suppose to be in operation, P is the primary 
 supplied from an active circuit, through which 
 current is constantly flowing. S is the secondary,
 
 TRANSFORMERS. 
 
 Pifc.il 
 
 1 
 
 . 12.
 
 THEOKETIC CONSIDERATIONS. 37 
 
 i. e. lamp circuit, and L L are lamps. The primary 
 and secondary are separated from one another 
 solely to render the diagram more distinct. 
 
 It is evident that if the amount of current 
 flowing through P, were dependent solely upon its 
 so-called " dead resistance" or the Ohms* in its 
 circuit, then the same amount of energy would be 
 expended in it continuously, whether the lamps in 
 S r circuit were burning or not, this would be ex- 
 tremely wasteful for the dynamo, and therefore the 
 engines at the station would be doing just as 
 much work, and just as much coal would be burned 
 when no lights were turned on, as when all were 
 in use. Then too, so little current could be per- 
 mitted to pass through P, that it could be practi- 
 cally of no use in transferring energy to S. Here 
 self-induction comes into play. Supposing all of 
 the lamps (.Z/) to be turned off, S is obviously an 
 " open " or an incomplete circuit, and no current 
 can traverse it. The self-induction of P then has 
 full play to react on P coil, choking down the 
 prime current to almost nil, the amount of power, 
 in fact present in P being about equal to the dif- 
 ference in energy between the prime current on 
 full load and the reaction of self-induction. 
 
 This difference represents, practically, the losses 
 in the transformer, with which we shall presently 
 
 * For a definition of terms which may be unfamiliar to the non- 
 electrical reader, see glossary at end of volume. 
 
 73385
 
 88 TRANSFORMERS. 
 
 deal, almost all resulting, however, directly or in- 
 directly in heat. The engine and dynamo are, 
 therefore, doing very little work at this point, 
 when no lights are in use. 
 
 Now suppose one light is turned on in S circuit. 
 A little current now flows in $, dependent upon 
 the resistance of S circuit. This sets up a slight 
 mutual inductive reaction of S upon P, counter- 
 balancing a portion of the self-induction of the 
 primary. More current flows in P, and the engine 
 and dynamo are doing just that much more work. 
 And so the action goes on, as light by light is 
 brought into service, till at full load (or the point 
 of safe carrying capacity of the two wires) the 
 self-induction of the primary is proportionally 
 balanced by mutual induction. If the secondary 
 coil were short circuited, as frequently happens, 
 that is, if the two ends of S were joined without 
 intervening resistance, such as lamps, then more 
 current would pass than the wire could stand, and 
 it would melt in two at its weakest point, quite as 
 likely in the primary as the secondary. The same 
 thing would happen if too many lamps were placed 
 in the secondary. 
 
 To prevent the destruction of transformers in 
 this way " fuses " are introduced, both in the pri- 
 mary and the secondary circuits. A fuse is a short 
 piece of soft, easily fusible metal, usually lead and 
 tin, calculated to melt, and thus -break the circuit
 
 before the danger limit of the winding or wire is 
 reached. When a fuse burns out, or *' blows," all 
 of the lights go out which are on that transformer, 
 but nothing is injured. 
 
 The reader may have wondered why the curve 
 of rise and fall of current shown in Fig. 5 of pre- 
 vious chapter did not abruptly break and com- 
 mence, instead of gradually waxing and waning. 
 This effect is due almost wholly to self-induction, 
 and may be now readily understood. Self-induc- 
 tion may, in this connection, be looked upon as a 
 kind of electrical inertia. A most ingenious 
 device, known as a reactive coil, and used for turn- 
 ing lights up or down to any required brilliancy, 
 is dependent for its action solely upon the effects 
 of mutual and self-induction. It will be found 
 fully described later, for it is an excellent example 
 of these effects. 
 
 But we cannot deal with it until we have con- 
 sidered a matter on which we have not yet 
 to ached, but which is of prime importance: namely, 
 the presence of iron in the transformer; on this 
 entirely depends the efficient application of the 
 phenomena already discribed. 
 
 We have dealt with electrical currents and their 
 direction of flow: we must now introduce a new 
 factor, that of magnetic currents, heretofore spoken 
 of as lines of force. These are quite distinct from 
 electric currents, but they go hand in hand, the
 
 40 TRANSFORMERS. 
 
 one being dependent upon the other. Like electric 
 currents they have ( we suppose ) a direction, we 
 can scarcely say a direction of flow. Every 
 known medium, even a vacuum is a conductor, but 
 average iron is some 700 times better than any 
 other known substance, all other mediums being 
 about equal, save some comparatively rare metals 
 of the iron group, such as nickel, cobalt, etc. 
 
 The direction of movement of these lines of 
 magnetic force is at right-angles, to the direction 
 of current flow, as shown at Fig. 12. With a 
 single wire this "field " consists, as we have already 
 seen, of concentric lines of force rotating around 
 the conductor. These lines of force are of course 
 purely imaginary, but something must be assumed 
 to render the phenomena of electro-magnetism 
 capable of being grasped. A very pretty demon- 
 stration of the presence of the force which we 
 term magnetism may be made by thrusting a wire 
 through a card, and then sprinkling the card freely 
 with fine iron filings. On passing current through 
 the wire, the filings will arrange themselves in 
 concentric circles around the wire, exactly as we 
 have described and shown in Chapter I. 
 
 Magnetic resistance is, in its way quite as dis- 
 astrous to efficient operation, as is electrical ; 
 the magnetic conductivity, or as it is termed the 
 permeability of the medium through which the 
 lines of force are to be forced or circulated, has
 
 THEORETIC CONSIDERATIONS. 41 
 
 everything to do with the economical operation 
 of all electro-magnetic apparatus. Thus in Fig. 
 12, No. 2, a single coil of live wire, the circulation 
 of magnetic lines will be about as indicated by 
 the arrows. Now if this coil be simple wire 
 wound up hollow, or on wood, or brass, or in fact 
 any substance other than metals of the iron group, 
 then it will take, roughly speaking, some 700 
 times the electrical energy in the coil to maintain 
 a given strength of magnetic circuit that it would 
 if the magnetic circuit were of iron of sufficient 
 sectional area. The intensity of the magnetic circuit 
 is generally expressed by the number of "Kapp 
 lines " per square inch of section, in other words 
 the strength of the magnetic circuit, is supposed 
 to increase according to the proximity of the lines 
 to one another, that is, with the number within a 
 given sectional area. It is obvious therefore that 
 in constructing a transformer, the path for the 
 magnetic circuit must be a complete circuit of 
 iron* if the apparatus is to be efficient. 
 
 As the resistance of the magnetic circuit in- 
 creases with the length, this must of course be 
 kept as short as possible, consistent with good 
 mechanical construction. The permeability of 
 iron also varies very greatly; this may be expressed 
 by the number of Kapp lines per square inch, per 
 
 * This theory is contradicted provisionally by the inventor of the 
 " Hedgehog " transformer, for which see appendix.
 
 42 TRANSFORMERS. 
 
 ampere turn* in a given length of magnetic cir- 
 cuit. Therefore it is necessary for the best 
 results, to select iron of the greatest magnetic 
 permeability, that is iron in a given sectional area 
 of which the greatest number of Kapp lines can be 
 induced with the least expenditure of energy. 
 
 The number of lines per square inch does not 
 increase in a direct ratio with the number of 
 ampere turns; there is a point at which the number 
 of ampere turns must be vastly increased to secure 
 even a slight increase of magnetic strength, and 
 it is evident that it could be neither economical 
 nor advantageous to build commercial apparatus 
 to work, at, or above, this magnetic strength. The 
 point at which the greatest number of lines per 
 square inch can be induced with a relatively 
 economical expenditure of current, is obviously 
 the best magnetic strength to use in commercial 
 apparatus. This is known as the " working point " 
 of the iron, and varies greatly with the grade of 
 material. 
 
 There is also another point, at which the mag- 
 netic strength practically ceases to increase, 
 irrespective of any increase of ampere turns. 
 This is the " point of saturation." At this point the 
 number of Kapp lines per square inch of section 
 have reached the limit of that particular piece of 
 
 * One complete turn of wire carrying one ampere, or two turns 
 carrying J an ampefe, etc.
 
 THEOKETIC CONSIDERATIONS. 43 
 
 iron. The point of saturation is never reached in 
 commercial transformer practice. 
 
 In testing iron, all of the above characteristics 
 are clearly expressed by plotting curves, similar to 
 that shown in Fig. 13. 
 
 This curve shows wrought iron, having excel- 
 lent qualities for transformer construction; its 
 magnetic strength increases rapidly up to the 
 working point as compared with the rate of increase 
 of magnetic turns to obtain the result. It is of 
 course understood by the reader that the greater 
 the density of lines of force in the iron, the greater 
 the energy induced in the secondary. Herein lies 
 the whole question of transformer proportioning. 
 How many turns in the primary, how many turns 
 in secondary, and what section of iron to obtain 
 certain results ? This will be dealt with practi- 
 cally and theoretically further on. 
 
 Cast-iron gives the poorest results and is never 
 used for transformers. Soft rolled charcoal iron 
 has generally been considered the best, but the A 
 most recent practice seems to indicate that some |i\ 
 grades of very soft rolled steel have the highest J \ 
 efficiency of all magnetic mediums iron, however, 
 is generally used. 
 
 A transformer constructed with cast-iron would 
 be comparatively inefficient, or else the iron would 
 have to be so increased in sectional area to obtain 
 the necessary number of total Kapp lines, that
 
 TBAN8FOEMEB8. 
 
 I
 
 THEORETIC CONSIDERATIONS. 45 
 
 bulk and weight would be prohibitively large. In 
 other words, the working point of this iron is far 
 too low. 
 
 Before considering transformer losses, mention 
 should be made of the reactive coil already referred 
 to, for in it is embodied a practical illustration of 
 the application of all of the laws as yet referred to. 
 
 Fig. 14 shows one of these coils, which consists 
 of three main portions, a ring of iron highly lami- 
 nated (^4.), around which is wound a considerable 
 number of turns of wire, in series with the primary 
 or secondary circuit which it is to control. A 
 circular block or plug of iron (.#), a ^ so highly 
 laminated, fitting closely, but without touching, 
 into the hole through A, and supported and rotat- 
 ing in bearings at F. A solid copper casting ((7), 
 forming a complete secondary circuit of one turn, 
 rigidly fastened to D. C is provided with a 
 handle. D and are, of course, movable in the 
 bearings F, and can be placed in anjposition in 
 the arc of A. Now when C is turned until directly 
 opposite B, it is obvious that the lines of force set 
 up by the current in B will not pass around A, 
 but will take a short cut across through D, and 
 thus round and round, for lines of magnetic force, 
 like electricity, always take the path of least 
 resistance. Since the lines of force do this, C, in 
 its present position does not come within their in- 
 fluence, and remains inactive. Thus the current
 
 THA.NSFOKMERS. 
 
 in B reacts solely upon itself, and by self-induction, 
 as already explained, chokes down the flow of cur- 
 rent, and the lamps in its circuit glow very dimly. 
 As C is drawn over towards B, it begins to en- 
 close more of the lines of force, whereupon cur- 
 rent begins to flow in it, and it becomes a sec- 
 ondary. The secondary current in C, reacts upon 
 B, by " mutual induction," and permits more cur- 
 rent to flow, the lamps becoming brighter. The 
 nearer to B, C is drawn, the more lines it encloses, 
 the more current is set up in it, and the more it 
 reacts on B, till when C rests directly over JB, all 
 of the lines of force are brought into play, the 
 action and reaction of mutual, and self-induc- 
 tion about balance, and practically full current 
 flows through B to the lamps. 
 
 We have now to consider the various causes of 
 loss, incident to the transfer of energy, from the 
 primary to the secondary of a transformer. First, 
 and most serious is the C 2 R loss in the primary 
 and secondary. This is dependent solely upon the 
 Ohms resistance of the two coils, and is in accord- 
 ance with Ohms law with which we assume the 
 reader to be familiar. This loss is unavoidable, 
 but may be kept at a minimum by using as few 
 turns as possible to accomplish a given result, and 
 by using copper of great purity, and of ample sec- 
 tional area to carry the required current. The
 
 THEORETIC CONSIDERATIONS. 
 
 47 
 
 FIG. 14.
 
 4S TRANSFORMERS. 
 
 OR loss may be computed by means of Ohrns 
 law. after measuring the Resistance (in Ohms) of 
 primary and secondary coils. 
 
 The next important loss in the transformer is 
 due to Eddy or Foucault currents, which occur in 
 the iron, and to some extent in the coils them- 
 selves. As has been stated in the foregoing 
 chapter, the character of these currents is pretty 
 clearly emphasized by their name. They are very 
 similar in character to the local eddies and currents 
 which are to be found everywhere in running 
 streams of water. The iron of transformer cores 
 being of course conductive, current in the primary 
 coil has an inductive effect upon it, quite apart 
 and distinct from its magnetising influence ; this 
 effect is quite analagous in character to the 
 induction of current in the secondary, small local 
 currents of electricity being set up throughout the 
 mass of iron. Energy is naturally required to 
 generate these Eddy currents, and this energy is 
 taken from the primary and secondary circuits, 
 which induce them, whilst as these Foucault cur- 
 rents do no useful work, the force expended in 
 generating them is absolutely wasted. If steps 
 were not taken to prevent the presence of Eddy 
 ciirrents in the core, in dangerous quantities, they 
 would at once prove disastrous to the efficient 
 operation of the transformer; this would be the 
 case if the iron were one solid mass, for they would
 
 TIIEOHKTIC COXSIDEUATIOXS. 
 
 4!) 
 
 then have a free path through which to circulate. 
 Fortunately it is a comparatively easy matter to 
 guard against and prevent the generation of Eddy 
 currents in dangerous quantities by simply so sub- 
 dividing the core that the path through which 
 these eddys would natually flow, is broken at 
 short intervals by minute non-conductive spaces, or 
 spaces filled with matter of very high electrical 
 
 25. 
 
 resistance. Since the potential of Foucault cur- 
 rents is always very low, these insulating spaces 
 may be comparatively imperfect. 
 
 Since Foucault currents must always have a 
 direction consistently in the plane of direction of 
 the inducing current ; and since the magnetic 
 circuit follows a direction at right-angles to the 
 electric circuit, it is only necessary to build up the 
 iron core of thin sheets, or of a large number of turns 
 of iron wire partially insulated from one another 
 by tissue paper, corrosion, or other means, to
 
 50 TRANSFORMERS. 
 
 effectually prevent the presence of Eddy currents 
 in dangerous quantities, without breaking or 
 interrupting the path of the lines of force. This 
 may be better understood by referring to Fig. 15, 
 A being a coil of wire carrying an alternating 
 current, and wound around an iron ring, B. The 
 arrows, (7, indicate the plane of direction of 
 Foucault currents, whilst the arrows, -D, show the 
 plane of direction of the magnetic circuit. The 
 iron ring is made up of a number of punched 
 washer-like pieces of sheet iron, separated by tissue 
 paper or other means, and closely clamped together. 
 It will be seen that the path of the Eddy currents 
 is obstructed at short intervals by the layers of 
 insulation, whilst the path of the magnetic lines 
 is closed and unobstructed. The same would hold 
 good if the core was constructed of a bundle or 
 coil of iron wire, and such a core was quite largely 
 used in early practice, but a laminated core of 
 sheet iron is now almost universal, because, aside 
 from economy, the magnetic discontinuity of a 
 closed circuit wire core is obvious, besides which, 
 plates or punchings are much more readily 
 clamped firmly together. 
 
 Practically all of the energy which is present 
 in the core in the shape of Foucault currents, is 
 expended in heating the iron ; this heat is objec- 
 tionable, since it warms the copper wire of primary 
 and secondary, and since the resistance of copper
 
 TIIEOUETIC CONSIDERATIONS. 51 
 
 increases with the rise of temperature, the C 2 R 
 loss is magnified, and it becomes necessary to use 
 somewhat larger wire than would be required if 
 there were no heating. 
 
 Minute Eddy currents are also induced within 
 the copper of primary and secondary, being quite 
 distinct from the prime currents, but, except in 
 transformers, having secondaries of large current 
 capacity and containing therefore, considerable 
 masses of solid copper, their influence is almost 
 negligable, tending, however, to increase the tem- 
 perature of the coils. There are several excellent 
 formulae for computing the losses due to Foucault 
 currents. Being somewhat complicated and in- 
 volved, however, they have been omitted.* 
 
 Foucault current losses, in combination with 
 other heating effects, may be closely and easily 
 computed by means of the calorimeter, which will 
 receive further mention a few pages later. It is 
 obvious that, in view of the prejudicial effect of 
 heat, due allowance should always be made for it, 
 and such opportunity for radiation be provided as 
 is consistent with mechanical construction. 
 
 Hysteresis, the last of the loss producing forces 
 to be considered, is far less clearly defined in 
 character, and much more difficult to grasp, than 
 any of the effects as yet treated of. It is only 
 
 * For an able theoretic treatise on Eddy currents, we would refer 
 the reader to the Phil. Mag., ( England ) Jan. and Feb. 1884.
 
 52 TRANSFORMERS. 
 
 very recently that hysteresis has been individual- 
 ized and separated from other influences of similar 
 effect. 
 
 Hysteresis may probably justly be considered 
 as being the direct outcome of magnetic inertia. 
 Loss by hysteresis may, in fact, be looked upon as 
 the energy used in overcoming a kind of internal 
 friction between the molecules of iron, which, 
 according to the accepted theory, change their 
 position every time the polarity of the iron is 
 reversed, and this theory may the more readily be 
 accepted, since the loss due to hysteresis decreases 
 with the increase of mechanical vibration, which 
 serves to agitate, and therefore assists in moving 
 the molecules. Let us suppose a piece of iron to 
 be so enormously magnified that the eye can 
 distinguish the molecules, one from another, the 
 whole mass having the appearance of an enormous 
 number of particles grouped together and held in 
 the form of a homogeneous mass only by molecular 
 attraction, as is really the case. When this mass 
 is subjected to the influence of an electric current, 
 it becomes magnetically polarised, that is, each 
 molecule has a north and a south pole. Now if the 
 current be reversed in direction, the magnetism 
 necessarily reverses, not, as it might seem, by the 
 molecules being first demagnetized and then mag- 
 netized in an opposite direction, but by turning all 
 of the molecules half-way round, so that their ex-
 
 THEORETIC CONSIDERATIONS. 53 
 
 isting poles may coincide with the reversed con- 
 ditions.* This reversal is necessarily accomplished 
 by the expenditure of a certain amount of energy, 
 which results directly in heat, due ( we assume ) 
 to molecular friction. This waste of energy, is the 
 loss due to hysteresis. Hysteresis losses naturally 
 increase with the frequency of the alternations, 
 and almost in a direct ratio. 
 
 The heat losses due to Foucault currents, and 
 to hysteresis, can most conveniently be ascertained 
 by the Calorimeter test. The principle of this 
 method is simply the measurement of the rise in 
 temperature due to these effects, whose value is 
 then established by comparison with a scale of 
 temperatures due to known expenditures of energy. 
 The transformer, or other piece of apparatus in 
 which these losses are to be measured, is placed 
 within a closed box, whose temperature is normally 
 fixed, owing to its being packed in ice, or sur- 
 rounded by circulating water or air. The total 
 rise in temperature above normal, of the air within 
 the box, then represents the total heat losses. 
 
 This method, which is extremely exact, is not 
 more fully described, as it is commonly used in the 
 most careful and exact tests, as carried out in the 
 leading technical schools. 
 
 * It must be borne in mind that we are treating of a theory, which is 
 however, generally accepted.
 
 54 TRANSFORMERS. 
 
 The commoner method (though less exact) is to 
 measure the total losses, by simply ascertaining the 
 watts expended in the primary, when the second- 
 ary is idle, due allowance being made for magnet- 
 ism and C 2 R losses. The per cent of loss (at full 
 load) in the average transformer of today varies 
 from about 10 to 15 per cent in very small, to 3 to 
 5 per cent in large converters. It is probably 
 easier to design a transformer of, say 10,000 watts 
 capacity, with an efficiency of about 97.5 per cent, 
 than a 250 watt converter, with an efficiency of 90 
 per cent. For this reason the fewer the number 
 and the greater the size of the transformers, for 
 a given number of lights, the greater the efficiency 
 of the system, provided, always, that the converters 
 are working at full load in both cases. A con- 
 verter is, of course, most efficient at full load, for 
 the waste (except the C 2 R losses) is practically a 
 fixed quantity at all loads. For this reason trans- 
 formers should not be installed, having a capacity 
 in excess of what they are called upon to supply. 
 Further information relative to this will be found 
 in chapter VI. 
 
 The reader may note that the question of lag 
 has been neglected. This has been done intention- 
 ally, and after due consideration, as purely techni- 
 cal matters do not seem to be in keeping with the 
 character of the work in hand.
 
 MATHEMATICAL CONSIDERATIONS. 55 
 
 . 
 
 '// 
 
 CHAPTER III. 
 
 THE THEORY OF THE TRANSFORMER MATHEMATI- 
 CALLY CONSIDERED. 
 
 WHILST it has been deemed advisable to avoid, 
 as far as possible, all purely technical matters in 
 this work, it could scarcely be considered complete 
 without at least a brief consideration of the math- 
 ematical theory of the converter. After due con- 
 sideration Hopkinson's Formulae for the Trans- 
 former, as set forth by Prof. S. P. Thompson, has 
 been selected, as being at once the simplest and 
 the most direct. 
 
 The following equations are not, in the remotest 
 sense, original in any respect. They are presented 
 purely for the convenience of the reader, and in 
 the simplest form possible. A knowledge of alge- 
 bra only is presumed. 
 
 The following symbols have been assumed: 
 A Sectional area of core in square centimeters. 
 EI The whole applied electro- motive force in 
 
 primary. 
 
 C Current in amperes in external circuit. 
 Ci Current (absolute C. G. S. units) in primary,-
 
 56 TRANSFORMERS. 
 
 C 2 Current (absolute C. G. S. units) in secondary. 
 H Intensity of magnetic field. 
 L! Coefficient of self-induction in primary coil. 
 L 2 Coefficient of self-induction in secondary coil. 
 1 Length of magnetic circuit in core. 
 M Number of magnetic lines per square centi- 
 meter. The magnetic induction. 
 N Total number of lines of force traversing the 
 
 core, 
 p Coefficient of transformation ratio between 
 
 windings. 
 
 R Resistance of external circuit in ohms. 
 Ri Resistance of primary in ohms. 
 R 2 Resistance of secondary in ohms. 
 r 2 Internal resistance of secondary in ohms. 
 Si Number of turns of wire in primary coil. 
 S 2 Number of turns of wire in secondary coil. 
 t Time measured in seconds. 
 p Coefficient of magnetic permeability of iron. 
 
 It is, of course, understood by the reader that IT 
 is the accepted symbol for the ratio of the circum- 
 ference to the diameter of a circle, which equals 
 about 2 7 2 of the diameter; or, more exactly, 3.14195.* 
 Hopkinson's theory of the transformer has, as its 
 basis, the magneto-motive forces; or, to state it 
 more clearly, the applied magnetism-creating 
 energy at work in the iron core. Considered as 
 
 * It is well to remember the following expressions : Circumference 
 of a circle = 2jrr. Area of a circle = rrr 3 . r, denoting the radius.
 
 MATHEMATICAL CONSIDERATIONS. 57 
 
 resulting from the algebraic sum of the ampere 
 turns in the primary and secondary, from this the 
 voltages resulting from the variations in the mag- 
 netic induction of the core, are deducted. 
 
 Hopkinson first considers the magneto-motive 
 force necessary to project N lines of force through 
 a core, having 1 length and ^ permeability this, 
 of course, being entirely dependent on the quality 
 of the iron. The total number of magnetic lines 
 
 __ magneto-motive forces. 
 magnetic resistance. 
 
 or we may, according to Hopkinson's formulae, 
 write : 
 
 If we know the cross sectional area of the iron 
 (A), and the number of lines per square centimeter 
 (B), we may say: 
 
 N=AB (,) 
 
 in which case the magneto-motive force is 
 
 Next state two equations for the pressures 
 created in the primary and secondary circuits, as 
 follows : 
 
 0=(r 2 +R 2 )i 2 -S 2 f =(r, +R a )i S,A*g ( 5 )
 
 58 TRANSFORMERS. 
 
 . K 2 being the resistance of the lamp circuit and 
 .r 2 the resistance of the secondary coil itself. We 
 now multiply the fourth equation by S 2 and the 
 fifth by Si, getting : 
 
 S,E 1 =S,R l ii-^3i(r,+R 1 )i, () 
 this, with the third equation, gives us : 
 
 i t j SiR!+S?(r 2 +R 2 ) } =S*E 1 -}-S 1 (r 2 +R 2 )(10Hl-M7r) 
 
 i,{SSRi+Sf(r,+R,)}= SiS-, 
 
 and 
 
 AdB _ (r,+B,)S 1 E 1 J_ 10H1 
 
 A dt S|R 1 +Stfr t +B i '~T 4, ' 
 
 The second term ( 8 ) may be ignored, since HI 
 is insignificant in comparison with jjj^ii ; it being 
 only the difference between Siii and S 2 i 2 (which 
 are large factors), that is needed as a magneto- 
 motive force, which is small, provided the permea- 
 bility of the iron is great as is customary in actual 
 practice.* 
 
 We have, therefore : 
 
 AdB dN (r a +R J )S 1 E, ( -\ 
 
 ""dt dt S}R,+SKr a +R s ) ^ 
 
 bearing in rnind that : 
 
 ^F^ffjg (10) 
 
 *It is not generally customary to carry the number of lines much 
 above 10,000 lines per square centimetre.
 
 MATHEMATICAL CONSIDERATIONS. 59 
 
 we have : 
 
 Since the effective electro-motive force in the 
 secondary circuit (E 2 ) is equal to (r 2 -{-R2)i2, it 
 follows'that : 
 
 Eigf GO 
 
 g^+r.+R, ' 
 
 Assuming (as is practically true) that the 
 voltage at the terminals of the primary coil of the 
 transformer is constant, we may take E. and .R., as 
 relating purely to that which is internal in the 
 primary coil ; and it is quite plain that the action 
 of the converter is to reduce (or increase) the 
 pressure in direct proportion to the ratio of the 
 winding, for example, to reduce a primary pressure 
 of 1,000 volts to a secondary E. M. F. of 50 volts 
 we might^say, 50 : 1,000 : : 100 : x or 2,000 turns of 
 primary^to 100 turns secondary, or 20 to 1, follow- 
 ing a like argument in all required secondary volt- 
 ages. We also see from the foregoing arguments 
 that the effect of the transformer is to add to the 
 resistance of the secondary circuit a term equal to 
 the resistance of the primary, reduced in proportion 
 to the square of the ratio of the windings. It is 
 also obvious that, owing to the negative sign in
 
 60 
 
 TRANSFORMERS. 
 
 equation number twelve, the two currents are in 
 exactly opposite phases. 
 
 It is a general, and a pretty safe, rule to assume 
 that the weight of copper in primary and in 
 secondary are equal, this being so the following 
 table of relations and proportions will be found to 
 be correct. We quote this table from "Dynamo 
 Electric Machinery " by Prof. S. P. Thompson. 
 
 
 PRIMARY. 
 
 SECONDARY. 
 
 RATIO. 
 
 Windings 
 
 8, 
 
 S 2 
 
 P 
 
 Kesistance .... 
 
 TI 
 
 r 2 
 
 P 2 
 
 Self-induction. . . 
 
 L, 
 
 L 2 
 
 P 2 
 
 E. M. F 
 
 E, 
 
 E 2 
 
 P 2 
 
 Current 
 
 ij 
 
 is 
 
 P 
 
 Heat Waste. . . . 
 
 il'ri 
 
 i 2 2 r 2 
 
 1 
 
 As has been seen, the question of proportioning 
 the iron, depends almost absolutely on its perme- 
 ability, and the length of the magnetic circuit, the 
 following weight, however, is a fair average for the 
 various medium sizes of standard commercial 
 transformers.* Weight, per useful applied watt at 
 secondary terminals equals .103 Ibs. This includes 
 weight of copper, etc., which must be deducted to 
 get iron weight where required. Very small trans- 
 formers are heavier in proportion, whilst very 
 large ones are generally much lighter. 
 
 * 15 to 30 lights.
 
 MATHEMATICAL CONSIDERATIONS. 
 
 61 
 
 A 
 
 
 
 
 v 
 
 
 
 
 
 
 
 
 
 
 0) 
 CO 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 A, 
 
 
 
 
 
 
 
 
 
 3 
 
 _A^ 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 X 
 
 
 \ 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 V 
 
 ^ 
 
 ^ 
 
 
 
 9 
 
 
 
 
 
 
 s 
 
 (, 
 
 Ll 
 
 ? 
 
 s 
 
 i 
 
 Vj 
 
 =? 
 
 
 
 
 

 
 62 TRANSFORMERS. 
 
 It is of course, understood that the efficiency of 
 a transformer vanes "With the""loat!r'the least 
 efficient point being the lowest load, and the most 
 efficient, the highest. TKe efficiency is graphically 
 shown by plotting the results _of tests as an 
 efficiency curve. A good example of this is shown 
 at Fig. 16. 
 
 The horizontal lines representing the per cent of 
 efficiency, and the. verticals-lines, the load in 
 thousands of watts. In this particular case it will 
 be noted that -the- maximum efficiency shown is 
 97 per cent, the remaining 3 per cent representing 
 the losses. The potential of the secondary also 
 varies somewhat with the load, it being highest on 
 very low loads, and lowest on full loads. The 
 average variation between one lamp and full load, 
 being about 1.9 volts, in average sizes of conver- 
 ter.
 
 EVOLUTION OF TKAXSFOKMEIi. 
 
 CHAPTER IV. 
 
 EVOLUTION OF THE ELECTRICAL TRANSFORMER. 
 
 IN these clays the alternating system has come 
 into universal use. The high potential current of 
 the dynamo is carried great distances with but 
 little loss, and is reduced to the low pressuie 
 necessary for house lighting, by means of a trans- 
 former one of a dozen or more kinds all of them 
 efficient, sure in their action, and entirely reliable. 
 Transformers are sold as a mere commercial article, 
 and lighting companies order a dozen of them as a 
 cook might a dozen eggs. 
 
 In view of all these' facts, it is, perhaps but 
 natural in us to forget that the success of the 
 transformer, in fact the discovery of its principles, 
 is a matter of comparatively recent date, and that 
 it has gone through very many changes before 
 reaching its present comparatively perfect form. 
 Many people, if asked who first invented and used 
 the induction -coil, would answer, if they were only 
 casual readers on the subject "Ruhmkorff." It is 
 very easy for us in these days to forget Faraday 
 and his discoveries. Still easier to forget that the 
 origination of the transformer is due to him. Yet
 
 64 TRANSFORMERS. 
 
 the original transformer of Faraday embraced all 
 of the essential features of the best transformers 
 of today. 
 
 Fig. 17, represents the original "Perfected In- 
 duction Coil" of Faraday. A is a ring of cast-iron, 
 
 B 
 
 FIG. 17. 
 
 B the secondary, C the primary. It will readily 
 be seen that we have here the most essential 
 feature of the modern transformer, namely, a 
 closed magnetic circuit. 
 
 The Ruhmkorff induction coil was invented and 
 patented in 1842 (see Fig. 18 ). 
 
 Its construction is familiar to all a straight rod 
 of iron forming a core, a primary wound upon it, 
 and a secondary upon that. 
 
 This coil was designed, and is very extensively 
 used for inducing high from low potential inter-
 
 EVOLUTION OF TRANSFORMER. 65 
 
 mittent or interrupted currents, in connection with 
 batteries, in laboratory work, and also for medical 
 and electro-chemical purposes. 
 
 The Faraday and Ruhmkorff coils, are types of 
 the two classes of converter prevailing today the 
 opened and the closed circuit transformers. It 
 will readily be seen however that the Ruhmkorff, 
 having a straight core, must complete its energetic 
 circuit through the air ; thus its magnetic circuit 
 is of high resistance and presents a strong contrast 
 to the Faraday coil, which has a complete mag- 
 
 FIG. 18. 
 
 netic circuit of iron, and hence of low resistance. 
 Generally speaking, the lower the resistance of the 
 magnetic circuit, the greater the efficiency of the 
 transformer, provided other parts are properly 
 proportioned. The converters of the Faraday type 
 are therefore the more efficient, and the trans- 
 former of today is almost universally of this class. 
 To shorten and lower the resistance of the mag- 
 netic circuit is one of the chief aims of the modern
 
 66 TKAJJSFOKMEKS.- 
 
 transformer builder. This is done in four ways; 
 first, by. making the circuits entirely of iron ; 
 second, by shortening these magnetic circuits as 
 much as possible ; third, by increasing the sec- 
 tional area as much as is consistent with weight ; 
 and fourth, by using iron of the greatest magnetic 
 permeability. With these features always in view, 
 and due care with regard to insulation, and the 
 best ventilation possible, a transformer with prop- 
 erly proportioned windings, can be made extremely 
 efficient. It is interesting to note the various 
 changes and improvements which the transformer 
 has passed through in attaining its present high 
 character. 
 
 Jablochkoff and Goulard & Gibbs, used trans- 
 formers of the Ruhmkorff type in the first distri- 
 bution which was attempted by the alternating 
 system. These early transformers were used in 
 series, a practice which has long since been discon- 
 tinued except in connection with one or two 
 systems of street lighting, and for special pur- 
 poses. Probably the first improvement on the old 
 Faraday coil (and it was a very slight one), was 
 made by Kennedy in '83 ; he employed the same 
 large iron ring, with alternate windings of primary 
 and secondary, the improvement being in the ring, 
 which was of iron wire instead of cast-iron, thus 
 gaining greater magnetic permeability, it being 
 wrought, iron highly laminated. Ferranti some-
 
 EVOLUTION OF TRANSFORMER. 67 
 
 what improved this a little later, by winding strip 
 iron around the periphery over the coils. This 
 somewhat reduced the magnetic resistance, since 
 it increased the cross-section of iron employed ; it 
 also served to inclose any magnetic leakage which 
 might occur. The merits of this device were 
 questionable however, the additional iron neces- 
 sarily increased the weight, whilst the ventilation 
 
 FIG. 18 A. 
 
 being reduced, heating naturally resulted, and the 
 liability of a "burn out" increased. 
 
 Stanley a little later made a transformer, which 
 he claimed to be a marked improvement over any- 
 thing then existing. He made a ring of finely 
 laminated iron of the shape shown in Fig. 18 A,
 
 68 TRANSFORMERS. 
 
 This was an adoption of the Paccinotti ring arma- 
 ture. It is difficult to say what were the advantages 
 expected to result from this construction. The 
 leakage from the teeth of the ring was anything 
 but desirable, a considerable amount of magnetic 
 energy doubtless being thus dissipated. It how- 
 ever served to give an idea to Dick and Kennedy, 
 
 FIG. is B. 
 
 who in '86 introduced a transformer which showed 
 the first really vital improvements since the days 
 of Faraday. 
 
 Starting with Stanley's Paccinotti ring, they 
 wound the periphery with thin sheet iron, the 
 result being a transformer in appearance like Fig. 
 18 B. A coil of considerable efficiency and correct
 
 EVOLUTION OF TRANSFORMER. 
 
 principle. Much of the efficiency which should 
 have been gained by the reduction of the length 
 of the magnetic circuits, was sacrificed by the 
 magnetic resistance due to these circuits not fol- 
 lowing the direction of the lamination in the 
 peripheral iron. 
 
 But at this point in the history of the transfor- 
 mer, the question of cost began to demand serious 
 attention. Up to this time all of the closed cir- 
 cuit transformers had hand wound coils. The 
 shape of the iron punchings made them wasteful 
 and expensive. The iron on the periphery was 
 difficult to wind, and any repairs necessitated 
 tearing the whole thing to pieces. But though too 
 expensive for commercial use, the Dick and Ken- 
 nedy transformer of '86, was in some respects 
 excellent in principle. The magnetic core was 
 cut up into a number of short circuits of theoreti- 
 cally low resistance, as indicated by the arrows in 
 Fig. 18 B. The magnetic lines of force were quite 
 well enclosed, opportunities for magnetic leakage 
 being almost entirely absent, and the ventilation 
 was good. A number of tests of this transformer 
 carefully conducted, proved its efficiency to be 
 fairly high. There is a further objection to this 
 coil however, which applies equally to all of the 
 early ring-shaped transformers. The space occu- 
 pied was quite disproportionate to the work done. 
 This, together with the difficulty and expense of
 
 70 TRANSFORMERS. 
 
 winding, eventually led to the abandonment of the 
 ring transformer in its original form of an endless 
 jointless iron ring. 
 
 One of the best, perhaps the very best, of the 
 early ring-shaped transformers was presented to 
 the public in 1885, by Messrs. Zipernowski and 
 Deri. In this transformer, which electrically was 
 excellent, but mechanically very faulty, the posi- 
 tions of coils and iron were, so to speak, reversed. 
 The primary and secondary coils, both thoroughly 
 insulated being wound into a kind of solid core, 
 which was over-wound with a heavy layer of iron 
 wire. Owing to the shortness and abundant sec- 
 tional area of the magnetic circuit, this transfor- 
 mer had excellent self-regulation, and owing to 
 the large radiating surface of the iron, the heating 
 effect was considerably reduced. Mechanically, 
 the construction was obviously open to many 
 objections. 
 
 About the date of the perfecting of the Ziper- 
 nowski & Deri transformer, the more modern 
 "block shaped" class of converter commenced to 
 attract attention. 
 
 These transformers are of many kinds, no one 
 differing very essentially from another except in 
 details of construction, proportioning, efficiency 
 and workmanship. This class of transformers 
 generally has the coils and iron arranged about 
 as shown in the accompanying drawing, Fig. 18 C.
 
 EVOLUTION OF TRANSFORMER. 
 
 71 
 
 Thus the coils are entirely surrounded by lami- 
 nated iron save at their ends, whilst the magnetic 
 circuits are comparatively short and in two direc- 
 tions, the whole apparatus being mechanically 
 simple, and easy to assemble. 
 
 Block transformers are those chiefly in use at 
 the present day, and whilst the various makes 
 vary in detail, their general plan is similar. The 
 efficiency of the modern transformer is very high, 
 
 \ / 
 
 FIG. 18 C. 
 
 and its regulation is admirable. It may be 
 taken as an unquestionable fact that more light 
 is being distributed today by means of the conver- 
 ter than by all of the direct current systems com- 
 bfnecT There are still some strong supporters of 
 the open magnetic circuit transformers of which 
 the " Ruhmkorff coil " is a type, and by careful 
 construction and proportioning, some remarkable 
 results have been obtained from time to time.
 
 72 TRANSFORMERS. 
 
 A notable example of this is the English " Hedge- 
 hog " transformer of Swinburne, which attracted 
 so much attention some time since.* 
 
 As a class however, the open circuit trans- 
 formers from their very principle never have and 
 never can become generally successful. There 
 are one or two transformers now in use of rather 
 unique design, which show no especial feature of 
 excellence, and which form no link in the chain of 
 improvement which we have attempted to follow, 
 but which are none the less reliable and efficient, 
 though perhaps less so than others of the present 
 day. A good example of these is the Ferranti 
 converter, which is referred to elsewhere. 
 
 The chief direction of improvement in transfor- 
 mer construction today, tends first, toward greater 
 efficiency, and secondly toward the bettering and 
 simplifying of case and fixtures, such as fusing 
 devices. 
 
 Whilst we can scarcely look for the 105 per cent 
 efficiency, which some printed matter has now and 
 then seemed to hint at, still the transformer, of 
 1892 has crept up very close indeed to the other 
 side of the 100^. 
 
 * See Appendix.
 
 TRANSFORMER CONSTRUCTION. 73 
 
 CHAPTER V. 
 
 TRANSFORMER CONSTRUCTION. 
 
 IN considering the transformer and its structure 
 from a purely mechanical point of view, the student 
 is at once confronted with a multitude of forms r 
 each possessing its own especial advantages. Few 
 it must be confessed, can be found, which do not 
 show some noticeable defects. In preparing this 
 chapter it has been our endeavor to classify the 
 different steps in the construction, treating them 
 progressively, rather than as a whole. 
 
 The general character of the transformer is at 
 once determined by the form of core employed, 
 and in this portion of the apparatus, good mechani- 
 cal construction, probably does more to influence 
 the excellence of the results than in any other 
 portion of the work 
 
 On pages 74-75 are shown some typical forms of 
 cores, all more or less in use at the present day. 
 
 No. 1 is without doubt the most typical and 
 characteristic form of the modern transformer, 
 the "block-shaped" converter of the preceding 
 chapter, A and B are respectively, the iron and the
 
 74 
 
 TRANSFORMERS.
 
 TRANSFORMER CONSTRUCTION. 
 
 no. s.
 
 76 TRANSFORMERS. 
 
 winding. A is laminated, in the direction shown 
 in cut, which is of course, a section. It will be 
 noticed at once that the coils in this form of con- 
 verter are easy to wind and most compact. Some 
 manufacturers wind the primary and the secondary 
 into two coils of equal internal diameter, placing 
 them upon the iron side by side, others again wind 
 either the primary or secondary of greater internal 
 circumference so that the one may fit snugly over 
 the other ; whilst in some instances the two are 
 combined in one solid mass. In this latter case 
 the primary or secondary is first wound upon the 
 mandril and completed, this then in its turn, is 
 overwound with due insulation, upon which is 
 placed the second coil, the secondary generally 
 forming the outer layer. It is obvious that in this 
 form of converter, the iron must be divided at 
 some point to permit of the introduction of the 
 coils. 
 
 Very many different methods have been, and, 
 in fact, are still, in use for joining the iron, the 
 object, in all cases, being to obtain as good a 
 contact as possible between the abutting ends, and 
 thus avoid the introduction of unwarranted mag- 
 netic resistance resulting from a gap or break in 
 the path of the lines of magnetic force. Some 
 manufacturers prefer to break each plate of the 
 lamination at the same line, thus permitting the 
 two blocks formed by clamping the plates
 
 TRANSFORMER CONSTRUCTION. 77 
 
 together, to abut as two solid masses, whilst others 
 break the plates at alternate lines, thus overlap- 
 ping each break with the solid portion of the 
 next sheet above it. 
 
 A clear idea of what is meant may be had by 
 glancing at cut 21. In the latter method the 
 iron necessarily has to be built up within the coils, 
 which probably renders the assembling of the 
 transformer somewhat more laborious. In either 
 case a very perfect magnetic joint is secured by 
 proper care, and at this point as in almost every 
 
 other detail of construction, care in the manufac- 
 ture bears a very important relation to the final 
 efficiency obtained. Cut 22 shows some of the 
 commoner lines at which the cores of block-shaped 
 transformers are broken. In transformers of 
 medium and large size, it is a very common practice 
 to wind several coils, each of primary and second- 
 ary. 
 
 Cut 2, Fig. 19, shows us another form of trans- 
 former in use at the present day. This as well
 
 78 
 
 TRANSFORMERS. 
 
 as No. 8, Fig. 20, ( which it closely resembles) is a 
 ring-shaped transformer, having the iron (which 
 we can scarcely term a core in this case) on the 
 outside. In transformers of this type, the primary 
 and secondary are formed into a solid internal 
 ring, the two circuits being of course duly separ- 
 ated one from the other. Upon this ring of wire 
 is placed the iron. In the case of type 8, this con- 
 
 32. 
 
 sists of iron wire overwound upon the coils by 
 hand, a tedious and expensive operation. 
 
 This form of construction is seldom, if ever used. 
 Commonly the iron consists of sheets or plates, as 
 in the block form, and as before, the plates have, of 
 course, to be broken at some line, to permit of the 
 free introduction of the winding. 
 
 This form of transformer presents the merit of 
 a short magnetic circuit. The external circum- 
 ference of the ring being of course greater than 
 the internal, it follows (the plates of iron being of
 
 TRANSFORMER CONSTRUCTION. 79 
 
 uniform thickness) that the peripheral edges of 
 the plates must be somewhat separated, even 
 though the inner edges be clamped very closely 
 together, and this is a mechanical difficulty to be 
 overcome, although it is conclusive to very perfect 
 lamination. 
 
 Cut 4, Fig. 19, shows us a section through a 
 transformer of rather unique construction. The 
 iron in this case consists either of bundles of very 
 thin soft hoop iron well bound together, or, in a 
 few isolated instances of soft iron wire. A num- 
 ber of bundles of this iron being brought together 
 parallel, and in the same plane, they are overwound 
 and bound together by insulation at their central 
 portion. Over this insulation is wound the 
 secondary, and over this again is placed the pri- 
 mary, generally in the form of ready wound coils, 
 due insulation being interposed. The soft iron is 
 then turned back and over from each end, the ends 
 of the strips lapping one over the other, till the 
 middle of the bundle is reached, when the last two 
 ends are turned back and made fast ; the remain- 
 ing half of the iron is then turried back similarly 
 in the opposite direction, the iron, when in posi- 
 tion, enclosing the coils as in the block trans- 
 former. 
 
 This design presents some merits, chiefly 
 mechanical, probably the most marked being the 
 ease with which injured coils may be removed and
 
 80 TRANSFORMERS. 
 
 replaced. Where iron wire is used for the core, 
 the device differs somewhat, in that the iron is 
 turned back around the coils in all directions, in- 
 stead of in two directions only, and in this case, as 
 in most wire-cored converters, the coils are gen- 
 erally circular. Wire-cored transformers of this 
 type are, however, only made in small sizes and 
 for especial purposes. As a type, this general style 
 of converter has been somewhat popular abroad, 
 and has been adapted to very high voltages, large 
 capacities, and step-up purposes. 
 
 No. 3, Fig. 19, shows us another form of the 
 popular block-shaped transformer. It will be 
 noticed that the iron only encloses one-half of the 
 wire, hence the converter is, generally speaking, 
 less efficient for the same weight of iron than the 
 usual block form. Transformers of this type are 
 used solely for special and experimental purposes. 
 Its only merit lies in its extreme simplicity. 
 
 In cut 7, Fig. 20, we have the modern form of 
 the original Faraday "Perfected Induction Coil," 
 a typical ring-shaped converter. The iron core is 
 of wire, wound upon a form and bound closely 
 together with insulating material, which serves to 
 isolate it from the coils. The primary and second- 
 ary are over-wound upon this, necessitating much 
 hand labor. At times one-half of the ring is 
 devoted to the primary winding and the other to 
 the secondary, as shown in the illustration. Some-
 
 TRANSFORMER CONSTRUCTION. 81 
 
 times the two circuits are divided into a number 
 of short coils, placed alternately, and again one 
 coil is wound over the other. Transformers of 
 this type are not common in practice, as they 
 present numerous disadvantages, not the least of 
 which is in construction. They have been used, 
 however, in connection with series lighting, and 
 for special converters of small capacity. 
 
 Types of the open circuit transformer are shown 
 in Fig. 20, Nos. 5 and 6. The core is almost in- 
 variably of wire, straight and non-continuous, the 
 magnetic circuit being completed through the air. 
 No. 6 is the popular form of the early Ruhmkorff 
 coil, already mentioned in previous chapters. Its 
 general plan of construction is so simple as to need 
 no comment, more especially as it bears no especial 
 relation to electric lighting, although in the early 
 days it was used in series distribution. The same 
 modifications of winding are used in this, as in the 
 ring transformer described as No. 7. Although 
 quite typical and very useful for many purposes, 
 transformers of this class may safely be put aside 
 as essentially inefficient. 
 
 Cut No. 5 illustrates the general plan of con- 
 struction of the so-called " Hedgehog " transfor- 
 mer. Notwithstanding its similarity to No. 6 this 
 converter may be considered as a type, its charac- 
 teristic being the manner in which the ends of 
 the core are finished. The iron wire is permitted
 
 TRANSFORMERS. 
 
 to extend considerably beyond the coils, the wires 
 being bent into a radiating form, so that ach 
 individual wire is separated from its neighbors, the 
 whole having the general appearance of a brush. 
 This construction serves to equally disseminate 
 the lines of force through the surrounding space, 
 and results, according to the inventor, in especial 
 efficiency. This theory deserves attention, and is 
 the subject matter of an appendix. 
 
 One of the most important points in transformer 
 construction is the selection of the iron. Gener- 
 ally speaking, it is safe to say that the selection of 
 the iron is one of the most important preliminaries 
 of transformer construction. Not only is the 
 purity of the iron important, but also its texture, 
 fibre and malleability. Iron of great purity should 
 be selected. In other words, iron that is as free 
 as possible from carbon, phosphorus, silicon, etc. 
 This may be determined by chemical analysis. A 
 high quality of Swedish soft sheet is probably as 
 good as anything that can be selected, although 
 some of the very high grade English irons are 
 excellent. The metal should be as soft and uni- 
 form as possible, that is, it should be thoroughly 
 well annealed. The iron which can be bent the 
 greatest number of times without breaking may, 
 as a rule, be safely selected as the best, provided 
 the analyses have proved it to be pure, which, with 
 rare exceptions, will be found to be the case.
 
 TKA.N.-J OJi.MEK CONSTRUCTION. 83 
 
 Such an iron should have rather high tensile 
 strength, and should show a high per cent of elon- 
 gation before breaking. Recent practice has 
 pointed towards certain grades of low steel, as 
 possessing many good qualities for transformer 
 construction, but suitable steel is probably more 
 difficult to obtain than is good iron. 
 
 The iron being selected, the next step in the 
 manufacture is the punching of the plates (we are 
 dealing with the now almost universal "block" 
 type of converter). This is accomplished upon a 
 power stamp or press, by means of dies. The 
 sheet metal is placed upon the lower die, which is 
 held stationary, whilst the upper die is brought 
 down upon it, punching out the requisite shape. 
 Where both an outer shape, and a hole, are to be 
 cut out, it is a common practice to accomplish the 
 operation by a single blow, instead of, as formerly, 
 using two dies and presses. This is done by using 
 what is termed a "leading " die, one form of which 
 is shown in Fig. 23. There are many modifications 
 of this plan, and it is used in the manufacture, 
 from sheet metal, of many hundreds of articles. 
 There is almost always some automatic device for 
 freeing the metal after punching. A good set of 
 dies should leave only a very light burr at the 
 edges of the punchings, whilst poor dies will leave 
 a heavy one. 
 
 The plates being punched, the next step is the
 
 84 
 
 TRANSFORMERS. 
 
 building up of the core. The direction of the 
 lamination is already understood by the reader, it 
 having been dealt with in Chapter II, when dis- 
 cussing Eddy currents. If the transformer core 
 is of the class which has abutting edges at the 
 break, then the iron is built up quite independent 
 of the coils. If of the class which has overlapping 
 edges, then it is built up within the coils. In 
 either case the general plan is the same. 
 
 As has been stated, next each sheet of iron must 
 be placed a layer of insulating material. Only a 
 
 very low insulation is needed, for the potential of 
 Foucault currents is invariably low. Practice 
 differs as to the best and easiest method of accom- 
 plishing this end. Some makers shellac or enamel 
 the plates, and then bake them in an oven to set 
 the coating. Others rust, or oxidize the plates, 
 rust being an insulator, or, rather, a semi-conductor 
 of sufficiently high resistance for this purpose. 
 Probably, however, thin tissue paper is more uni-
 
 TRANSFORMER CONSTRUCTION. 85 
 
 versally used for this purpose than any other 
 insulating material. This paper is sometimes 
 glued or pasted to the iron, but more commonly 
 laid in loosely, being cut to the required shape and 
 size, as are the iron plates. 
 
 In building up the plates are laid within a rack 
 or former, which they fit, or if there are holes in 
 the plates, as there commonly are, for the binding 
 bolts which are frequently used to hold the block 
 together, then the plates are laid over pins, which 
 correspond to these holes, or even over the bolts 
 themselves. The necessary rack, or pins, being 
 provided, the plates are placed in position, a layer 
 of paper or other insulation alternating with each 
 layer of iron. 
 
 When the requisite number of plates have been 
 placed, the whole pile is subjected to heavy pres- 
 sure, thereby reducing it as nearly as possible to 
 the nature of a solid mass. After this operation 
 it is tightly bound together, either by binding 
 bolts passing through all of the plates, or by 
 clamps on the outside. Where the binding bolts 
 are used they should be surrounded by an insu- 
 lating bushing, interposed between them and the 
 core, and should have a washer of insulating 
 material under both the head and the nut. 
 
 The iron being now ready for assembling, 
 we may turn our attention to the winding and 
 insulation of the coils. This is a comparatively
 
 TRANSFORMERS. 
 
 simple matter, and is carried out in the same 
 general manner, whether the coils (primary and 
 secondary ) are made up independently, or together, 
 to fit over one another, or to be placed side by side. 
 
 Insulation is, generally speaking, the most vital 
 consideration, but this important question will be 
 considered later. 
 
 The coils are wound in a lathe, upon a former 
 or mandril, having perpendicular ends like a spool, 
 to hold the wire on, and to determine the width of 
 the coil. One of these ends is removable, to 
 permit of the coils being taken off when completed. 
 Double cotton covered wire is used, generally 
 spoken of as D. C. C. wire. One end of the wire, 
 to be wound up, being made fast, the "former," 
 or mandril, is set in motion in the lathe, the wire, 
 as it is wound on, being drawn through tension 
 pulleys to insure its being wound on straight and 
 level. The wire is watched and guided at the 
 mandril, to insure its being wound on smoothly 
 and progressively. When a layer is completed it 
 is quite customary to give the surface a thin coat of 
 shellac before winding the wire back over itself. 
 In winding the primary coils of large high voltage 
 converters, it is quite customary to insert some 
 kind of insulating material, in addition to the 
 shellac, between each layer, to prevent the differ- 
 ence of potential between the layers from short- 
 circuiting, inner and outer turns.
 
 TRANSFORMER CONSTRUCTION. 87 
 
 When the requisite number of turns have been 
 wound onto the former, the wire is cut off and the 
 coil removed, it having previously been tied firmly 
 together at four or five points, by pieces of twine 
 or tape placed on the mandril before the winding 
 began. With the exception of the care required 
 in insulating, primary coils are far less difficult to 
 wind than secondary, owing to the fact that the 
 wire is much finer and more manageable. As the 
 heavy wire necessary for the secondaries of large 
 capacity, low potential transformers is often most 
 unmanageable, amounting almost to rod, copper 
 tape is at times substituted for this heavy wire, 
 with good results. The most popular plan, how- 
 ever, is to wind a number of wires in multiple, or 
 even to place several coils in multiple. 
 
 Too much care cannot be exercised to prevent 
 any portion of wire, however small, which is bared 
 of its insulation, from being wound on. Such a 
 spot, if present, would almost certainly make 
 trouble eventually by coming in contact with other 
 bare metal, and causing a short circuit. When 
 such a point is discovered in the wire it should be 
 carefully wrapped with adhesive rubber tape and 
 shellaced before being wound on. If the rubber 
 tape is too bulky, as it frequently will prove to be, 
 then narrow silk tape or ribbon may be used, it 
 being made to adhere by means of shellac. If it 
 becomes necessary, for any purpose, to make a
 
 TRANSFORMED. 
 
 joint in the wire of an incompleted coil, it should 
 always be soldered, as well as twisted, and should 
 be over-wound with tape as already described. 
 Above all things, great care should always be used 
 to see that there are no sharp edges or projections 
 at the joint, which might chafe through the insu- 
 lation and expose the metal. 
 
 In the case of high potential transformers of 
 large capacity, it is a common and excellent prac- 
 tice to wind the primary in several separate and 
 distinct coils, which can afterwards be connected 
 in series, thus preventing any two turns of wire, 
 with a considerable difference of potential, from 
 crossing. If they did so the insulation would 
 almost certainly chafe through and a short circuit 
 result 
 
 All of the above-mentioned contingencies having 
 received due and proper attention, the coil may be 
 removed from the "former" without fear of future 
 trouble. 
 
 The next operation is the taping of the coils. 
 An ordinary cotton tape, from an inch to an inch 
 and a half in width, is commonly used for this 
 purpose. One end of the tape is made fast to the 
 coil with melted shellac, the covering then being 
 wound on and over the coil progressively. When 
 the coil is intended to carry, or be brought into 
 close proximity to high pressure, it is preferable to 
 insert a layer of insulating material below the
 
 TRANSFORMER CONSTRUCTION. 89 
 
 outer covering of tape, mica is probably the best 
 substance for this purpose. It should be put on over 
 a layer of heavy shellac, and should have another 
 coating of thin shellac before the tape is applied. 
 Mica is very brittle, and great care should be 
 exercised to prevent its breaking and chipping off 
 at corners and curves. Other substances are 
 frequently used instead of mica, such as asbestos 
 paper, vulcanized fiber, shellaced paper, etc., but 
 whilst far more easy to manipulate than mica, 
 they are inferior to it, owing to the fact that they 
 absorb moisture more or less rapidly, which 
 speedily destroys their insulating qualities until 
 they are again dry. 
 
 No mention has been made as yet of what is 
 done with ends of primary and secondary coils. 
 They are, of course, brought out through the tape, 
 and this should be done in such a way, that their 
 positions may be comparatively remote from one 
 another, since contact between the two terminal 
 wires, especially of the primary coil, even though 
 well insulated, would be most dangerous, and 
 might at any time result in a short circuit and a 
 burn-out. The terminal wires should be carefully 
 taped and shellaced, both for the sake of the in- 
 sulation, and the protection which the covering 
 affords to the wire. It is a common plan to bring 
 out the ends of the primary coil at one end of the 
 transformer, and the secondaries at the other. A
 
 90 TRANSFORMERS. 
 
 sufficient length of wire is left to permit of making 
 the necessary connections to the binding posts, or 
 terminal blocks, which will be presently mentioned. 
 
 On completion, the coils are given a final coat- 
 of shellac, after which they are sometimes baked in 
 an oven, to thoroughly dry them out and harden 
 the shellac, but this is not always necessary, since 
 the completed transformer is generally baked as a 
 whole before being enclosed within its case. 
 
 The finished coils are next placed upon the iron 
 in their proper position, where they should fit 
 closely, due insulation (dependent upon the pres- 
 sure for which they are intended) being interposed. 
 Mica, rubber, vulcanite, fibre, asbestos and wood, 
 are the substances commonly used for insulation 
 in transformers. 
 
 There is often a space between iron and wind- 
 ings, at the bend at each end of the coils, caused 
 by the coils not making a square turn and lying 
 flat upon the core. A wooden block, or wedge, is 
 frequently inserted at these two points, to ren- 
 der the whole solid. Teak, or failing that, a 
 very resinous pine is probably the best wood as 
 regards insulating qualities, for use in transformer 
 construction. The coils being in position and the 
 two portions of the core being brought together 
 and firmly fastened, either by clamping screws or 
 bolts, the transformer is complete and ready for 
 baking, to which process it is at once submitted,
 
 TRANSFORMER CONSTRUCTION. 91 
 
 being placed in an oven which is at a temperature 
 as high as the insulation will bear without injury. 
 This thoroughly dispels all moisture and sets the 
 shellac hard, so that it resembles lacquer. On 
 being removed from the oven, and cooled, the 
 transformer only lacks testing, casing and con- 
 necting, to be ready for commercial use. 
 
 A thin cast-iron box or case is used to enclose 
 the transformer. Although this case varies greatly 
 in details of construction, it is always water-proof 
 and usually has some provision for ventilation, a 
 ring bolt for lifting, lugs and holes for lag screws, 
 by which it may be fastened up. Terminal blocks 
 are set into it, but insulated very carefully from 
 it, and these blocks should be so placed as to pre- 
 clude any possibility of a short circuit between 
 them. The two ends of the primary and secondary 
 coils are fastened to these blocks from the inside, 
 and the ends of the leading-in wires from the main 
 circuit to the primary, and the leading-out wires 
 from the secondary to the lamps, are fastened into 
 these terminals by means of set screws when the 
 transformer is put into use. The greatest care 
 should be exercised to prevent the necessity of 
 crossing either the primary or secondary inside 
 leads to get them to their binding posts. This is 
 especially important in connection with the pri- 
 mary, as, if they cross, a short circuit is very likely
 
 TRANSFORMERS. 
 
 to result, if the insulation of the leads becomes 
 chafed at the point of contact. 
 
 It is a common practice to have an internal 
 arrangement whereby fuses, which can readily be 
 replaced, are interposed between the terminal wires 
 of the coils and the binding posts, sometimes in 
 connection with the primary coil only and some- 
 times with both. Such arrangements are generally 
 mounted on porcelain, and placed within a sepa- 
 rate compartment of the transformer case, which 
 can readily be opened to replace a burned out fuse. 
 Secondary transformer fuses, though often found, 
 are, generally speaking, uncalled for. Although, 
 till recently, fuses have almost invariably been 
 placed within the transformer, it is probably better 
 to have them outside instead, in an accessible po- 
 sition, and this plan is rapidly finding favor. 
 
 Many manufacturers have their transformei 
 cases so constructed that the fuses cannot be 
 reached without breaking, or opening the circuit. 
 As a safety device this is excellent, as many 
 accidents have occured from carelessness in put- 
 ting in new fuses on " live " circuits. 
 
 One of the most radical, and marked improve- 
 ments in modern transformer construction, is the 
 introduction of oil insulation, the most perfect and 
 efficient insulation known for this class of work. 
 Where this is used, the entire transformer is sub- 
 merged in a heavy low grade oil, specially
 
 TRANSFORMER CONSTRUCTION. 93 
 
 prepared for the purpose, and closely resembling 
 a low grade of cylinder oil. This dense fluid 
 speedily penetrates every crevice of the transform- 
 er, and becomes a perfect insulating film at every 
 point of danger, precluding besides any possibility 
 of moisture. 
 
 If a puncture or short circuit occurs at any 
 point, from coil to coil, or from coils to iron, the 
 oil at once penetrates and re-insulates the crevice, 
 rendering the point where the discharge took place 
 as perfect as before, thus it may be said that trans- 
 formers, with oil insulation, are self -repairing. 
 
 Converter testing in practice, is a comparatively 
 simple matter. The points to determine, are the 
 thorough insulation of the coils from one another, 
 and from the iron and casing. The fitness of the 
 transformer to carry its normal load indefinitely, 
 its ability to bear its normal voltage, or a greater 
 pressure, without breaking down ; and the ratio of 
 conversion, to make sure that the right number of 
 turns are present, or that the voltage of the second- 
 ary is correct, with normal primary pressure. 
 
 To test the insulation, a magneto capable of 
 ringing through a high resistance is applied. The 
 contact points of the magneto should be applied 
 first to the end of the primary and secondary coils, 
 then to primary and iron, and to secondary and 
 iron. If the insulation is perfect no ring will be 
 obtained. Some makers test by attaching primary
 
 TRANSFORMERS. 
 
 to one side and secondary to the other side of a 
 very high potential circuit, of say five or six 
 thousand volts. The insulation should stand this 
 without breaking down. 
 
 To test the ability of the transformer to carry 
 normal load, and bear normal pressure, it is only 
 necessary to place it on a standard primary pres- 
 sure, load it to its full capacity of lamps and leave 
 it there for some hours, testing the voltage on 
 primary and secondary at the end of the run to 
 determine that the ratio of conversion is correct. 
 
 The converter having stood these tests, is fin- 
 ished and ready for shipment and service. 
 
 We have endeavored to carry the reader with us 
 through the path of construction in its main feat- 
 ures, from the transformer's component parts to 
 its completion, and it now only remains to us, to 
 add a few points of great importance, which should 
 be borne constantly in mind, to obtain the best 
 results. 
 
 1st. Never permit two wires with any great dif- 
 ference of potential to cross each other. Such 
 construction invites a short circuit. 
 
 2d. Allow ample sectional area of copper, (800 
 to 1000 circular mills per ampere, ) otherwise the 
 converter will run too warm, and give poor results. 
 
 3d. Better use too much than too little 
 insulation. 
 
 4th. Do not attempt to work the iron at too
 
 TRANSFORMER CONSTRUCTION. 95 
 
 high a point of saturation, that is at too high an 
 induction, it is not economical. 
 
 5th. Clamp the iron and bind the coils rigidly, 
 if this is not done the transformer may hum. 
 
 6th. Ventilation is an excellent thing, but 
 moisture is dangerous and distinctive. 
 
 7th. Keep the primary and secondary terminals 
 well apart. 
 
 8th. Do not let any iron chips, or filings, get 
 into the transformer, they will make their way 
 through almost any insulation by magnetic pro- 
 pulsion, and perhaps cause a short circuit.
 
 96 TRANSFORMERS. 
 
 CHAPTER VI. 
 
 THE TRANSFORMER IN SERVICE. 
 
 IF the theory of the transformer is in some of 
 its details somewhat intricate, its application is 
 simple and easy to a degree. We have in the 
 foregoing pages treated here and there of various 
 matters which have direct bearing on the subject 
 of this chapter, and we therefore shall here limit 
 the discussion to various important points of detail, 
 rather than to any general plans of arrangements 
 of circuits, which are undoubtedly already well 
 understood. 
 
 The question of the location of a transformer 
 is an important one, and is scarcely amenable to 
 any general rules which could be formulated. 
 
 Local custom has much to do with the positions 
 selected for installing converters. In Europe it 
 is the general practice to install them within the 
 building whose lighting service they are to main- 
 tain, and for this reason the transformers manu- 
 factured in Europe are not, as a rule, enclosed 
 within waterproof cases. With us, however, it is 
 the custom to install converters on the outside of 
 buildings, or in some position in the open air, and
 
 TRANSFORMER IN SERVICE. 97 
 
 this is generally demanded by insurance require- 
 ments. 
 
 There is an opportunity for the exercise of 
 much judgment in selecting the capacity of the 
 transformer required to do certain work. It should 
 always be borne in mind that the nearer a trans- 
 former operates to full load the greater its effi- 
 ciency will be, but that an overload is a bad thing, 
 endangering the transformer and causing poor 
 regulation. If the transformer is to operate on a 
 dwelling house circuit, then the number of lamps 
 upon its secondary will be considerably in excess 
 of the actual capacity required, for there will 
 seldom, if ever, be an occasion when all of the 
 lamps will be in service at the same time. In a 
 case of this kind a close estimate of the number of 
 lamps likely to be in use at one time will represent 
 the capacity required. If a store, hall, or factory 
 is to be lighted it is probable that all of the lights 
 will be in use at the same time and, therefore, the 
 total number of lights installed will represent the 
 size of converter required. 
 
 Large transformers are more efficient than small 
 ones, therefore they should always be used when 
 possible. It frequently happens that a number of 
 stores or dwelling houses requiring light are near 
 together. In a case of this kind one transformer 
 can be made to do the work for the Avhole number, 
 it being installed as nearly as possible at the centre
 
 98 TRANSFORMERS. 
 
 of consumption, and its secondary being divided 
 and carried to the various installations, as sepa- 
 rate secondary circuits. This practice, if generally 
 followed, will result in economy, though, of course, 
 the secondary must not be carried through any 
 considerable distance, or the C 2 R loss will become 
 excessive and wasteful. 
 
 There is an opportunity also for the exercise of 
 much judgment in determining the best secondary 
 potential to be used. If the total secondary circuit 
 is short, then 50 to 52 volts is the best pressure to 
 use, for, though the amount of copper required 
 will be considerable, lamps of this potential are 
 somewhat more efficient. If the length of second- 
 ary wiring be considerable, then it is generally 
 preferable to use a pressure of 100 to 104 volts, 
 because of the resultant saving in wire, and the 
 reduction of C 2 R loss. Where a large building is 
 to be wired and the total length of secondary 
 wiring is great, it is a frequent and excellent prac- 
 tice to bank two transformers and do the wiring 
 on the three-wire plan. Directions for doing this 
 will be found later in this chapter. 
 
 Where the number of lights to be carried on one 
 circuit is greater than the capacity of a single 
 converter, a number may be arranged in multiple, 
 as shown in Fig. 24. When this is done however, 
 the transformers used, should always be of the 
 same capacity, and the same style. It must be
 
 TKANSFOKMER IN SERVICE. 
 
 J J 
 
 
 
 
 r ) 
 
 1 fr 
 
 
 i l 
 
 
 
 
 
 
 
 fe 
 
 
 r 1 ^ 
 
 
 ri ^ 
 
 
 d 
 
 99999T <> 
 
 <> 
 
 a 
 
 u y 
 
 V 
 
 i 
 
 .....? j ..... 
 
 _9. 9 9 9 $ ^ 99999 
 Tig. Z5.
 
 100 TRANSFORMERS. 
 
 borne in mind, that if the primary fuse of one of 
 several banked transformers "blows" then the 
 secondary of that transformer becomes a load 
 upon the rest of the bank, in addition to the por- 
 tion of the load it was carrying. Under this 
 extra strain, the fuses of the remaining trans- 
 formers, will probably blow also, and darkness and 
 confusion result. 
 
 For this reason it is generally preferable to 
 divide all circuits when possible, using single 
 transformers, so that all of the lamps in a building, 
 may not be dependent on a single source. When 
 banking is unavoidable, it is an excellent plan to 
 insert a large main circuit fuse on the united 
 secondary of the whole bank, calculated to blow 
 before the individual transformer fuses, thus, if all 
 of the lamps go out, they can be re-lighted by 
 the insertion of a single fuse instead of one in 
 each transformer. It is of course understood that 
 there are always a number of sub-circuit fuses 
 between the lamps and the bank. 
 
 At times it becomes necessary to so bank trans- 
 formers, as to get an increased potential. For ex- 
 ample, a building may be ready wired on a basis 
 of 100 volts, the lamps being made for 100 volts 
 also, whilst the only transformers available are 
 designed to give a secondary potential of 50 volts. 
 In this case the transformer must be banked with 
 two secondaries in series, as shown in Fig. 25.
 
 TRANSFORMER IN SERVICE. 101 
 
 It is well to remember that when two or more 
 transformers are connected with their secondaries in 
 multiple, their united output is equal to the sum of 
 their rated current at their normal potential. 
 
 When banked with their secondaries in series, 
 their output is equal to the sum of their rated po- 
 tentials, with current equal to the rating of one 
 transformer. (All of the converters, of course, 
 being of the same capacity.) 
 
 To bank transformers to operate on the three- 
 wire plan they should be arranged with their pri- 
 maries in multiple, as is usual, and their secondaries 
 in series, with the neutral, middle, or third wire 
 taken off between them, as shown in Fig. 26. 
 
 This is generally done with 100 volt transform- 
 ers, and saves much wire, the greater portion of 
 the energy being distributed at 200 volts, and the 
 lamps (which are 100 volts) burning two in series, 
 except when more are in use on one side than on 
 the other, in which case the middle wire takes care 
 of the difference. The two outside wires are to be 
 figured upon on the basis of 200 volts, and the 
 middle wire should have from one-half to one-third 
 of the carrying capacity of the outside wires. The 
 lamps must, of course, be distributed about equally 
 between the two sides. The main fuse on the 
 middle wire must have a capacity in excess of its 
 calculated load, for if it burns out many of the 
 lamps on one side will, as a rule, be destroyed by
 
 102 TKANSFOEMERS. 
 
 i 
 
 H 
 
 a 
 
 rig. 
 
 26. 
 
 y U
 
 TRANSFORMER IK SERVICE. 103 
 
 the passage of excessive current, due to the un- 
 balanced condition of their resistance. The main 
 fuse on the middle wire is frequently altogether 
 omitted. 
 
 Transformers are frequently incorrectly ar- 
 ranged on the three-wire plan. A common but 
 very incorrect method is shown in Fig. 27. The 
 error here is in the primary connections, opposite, 
 instead of similar wires (as when correct), being 
 connected to the same side of the primary. There 
 is no saving of wire in this plan, over the ordinary 
 two-wire arrangement. It has no advantages, and 
 should be avoided. 
 
 Some commercial transformers are so con- 
 structed that by changing simple internal connec- 
 tions, they will give either 52 or 104 volts at the 
 secondary as required, the output representing the 
 same amount of power in either case. This plan 
 is an excellent one and of great convenience. 
 
 Transformers should be placed, when out of 
 doors, either 011 the side of a building, or, still 
 better, upon a pole when available, the latter posi- 
 tion being preferable on account of the vibration. 
 Jar, within certain limits, is an excellent thing for 
 transformers, and should be sought rather than 
 avoided, since, as we have seen in Chapter II, it 
 tends to reduca the hysteresis loss, and to render 
 the converter more efficient. 
 
 Many transformers are provided with hooks, or
 
 104 TRANSFORMERS. 
 
 a kind of elbow at their upper end, by means of 
 which they can be suspended from the arm of a 
 pole, or from a cleat attached to a wall. This is a 
 great convenience and saves much labor. Gener- 
 ally speaking, however, they are fastened up by 
 means of lag screws. 
 
 If it can be avoided, transformers should never 
 be placed against metal, such as a tin roof or metal 
 shingled wall, as there is always danger, in a case 
 of this kind, of a "' ground " on one side of the line 
 and consequent trouble. It is also a bad practice 
 to lay converters on their sides or backs, in posi- 
 tions where they are liable to be exposed to wet. 
 This is frequently done, but is none the less to be 
 avoided. The upper end of the case is designed to 
 withstand rain, while the remaining portions of 
 the box are not always absolutely waterproof in 
 case of a heavy downpour, and if water penetrates 
 within, trouble is more than likely to result. 
 
 In some localities, where violent electrical 
 storms are prevalent, notably the Southern coun- 
 tries, considerable damage is at times caused to 
 transformers by lightning. In such cases it is not 
 necessary that a lightning discharge should "strike" 
 the converter or its wires, though this at times 
 occurs. Often the trouble results from a secondary 
 current, induced by the lightning, generally in the 
 main primary circuit. The effect which this has upon 
 the converter varies greatly with the conditions.
 
 TRANSFORMER IN SERVICE. 105 
 
 At times it simply blows the primary fuses, at 
 others it melts a portion of the primary coil.* 
 More commonly, however, it punctures the insula- 
 tion, striking through either to the secondary coil, 
 the hon, or both, at times leaving no evidence of 
 the mischief which has been wrought; it is this 
 last contingency that is most to be dreaded, for it 
 may contain an element of clanger. 
 
 The primary having a path to the secondary on 
 one side may, if a " ground " occurs upon the other, 
 send a dangerous shock through the secondary, pro- 
 vided anyone should come in contact with a portion 
 of the secondary circuit in such a manner as to con- 
 nect it to the ground through themselves. Such a 
 combination of circumstances is of necessity rare, 
 however, and there is probably only one authentic 
 case of disaster on record, resulting from this state 
 of affairs. 
 
 It is in connection with punctures, due to 
 lightning, etc., that the system of oil insulation 
 is of greatest value, for every puncture is re-insu- 
 lated by the oil as soon as made, leaving the trans- 
 former practically as perfect as before, Another 
 safety device, which has been adopted in some 
 transformers, in connection with lightning dis- 
 charges, calculated to prevent injury to the coils, 
 is the grounding of the iron core. This is accom- 
 
 * This latter result is rare, but has been observed. When the fuses 
 blow, they must have drawn an arc, which maintained the circuit.
 
 106 TRANSFORMERS. 
 
 plished by connecting the laminated plates either 
 to water or damp earth by means of a copper wire. 
 Since lightning invariably seeks the shortest path 
 to the earth,* it will, when the core is grounded, 
 strike through the insulation to the lamination, 
 and thence go directly to earth, when if oil insu- 
 lation be present it will immediately re-insulate 
 the puncture, leaving the transformer in perfect 
 working condition. When the core is grounded 
 in this way it must, of course, be strictly insulated 
 from the case. It is naturally understood that the 
 potential of a lightning discharge is practically 
 irresistable. 
 
 The primary fuses in the transformer, and in 
 fact everywhere, should have a rubber covering 
 over the lead and tin wire of which they are made. 
 Primary fuses are now constructed in this way. 
 The rubber covering prevents the flash of a burn- 
 ing fuse from reaching beyond its proper limits, 
 and also tends to prevent the burning fuse from 
 establishing an arc between its terminals. 
 
 When not enclosed within the transformer, pri- 
 mary fuses should be placed out of doors, within 
 boxes manufactured for their reception, and in a 
 readily accessible position, conveniently near the 
 transformer. Both the main and sub-circuit sec- 
 ondary fuses should be placed indoors, in a position 
 where they can be readily reached. 
 
 * The shortest electrical path, i. e., the path of least resistance.
 
 TRANSFORMER IN SERVICE. 107 
 
 In very long distance distribution it has some- 
 times been found advantageous to introduce step- 
 up transformers, to raise the output of the dynamo 
 to a yet higher pressure (5,000, and even 10,000 
 volts), at which potential it is carried very great 
 distances with remarkably little loss, and on rela- 
 tively smaller wire, to the distributing point, where 
 it is again reduced to customary distributing pres- 
 sure, and carried over ordinary circuits to the 
 house transformers, the energy thus being trans- 
 formed three times before reaching the lamps. 
 This system is valuable in that it renders possible 
 the utilization of large water-powers, remote from 
 points of usefulness. The plan is quite popular in 
 Europe, but much less so in this country. Of 
 course the great difficulty to be met lies in pro- 
 viding the insulation requisite for such high 
 voltages. Further information relative to this 
 subject will be found in an appendix. 
 
 It will sometimes be noticed that transformers 
 emit a humming noise, which is occasionally so 
 loud as to be annoying. This will be especially 
 noticeable if the converters are carrying an over- 
 load. The noise is caused by the rapid vibration 
 of the iron plates of the lamination ; and sometimes 
 even by the coils themselves, resulting from the 
 rapid attraction and repulsion of the iron plates of 
 the lamination, and also of the individual turns of 
 the coils. The more tightly the plates are clamped
 
 108 TRANSFORMERS. 
 
 together, and the more closely the coils are bound, 
 the less the transformer will hum, in fact, well 
 constructed transformers seldom hum noticeably, 
 even upon considerable overloads. 
 
 There are several very important points in con- 
 nection with the installation of transformers which 
 should be carefully borne in mind. An endeavor 
 is here made to condense these into a few terse 
 and clear sentences. 
 
 1. Remember that the larger wires are the sec- 
 ondary connections and go to the lamps. 
 
 2. Do not put in the fuses till the transformer 
 is fixed in position. 
 
 3. Rubber covered fuses should be used on the pri- 
 mary circuit. 
 
 4. Do not handle live primary wires without 
 rubber gloves. It is dangerous. 
 
 5. Try to use only one hand when working about 
 live primary wires. It insures safety. 
 
 6. Locate all fuses where they can be readily 
 reached. 
 
 7. Place the main primary fuse out of doors if 
 possible, when not enclosed within the transformer. 
 
 8. Abide by the Underwriters' 1 Rules. (See Ap- 
 pendix.) 
 
 9. Avoid making all lights in a large building 
 dependent on a single bank of transformers. 
 Divide your circuits.
 
 TRANSFORMER IN SERVICE. 
 
 10. In banking use transformers of the same 
 type and capacity. 
 
 11. When banking connect similar leads of the 
 transformer to the same side of the primary line. 
 
 12. A slight amount of jar renders transformers 
 more efficient. 
 
 13. Large transformers are more efficient than 
 small ones. 
 
 14. Do not use larger transformers than neces- 
 sary. Transformers are most efficient on full load.
 
 110 TRANSFORMERS. 
 
 CHAPTER VII. 
 
 COMMERCIAL TRANSFORMERS. FERRANTJ. NA- 
 TIONAL. SLATTERY. STANLEY. THOMSON- 
 HOUSTON. WESTINGHOUSE. 
 
 IT has been thought best to devote a brief con- 
 cluding chapter to a few of the commercial trans- 
 formers which are to be found upon the market at 
 the present time. The transformers mentioned in 
 the following pages do not represent all that are 
 for sale at the present day by any means. They 
 have been selected either because of their promi- 
 nence and popularity, or because novel and typical. 
 They are all, with one exception, of American 
 manufacture. 
 
 The Ferranti Transformer is the only converter 
 of foreign manufacture of which mention is to be 
 made in this chapter. It has been selected as 
 typical, and a good example of European practice. 
 
 It will be noticed by glancing at the accompany- 
 ing illustration, Fig. 28, that this transformer is not 
 enclosed within a water-proof case, as is customary 
 with converters of American manufacture. Such 
 a case is not needed where transformers are to be
 
 COMMERCIAL TRANSFORMERS. 
 
 Ill
 
 112 TRANSFORMERS. 
 
 installed within buildings, as is usual in Europe. 
 This transformer is typical, being built on the plan 
 of type 4, Fig. 19, Chapter V. The frame which 
 holds and supports the actual converter is of cast- 
 iron, and is so constructed as to provide for stand- 
 ing the transformer upon the floor, a basement or 
 dry cellar generally being the location selected. 
 
 The primary and secondary terminals are at 
 opposite ends of the base, and are so constructed 
 that they cannot be tampered with, or the wires 
 loosened with an ordinary screwdriver. The 
 terminals are thoroughly insulated from the frame 
 by means of sulphur and glass insulation, poured, 
 while in a molten state, into the space between the 
 frame and each terminal block. No fuses are con- 
 tained within the converter, as it is intended that 
 they should be installed separately. 
 
 The iron used in the construction of these trans- 
 formers is extra soft Swedish sheet, and is 
 unusually thin. As it is not clamped together, 
 however, the iron is quite apt to be loose, and the 
 transformers have a tendency to hum. The gen- 
 eral plan of construction follows closely that 
 described under the head of No. 4, Chapter V, in 
 fact, this is probably the only commercial trans- 
 former of that type. Ferranti transformers have 
 been constructed for service on very high poten- 
 tials, 2400 volts being the commonest primary 
 pressure, while large central distributing trans-
 
 COMMERCIAL TRANSFORM Ki:-. 
 
 113 
 
 FIG. 29.
 
 114 
 
 TRANSFORMERS. 
 
 FIG. 30.
 
 COMMERCIAL TRANSFORMERS. 116 
 
 formers have been constructed for, and successfully 
 operated upon, 10,000 volts.* 
 
 This is the class of transformer used in connec- 
 tion with distribution from the celebrated Deptford 
 Station, near London. The chief merits claimed 
 for the Ferranti converter are high insulation and 
 the ease with which they can be repaired. 
 
 The National Transformer is manufactured by 
 the National Electric Manufacturing Co., of Eau 
 Glairs, Wisconsin. Its general appearance is 
 clearly shown by Fig. 29. The transformer is of 
 the ring type, described in Chapter V as No. 2. 
 The entire winding is surrounded by iron, all of 
 the wire in the transformer thus being active, and 
 high efficiency is claimed as the result. The true 
 character of the converter may be better under- 
 stood by referring to Fig. 30. 
 
 A novel feature is the fuse and connection box, 
 which is on the lower side of the case. The open- 
 ing of the fuse box door simultaneously breaks the 
 connection between the primary wires and the fuse 
 contacts, thus rendering it possible for the primary 
 fuses to be quickly replaced without danger of 
 shocks or short circuits, the transformer and its 
 secondary circuits being inactive when the fuse 
 box is open. This safety device is clearly shown 
 by Fig. 31. The case is so constructed as to be 
 
 * See Appendix. *
 
 116 
 
 TBANSFOKMKKS. 
 
 FlG. 31.
 
 COMMERCIAL TRANSFORMERS. 
 
 117 
 
 FIG. 32.
 
 118 TRANSFORMERS. 
 
 waterproof, and ample provision is made for ven- 
 tilation. 
 
 These converters are manufactured in sizes 
 ranging in capacity from two to one hundred 
 16-candle power lights. The safety device is not 
 furnished for transformers of less than five, or 
 more than fifty light capacity. The merits claimed 
 are high efficiency, safety, convenience, and close 
 regulation. 
 
 The Slattery Transformer. This converter is 
 manufactured by the Fort Wayne Electric Co., of 
 Fort Wayne, Indiana. It is a block-shaped trans- 
 former, belonging to the class described as No. 
 1, Chapter V. Its general character is clearly 
 shown by the accompanying cut, Fig. 32. 
 
 Both the primary and secondary leads enter at 
 the lower end of the case, being kept well apart, 
 however. The case is cast in two nearly equal 
 pieces, and is bolted together, as shown in the 
 illustration. Careful provision is made for proper 
 ventilation, thus insuring cool running. 
 
 The ratio of transformation adopted is either 20 
 or 10 to 1, 50 or 100 volts being the popular sec- 
 ondary potentials. Efficiency, simplicity, safety 
 and careful construction are the merits claimed for 
 this transformer. It is widely used and generally 
 gives excellent results. 
 
 The Stanley Transformer is manufactured by the 
 Stanley Electric Manufacturing Co., of Pittsfield,
 
 COMMERCIAL TRANSFORMERS. 
 
 119
 
 120 TRANSFORMERS. 
 
 Mass. This company devotes itself almost exclu- 
 sively to the manufacture of electrical converters. 
 The construction of the Stanley transformer is 
 clearly shown in the accompanying illustration, 
 Fig. 33. It is here shown with the fuse and con- 
 nection box open, showing the construction. No 
 secondary fuses are used. The primary fuses are 
 mounted on movable porcelain blocks, one of which 
 is shown removed from its position. 
 
 In inserting new fuses it is only necessary to 
 withdraw the blocks from their connections, which 
 breaks the circuit, after which a new fuse can be 
 set on the block without difficulty or danger. It 
 is customary for a station to have a number of 
 extra plugs, that there may be no delay in insert- 
 ing a new fuse. The same size of plug fits all 
 different sizes of transformers up to 100 lights. 
 The front plate of the fuse box is held in place by 
 two screws, and the lower flap is held up by a 
 thumb-screw, in contact with a lip cast onto the 
 front plate. The two fuses are entirely separated 
 by a porcelain partition in the fuse box, thus pre- 
 venting any possibility of a short circuit from one 
 to the other. The secondary connections are made 
 at the lower end of the transformer. 
 
 In replacing fuses the front plate is not removed, 
 but the lowei- flap, or door, is opened by loosening 
 the thumb-screw. The transformer is provided at 
 the back with a hook or elbow, forming part of
 
 COMMERCIAL, TRANSFORMERS. 121 
 
 the case, thus rendering it easily placed in position. 
 It is only necessary to hoist the transformer up by 
 the eyebolt and hook it over the cross-arm, of the 
 pole, or a timber on the side of a building. The 
 merits claimed for this transformer are high effi- 
 ciency, close regulation, small leakage, convenience 
 and safety. 
 
 The Thomson-Houston Transformer. This con- 
 verter is manufactured by the General Electric 
 Co., at their factory in Lynn, Mass. It is of the 
 block form, but embraces many novel features. A 
 clear idea of the general character of this converter 
 may be obtained by glancing at Fig. 34. The 
 leading-in wires for both primary and secondary 
 are at the top of the case, which is extremely con- 
 venient in making connections with circuits. 
 
 No fuses are placed in the converter. The pri- 
 mary fuses are placed within a primary switch and 
 fuse box, which is always furnished with all 
 transformers, and is to be placed in any convenient 
 position where the fuses can be readily renewed. 
 The secondary fuses are to be placed within the 
 building to be lighted, or near the point where the 
 secondary wires enter. The fuses being thus dis- 
 posed, it follows that the transformer case need 
 never be opened after the converter is installed. 
 
 This transformer is provided with the oil insu- 
 lation already referred to in previous chapters. 
 Glancing again at the illustration, Fig. 34, it will
 
 TRANSFORMERS. 
 
 FIG. 34.
 
 COMMERCIAL TRANSFORMERS. 123 
 
 at once be noted that the case is constructed in 
 the form of a kind of cup or pocket, the only 
 entrance being by means of a lid, or cap piece, at 
 the top of the case. Within this complete recep- 
 tacle the transformer is fixed, and when the appa- 
 ratus is installed the case is filled with oil supplied 
 for the purpose, thus completely surrounding and 
 permeating the transformer with a very perfect 
 self-renewing insulating medium. It is unnecessary 
 to discuss the merits and advantages of oil insu- 
 lation here, as they are fully considered in a 
 previous chapter. The amount of oil required 
 with each transformer varies from two quarts for 
 a 600 watt converter, to nine quarts in the 7500 
 watt size. 
 
 Another safeguard against loss of insulation is 
 added by the grounding of the core. The lami- 
 nated iron is fully insulated from the case, and is 
 connected thoroughly to earth by means of a 
 ground wire. Thus a lightning discharge, entering 
 the converter, would strike through the insulation 
 from coil to lamination, and pass down the ground 
 wire to earth, the puncture being immediately 
 re-insulated by the oil. This matter was thorough- 
 ly discussed in Chapter VI. 
 
 These transformers are wound to stand a test of 
 5000 volts without oil insulation. 
 
 The case of the Thomson-Houston transformer 
 is provided with hooks to permit of its being
 
 124 
 
 TRANSFORMERS. 
 
 FIG. 35.
 
 COMMERCIAL TRANSFORMERS. 125 
 
 readily installed upon cross-arms or transformer 
 timbers on houses. 
 
 The merits claimed for this converter are high 
 efficiency, perfect insulation, safety, long life, close 
 regulation, small leakage, convenience and sim- 
 plicity. 
 
 Previous to the introduction of the transformer 
 just described, which is known as Type F, the 
 Thomson-Houston Electric Co. manufactured a 
 transformer known as their Type E, an illustration 
 of which is shown in Fig. 35. This transformer is 
 widely used and generally popular. 
 
 The Westinghouse Converter is shown in accom- 
 panying illustration, Fig. 36, and is manufactured 
 by the Westinghouse Electric Co., of Pittsburg, 
 Penn. Like most of the other commercial trans- 
 formers it is block-shaped. 
 
 At the upper end of the case, and attached to 
 the front plate, is a fuse and connection box. This 
 contains a removable porcelain block, carrying the 
 primary fuses. The removal of this block breaks 
 the primary circuit, leaving the converter dead 
 while the fuses are being renewed. New fuses 
 may thus be placed with perfect safety and with- 
 out trouble. 
 
 The fuse box is opened by removing the front 
 plate, which is held in place by three thumb-screws, 
 as shown by the illustration. The front plate 
 cannot be dropped, being fastened to the converter
 
 126 
 
 TBANSFOKMERS. 
 
 by a chain. The secondary connections are made 
 at the lower end of the case, being remote and 
 thoroughly isolated from the primary circuit. 
 
 These transformers are manufactured in sizes 
 ranging from 250 to 6250 watts, and to give sec- 
 
 FIG. 
 
 ondary potentials of 50 or 100 volts, as is required. 
 The manufacturers claim for them high efficiency, 
 good regulation, high insulation, safety and care- 
 ful construction. 
 
 In describing the foregoing transformers we 
 have endeavored to confine ourselves strictly to 
 their more striking and conspicuous features. A
 
 COMMERCIAL TRANSFORMERS. 127 
 
 general description has, in every case, been ren- 
 dered unnecessary by the excellent cuts which we 
 have been enabled, through the kindness of the 
 various manufacturers, to present. 
 
 Although this may justly be considered as the 
 concluding chapter of this little work, there are 
 yet to follow a number of pages devoted to sun- 
 dry miscellaneous, but distinctly relevant, matters 
 which, while too important to omit, could scarcely 
 be classified under any of the previous headings.
 
 TRANSFORMERS. 
 
 APPENDICES. 
 
 1. HIGH VOLTAGES. 2. THE "HEDGEHOG" 
 TRANSFORMER. 3. THE WELDING MACHINE. 
 4. DIRECT CURRENT TRANSFORMERS. 5. STA- 
 TION TRANSFORMERS. 6. CONSTANT CURRENT 
 TRANSFORMERS. 7. UNDERWRITERS' RULES. 
 
 1. HIGH VOLTAGES. 
 
 REFERENCE has already been made to the use 
 of very high pressures obtained by means of step- 
 up transformers, for distributing large amounts of 
 power through very long distances with relatively 
 small loss. 
 
 The economy that would result from such a 
 system has long been understood, but the many 
 serious difficulties which have stood in the way of 
 its successful application have, until recently, with 
 isolated exceptions, caused its use to be confined 
 almost entirely to experiment. Among the earliest 
 experiments with high voltages from step-up trans- 
 formers (as distinct from static pressures), were 
 those conducted in London, just previous to the 
 building of the celebrated Deptford station. Their 
 purpose was to establish the feasibility of, and the
 
 APPENDICES. 129 
 
 necessary methods to be pursued in a system such 
 as that which subsequently resulted. 
 
 In the frontispiece of this volume a photograph 
 is re-produced, showing a discharge from the 
 10,000 volt circuit of a step-up transformer. 
 
 This experiment was to establish the distance 
 that this pressure would force current through an 
 ordinarily dry atmosphere. 
 
 The discharge depicted was between two copper 
 points, separated from one another by five-eighths 
 of an inch. Across this open space the enormous 
 pressure caused the electric fluid to leap without 
 difficulty. 
 
 Dry air being practically the best insulating 
 medium known, the difficulty of properly insulat- 
 ing such high potentials will readily be appreciated. 
 The flash at the light of the illustration is occa- 
 sioned by the blowing of a fuse immediately upon 
 the establishment of the circuit through the air. 
 There were, at the time these experiments were 
 first attempted, no thoroughly effective measuring 
 instruments for computing what voltages really 
 existed at such very high pressures. 
 
 The earliest method of definitely determining 
 that 10,000 volts had actually been obtained was 
 by illuminating 100 100-volt lamps in series. This 
 required, of course, 10,000 volts between the two 
 ends of the line of lamps, to bring them to full
 
 130 
 
 TKAXSFOKMEI5S.
 
 APPENDICES. 131 
 
 candle-power. Such an arrangement is shown in 
 the accompanying illustration, Fig. 37. 
 
 Many recent experiments, conducted to ascer- 
 tain the effects and results which could be obtained 
 with very high potentials, combined, as a rule, with 
 high frequency of alternations, have brought to 
 our knowledge a large number of very beautiful 
 and remarkable results. 
 
 Thus, with a very large induction coil or step-up 
 transformer, having an enormous number of turns 
 of very fine wire in the secondary, and a very rapid 
 rate of alternation being used, the two ends of the 
 secondary, though remote from one another, emit 
 a more or less brilliant scintillating light, surround- 
 ing the ends of the wire. This resembles in 
 general character and appearance that remarkable 
 natural phenomenon, which at times appears at 
 the ends of ships' masts and spars, known as St. 
 Elmo's fire. 
 
 Incandescent lamps, having only a single thread 
 of filament, become brightly illuminated on being 
 attached to one of the wires, while on sufficiently 
 intensifying these conditions a simple incandescent 
 lamp of peculiar construction may be illuminated 
 without being brought in contact with any wire 
 or source of current whatever, and without being 
 introduced into any magnetic field. This is 
 accomplished by introducing the lamp between two 
 metal plates, forming, perhaps, the sides of a small
 
 132 TRANSFORMERS. 
 
 room, and each being attached to one terminal of 
 the secondary already referred to. 
 
 An endless variety of experiments in this line 
 have been made, chiefly by Mr. Nichola Tesla, to 
 whom chiefly belongs the credit of investigation 
 in this direction. 
 
 The results obtained are very interesting, and 
 all tend to prove that the rules governing the 
 action of electricity as we understand them, while 
 fixed and certain, only apply within what are 
 probably quite narrow limits. These experiments 
 point to the future utilization of electricity for 
 illuminating purposes in ways which are not as yet 
 known to us, and which will probably be as greatly 
 in advance of present systems as our existing 
 methods are in advance of those of one hundred 
 years ago. 
 
 Mr. Tesla has successfully reproduced, on a small 
 scale, many of nature's marvelous electrical phe- 
 nomena, and it is probable that yet more wonder- 
 ful discoveries are still to come.* 
 
 2. THE "HEDGEHOG" TRANSFORMER. 
 
 The "Hedgehog" transformer is the invention 
 of Mr. James Swinburne, the well-known English 
 electrician, and was the outcome of his theory that 
 
 * For further information relative to these experiments we would 
 refer the reader to the Proceedings of the American Institute of Elec- 
 trical Engineers, 1891.
 
 APPENDICES. 133 
 
 the iron and hysteresis losses were more serious in 
 the transformer than the C 2 R loss. He argued, 
 therefore, that an open circuit transformer, with 
 greatly reduced iron, would be more efficient, even 
 though the copper would have to be greatly in- 
 creased to obtain the same induction. 
 
 The iron loss is without doubt reduced in the 
 construction adopted by Mr. Swinburne (See 
 Chapter V), for not only is the sectional area of iron 
 reduced, but being an open circuit transformer the 
 outside iron is entirely wanting, thus reducing the 
 total amount of iron to about one-third that used 
 in a closed circuit transformer of similar capacity. 
 
 Thus far the theory is correct, for the " Hedge- 
 hog " transformer has shown, in careful tests, that 
 its actual iron losses amount to considerably less 
 than two per cent of full load. The change in the 
 arrangement of iron, however, greatly increases 
 the loss in the winding, more turns being needed 
 to obtain the same results. So that the actual total 
 efficiency of the " Hedgehog," according to Mr. 
 Swinburne himself, proved to be about 87 per 
 cent. To somewhat reduce the magnetic resist- 
 ance incident to the use of an open iron circuit, 
 the ends of the wire core are spread out as pre- 
 viously shown in Fig. 20, Chapter V,. to equally 
 distribute the magnetic lines through a broad path 
 in the surrounding air, for it is generally accepted 
 as true that the magnetic resistance of air is rela-
 
 134 TRANSFORMERS. 
 
 tively low when the magnetic induction, or the 
 magnetic current through a given area, is small. 
 
 The construction of the " Hedgehog " trans- 
 former is simple in the extreme. The iron wire of 
 the core is built up around a brass, or gun metal 
 back-bone, one end of which is spread out to form 
 legs, and the other to hold a connection board. 
 The windings being placed over the iron, the whole 
 transformer is enclosed within an earthenware jar 
 or case. Earthenware was selected for this pur- 
 pose because, being an open circuit transformer, 
 an iron case would at once become magnetized by 
 the core, while a case of any metal would be 
 subject to Foucault currents and would waste 
 much energy. 
 
 3. THE ELECTRIC WELDING MACHINE. 
 
 An ingenious and valuable application of the 
 transformer principle is the Electric Welding Ma- 
 chine. This is a transformer so constructed as to 
 generate enormous current volume in the second- 
 ary, with very little pressure. The secondary coil 
 consists, as a rule, of but one turn of heavy cop- 
 per, generally a casting. One end of this is 
 movable, being usually made in the form of a slid- 
 ing carriage, moving upon the solid portion of one 
 end of the secondary. 
 
 Each end of the secondary carries a clamp or 
 vice. These clamps are placed directly opposite
 
 APPENDICES. 135 
 
 one another, one being fixed and the other moving 
 with the sliding carriage. The two pieces of metal 
 to be welded together being fixed in the clamps 
 opposite and parallel, as required, the movable 
 clamp is drawn forward toward the other, which, 
 brings the objects to be welded together into con- 
 tact, completing the circuit of the secondary 
 through them. 
 
 The current passing through the secondary turn 
 being greatly in excess of the true carrying capac- 
 ity of the articles to be welded together and, owing 
 to their very small length between the clamps, 
 their resistance not being sufficient to materially 
 cut the current down, they are speedily heated to 
 the melting point at the point of contact and high- 
 est resistance. The two ends are then forced to- 
 gether while molten, and the two objects are at 
 once welded into one. 
 
 The welding machine is made in many styles, 
 according to the purpose for which it is designed. 
 Wire, chain, pipe, rod, axles, tires, projectiles, etc., 
 are all welded successfully and with perfect ease. 
 Cast, as well as wrought metals, may be welded, 
 as also may unlike metals, such as copper and steel. 
 
 The machines are designed according to the 
 work they are to do, the larger and more powerful 
 ones containing many mechanical features of great 
 interest. The closing of the weld is generally 
 accomplished in the larger machines by means of
 
 136 TRANSFORMERS. 
 
 hydraulic pressure, and water circulation is pro- 
 vided to keep the secondary and movable carriage 
 thoroughly cool. The welding machine has a wide 
 field of usefulness, and though but a recent inven- 
 tion has already become well known. 
 
 '4. THE DIRECT CURRENT TRANSFORMER. 
 
 The statement has been made throughout this 
 volume that electrical transformation could be 
 accomplished only by means of alternating, pul- 
 sating or intermittent currents. This is, perhaps, 
 not literally true, although practically so. 
 
 There is a system, however, whereby transfor- 
 mation is accomplished in connection with direct 
 current, but only mechanically. Stated briefly, it 
 consists simply of an electric motor, operating on 
 a high potential circuit, and driving a dynamo 
 which generates at any potential that may be de- 
 termined upon. Thus, instead of energy being 
 transformed directly from electrical power of one 
 pressure to electrical power of another, it must be 
 converted first into mechanical energy or motion, 
 and this again to electrical power of the pressure 
 required. 
 
 Machines have been designed whereby both 
 transformations are accomplished in one piece of 
 apparatus, both the motor (primary) and dynamo 
 (secondary) windings being placed upon a single 
 armature, the same fields serving both for dynamo
 
 APPENDICES. 137 
 
 and generator. The armature is, of course, pro- 
 vided with two commutators and sets of brushes, 
 one for the motor and another for the dynamo 
 winding, the two being as absolutely distinct as if 
 carried upon separate armatures. 
 
 While necessarily somewhat less efficient than 
 true electrical transformers of correct design, this 
 electro-mechanical converter is, in certain cases, 
 extremely useful and valuable, for, as has been 
 previously stated, alternating current has not, gen- 
 erally speaking, been successfully applied to the 
 distribution of power for mechanical purposes. 
 
 Thus, when it is desired to carry power for the 
 operation of motors for long distances, as, for ex- 
 ample, when a water power in a remote location is 
 to be utilized to operate stationary motors, or street 
 cars, then the energy may be transmitted as direct 
 current of high potential through long distances, 
 and over light wire, with small loss, being reduced 
 to the required pressure by means of one or more 
 motor-transformers situated at a convenient centre 
 of distribution. 
 
 5. STATION TRANSFORMERS. 
 
 It is customary to have upon the switch-board 
 of every alternating dynamo a small transformer, 
 which is mounted upon a base or back, instead of 
 being enclosed within a case, and these are known 
 as station transformers.
 
 133 
 
 TKANSFOHMEKS. 
 
 FIG. 38.
 
 APPENDICES. 139 
 
 These transformers are constructed with special 
 care as regards regulation, etc., and have the same 
 ratio of transformation as the transformers chiefly 
 in use upon the line. 
 
 Volt-meters, to measure high potentials, are 
 delicate and expensive, consequently the switch- 
 board volt-meter is placed upon the secondary of 
 the station transformer and indicates the potential 
 of the secondary circuits. The volt-meter and the 
 two or three lights which serve to illuminate the 
 switch-board constitute the load carried by the 
 station transformer, this load being, of course, 
 practically constant. 
 
 An excellent illustration of a station transformer 
 is shown in the accompanying cut, Fig. 38. 
 
 6. CONSTANT CURRENT TRANSFORMERS. 
 
 In addition to the transformers which have been 
 mentioned, designed to operate upon constant cur- 
 rents, transformers have been successfully con- 
 structed to give a constant current and varying 
 potential at the secondary, with the primary upon 
 a constant potential circuit. 
 
 Such transformers are designed to operate arc 
 lamps upon incandescent circuits. This result is 
 obtained by especial design and construction, and 
 further mention of it is purposely avoided, lest it 
 should prove confusing and irrelevant.
 
 140 TRANSFORMERS. 
 
 7. UNDERWRITERS' RULES. 
 
 The following rules are an extract from the 
 Rules and Regulations adopted by the New Eng- 
 land Insurance Exchange and the Boston Fire 
 Underwriters' Union, for electric lighting, as at 
 present in force. 
 
 The rules in use throughout the various States 
 of the Union vary somewhat from these in detail 
 but are the same in general character. These reg- 
 ulations are of great importance and should be 
 carefully noted and closely followed. 
 
 Only such rules are presented here as apply 
 directly to the application of transformers. For 
 the complete text of the regulations the reader is 
 referred to the official publication. 
 
 SECONDARY GENERATORS OR CONVERTERS. 
 
 Converters must not be placed inside of any building. 
 They may be placed on the outer walls when in plain sight 
 and easy of access, but must be thoroughly insulated from 
 them. If placed on wooden walls, or the woodwork of stone 
 or brick buildings, the insulation must be fire-proof. When 
 an underground service is used, the converter may be put in 
 any convenient place that is dry and does not open into the 
 interior of the building ; this location must have the approval 
 of the inspector before the current is turned on. 
 
 The converter should be enclosed in a metallic or non- 
 combustible case. 
 
 If for any reason it becomes necessary that the primary 
 wires leading to and from the converter should enter a 
 building, they must be kept apart a distance of not less than 
 twelve inches, and the same distance from all other con-
 
 UNDERWRITERS' RULES. 141 
 
 ducting bodies. The insulation of the wire must be of the 
 very best. 
 
 Safety fuses must be placed at the junction of all feeders 
 and mains, and at the junction of mains and branches where 
 necessary, also in both the primary and secondary wires of 
 the converter, in such a manner as not to be affected by the 
 heating of the coils. Secondary wires, after leaving the 
 converter, will be subject to rules already given for services, 
 inside wiring, etc. 
 
 Any provision for grounding the secondary circuit by 
 means of "film cut-out" or other approved automatic device 
 will be approved. A permanent ground will not be approved. 
 
 SECONDARY SYSTEMS. 
 
 In these systems where alternating currents of high elec- 
 tro-motive force are used on the main lines, and secondary 
 currents of low electro-motive force are developed in local 
 "converters" or "transformers," it is important that the 
 entire primary circuit and the transformers should be ex- 
 cluded from any insured building, and be confined to the 
 aerial line (the transformers being attached to the poles or 
 the exterior of the buildings) or to underground conduits if 
 such are used, or placed in fire-proof vaults or exterior 
 buildings. 
 
 In those cases, however, where it may not be possible to 
 exclude the transformers and entire primary from the build- 
 ing, the following precautions must be strictly observed : 
 
 The transformer must be constructed with or inclosed in 
 a fireproof or incombustible case, and located at a point as 
 near as possible to that at which the primary wires enter 
 the building. Between these points the conductors must be 
 heavily insulated with a coating of approved waterproof 
 material and, in addition, must be so covered-in and pro- 
 tected that mechanical injury to them, or contact with them, 
 shall be practically impossible. 
 
 These primary conductors, if within a building, must also 
 be furnished with a double-pole switch, or separate switches
 
 142 TRANSFORMERS. 
 
 on the ingoing and return wires and also with automatic 
 double-pole cut-out where they enter the building or where 
 they leave the main line, on the pole or in the conduit. The 
 switches above referred to should, if possible, be inclosed in 
 secure and fireproof boxes outside the building. 
 
 In the case of isolated plants using the secondary system, 
 the transformers must be placed as near to the dynamos as 
 possible, and all primary wires must be protected in the 
 same manner as is indicated in the second paragraph above.
 
 143 
 
 GLOSSARY. 
 
 ELECTRICAL TERMS NOT EXPLAINED IN 
 THE PRECEDING CHAPTERS. 
 
 Ampere. The unit of electric current. A current 
 of just sufficient strength to deposit .005084 grains of 
 copper per second. 
 
 Electro -Motive Force. Generally expressed as 
 E. M. F. (See Voltage.) 
 
 Kapp Lines. A unit of lines of magnetic force. 
 
 Multiple. Synonymous with Parallel. (See Chap- 
 ters I arid VI.) 
 
 Ohm. The unit of electrical resistance. A resist- 
 ance through which one volt can just force one 
 ampere. 
 
 Parallel. See Multiple. 
 Potential. See Voltage. 
 
 Series. Lamps or other pieces of apparatus are in 
 series when the same and whole current passes 
 through them all, one after another. (See Chapters I 
 and VI.)
 
 144 GLOSSARY. 
 
 Voltage. Synonyms : Potential, Electro-Motive 
 Force, Pressure, etc. The number of volts pressure 
 present at any given point in a circuit. 
 
 "Watt. The unit of electric power, or the volt- 
 ampere. The volts multiplied by the amperes in use 
 equal the watts.
 
 INDEX. 145 
 
 INDEX. 
 
 A 
 
 Alternating current 17, 19 ( 22 
 
 Analysis of iron .' 82 
 
 Asbestos paper, use of 89 
 
 Attraction, molecular * 52 
 
 B 
 
 Baking 90, 108, 109 
 
 Banking in multiple 98 
 
 in series 100 
 
 three wire 101 
 
 Binding bolts 85 
 
 Blocks, terminal 91 
 
 Box, transformer 91 
 
 Burr on plates 83 
 
 c 
 
 Calorimeter 51, 53 
 
 Carbon 82 
 
 Capacity, transformer required 97 
 
 Case, transformer 91 
 
 Coil, reactive 39, 45 
 
 induction 26 
 
 Faraday induction 64, 80 
 
 Ruhmkorff 64, 81 
 
 Coils, ends of 89 
 
 Cobalt 40 
 
 Connections, secondary ... 108 
 
 Commutator, function of 
 
 Constant current transformer 139 
 
 Construction, rules for 94 
 
 Core, form of 73 
 
 joining of 76
 
 146 INDEX. 
 
 Core, open circuit 65, 81 
 
 building of 84 
 
 grounding of 105, 123 
 
 wire 80 
 
 Copper tape, use of 87 
 
 Cost, influence of 67 
 
 Crossing of leads 91, 94 
 
 Currents, magnetic 39 
 
 D 
 
 Danger from grounds 105 
 
 Deptford Station 128 
 
 Devices, safety 92 
 
 Dies for punching 82 
 
 Dick and Kennedy's early transformer 67 
 
 Direct current 18 
 
 transformers 136 
 
 Direction of current 15 
 
 Discharge, 10,000 volt 128 
 
 Distribution, electrical 15 
 
 Dynamo, principle of 12 
 
 E 
 
 Eddy currents 12, 48 
 
 Efficiency of transformers . 62 
 
 curve 61 
 
 Electron 14 
 
 F 
 
 Faraday coil 64, 80 
 
 Faraday's discoveries 63 
 
 Ferranti converter 72, 110 
 
 Ferranti's early transformer 66 
 
 Fiber, vulcanized 89 
 
 Field of force 12 
 
 Filings, iron 40 
 
 Force, lines of 14, 40 
 
 Formulae, Hopkinson's 55 
 
 Fort Wayne Electric Co 118 
 
 Foucault currents 12, 48 
 
 Friction, molecular 53 
 
 Fuse, main circuit 100 
 
 Fuses 38 
 
 location of 106, 108 
 
 replacing of 92, 108
 
 INDEX. 147 
 
 Fuses, rubber covered 106, 108 
 
 sub-circuit 100 
 
 G 
 
 General Electric Co 121 
 
 Gloves, rubber 108 
 
 Goulard and Gibbs, distribution 66 
 
 Grounding of core 105, 123 
 
 H 
 
 Hedgehog transformer 72, 81, 132 
 
 High voltages 128 
 
 measurement of 129 
 
 Humming of transformer 95, 107 
 
 Hysteresis 51 
 
 I 
 
 Improvement, direction of 72 
 
 Induction, electrical 10, 14, 17 
 
 coil 26, 80 
 
 self 32, 39, 46 
 
 mutual 34, 46 
 
 Inertia, electrical 10, 39 
 
 magnetic 52 
 
 Intermittent current 17, 19 
 
 Interrupter 25 
 
 Insulation 90, 94 
 
 Insulation in lamination 84 
 
 oil 92, 105, 123 
 
 puncturing of 92, 105 
 
 of winding 86, 89 
 
 wood for 90 
 
 Iron, complete circuit of 41 
 
 analysis of 82 
 
 English 82 
 
 lamination of 50 
 
 filings 40, 95 
 
 permeability of 40 
 
 selection of 82 
 
 wrought, characteristics of 43 
 
 J 
 
 Jablochkoff, distribution 66 
 
 Jar, advantage of 103, 108 
 
 Joints in windings 88
 
 148 INDEX. 
 
 K 
 
 Kapp lines 41 
 
 Kennedy's transformer 66 
 
 L 
 
 Lag 11 
 
 Lamp, incandescent 17 
 
 Lathe, winding 86 
 
 Lamination of iron 50 
 
 Law, Ohms 46 
 
 Leads, crossing of 91, 94 
 
 Lightning, effect of 104, 106 
 
 discharge, potential of 106 
 
 Lines of force 14, 40, 41 
 
 Kapp 41 
 
 Location of transformers 93, 103 
 
 Loss, C 2 R 16 
 
 in converters 45 
 
 percentage of 54 
 
 M 
 
 Magnetic circuit, resistance of 66 
 
 currents 39 
 
 Magneto 93 
 
 -motive force 57 
 
 Mandril, winding 86 
 
 Mica, use of 89 
 
 Molecular attraction 52 
 
 friction 53 
 
 Multiple, banking in 98 
 
 arrangement 27 
 
 Mutual induction 34, 46 
 
 N 
 
 National Electrical Mfg. Co 115 
 
 transformer 115 
 
 Nickel 40 
 
 o 
 
 Ohm's law 46 
 
 Oil insulation 92, 105, 123 
 
 P 
 
 Paccinotti ring 87 
 
 Paper, asbestos 89
 
 INDEX. 149 
 
 Paper, shellaced gg 
 
 Parallel arrangement .....'., 27 
 
 Phosphorus 
 
 Pine '.'.'.'.'.'.'...'.'. 90 
 
 Plates, punching of ... 82 
 
 Polarity, change of 52 
 
 Potential, best to use 98 
 
 difference of 10 
 
 Primary circuits, handling of 108 
 
 coil, character of 15, 17 
 
 Pulleys, tension 86 
 
 Pulsating currents 17, 19, 23 
 
 Punching of plates 82 
 
 Puncturing of insulation 105 
 
 R 
 
 Ratio of diameter to circumference 56 
 
 of windings 59 
 
 Reactive coil 39, 45 
 
 Regulation, self 34 
 
 Resistance of circuits 16 
 
 Ring-shaped core 78 
 
 Ribbon, silk, use of 87 
 
 Rubber tape, use of 87 
 
 Ruhmkorff coil 64, 81 
 
 Rules for construction 94 
 
 for installation 108 
 
 Underwriters' 108, 140 
 
 s 
 
 Saturation point 42 
 
 Safety devices 92 
 
 Secondary coil, character of 15, 17 
 
 Sectional area of wire 94 
 
 Self-induction 32, 89, 46 
 
 Series arrangement 27 
 
 Silicon 82 
 
 Slattery transformer 118 
 
 Stanley Electrical Manufacturing Co 118 
 
 transformer 118 
 
 Stanley's early transformer ' 67 
 
 Steel, sheet 43 
 
 use of for cores 82 
 
 Step-up transformer 26, 107, 128 
 
 St. Elmo's fire 131 
 
 Station transformer 137
 
 150 INDEX. 
 
 Swedish iron 82 
 
 Swinburne's transformer 72, 132 
 
 Symbols assumed 55 
 
 T 
 
 Tape, copper 87 
 
 rubber 87 
 
 Taping of coils 88 
 
 Teak, use of 90 
 
 Tension pulleys 86 
 
 Terminal blocks 91 
 
 Tesla's experiments 128 
 
 Testing of transformers 93 
 
 Thomson-Houston transformer 121 
 
 Three-wire arrangement 101 
 
 u 
 
 Underwriters' rules 108, 140 
 
 V 
 
 Vibration and hysteresis 52, 103 
 
 w 
 
 Water, analogous to electricity 16 
 
 Weight of transformer per useful watt 60 
 
 Welding machine 134 
 
 Westinghouse converter 125 
 
 Winding, methods of 76, 85 
 
 Windings, proportioning of 60 
 
 ratio of 59 
 
 Wire, coiling of 32 
 
 Wood for insulation 90 
 
 Working point 42 
 
 z 
 
 Zipernowski and Derri's early transformer 70
 
 BWBIER'S 
 
 POPULAR ELECTRICIflN. 
 
 A Scientific Illustrated Monthly, 
 For the Amateur and. Public at Large. 
 
 Containing descriptions of all the new inventions as fast as they are patented, 
 also list of patents filed each month at the Patent Office in Washington, D. C. 
 Interesting articles by popular writers on scientific subjects written in a way 
 that the merest beginner in science can understand. 
 
 Pidee, Postpaid, SO Cents a Yeat*. 
 SfllWPIiE COPY, FIVE CENTS. 
 
 23p=" Send for it. You will be More than Pleased. ^j% 
 
 Bubier Publishing Company, Lynn, JVIass.
 
 THE LEADING INSTRUCTION BOOKS UPON 
 
 ELECTRICITY. 
 
 "Everybody's Hand-Book of Electricity." by Edward Trevert, with 
 Glossary of Electrical Terms and Table for Incandescent Wiring. 120 Pages, 
 50 Illustrations. Giving a description of all the latest electrical inventions up 
 to the present time. Over 25,000 sold already. 
 
 Price, Paper, 25 cents; Cloth, $1.00. 
 
 " How to Make Electric Batteries at Home" by Edward Trevert, fully 
 illustrated. The book gives just the information needed to make simple, yet 
 practical, electrical batteries, by which you can run electric motors, incandes- 
 cent lamps, bells, etc. Price, 25 cents. 
 
 " Experimental Electricity," by Edward Trevert. 176 Pages, 100 Illustra- 
 tions. Giving complete working directions for making electric batteries, electric 
 bells, induction coils, galvanometers, electric motors, dynamos, magnetos, etc. 
 
 Price, $1.00, Cloth Bound. 
 
 "Dynamos and Electric Motors," and All A tout Them, by Edward Trevert. 
 Fully Illustrated. This book gives complete directions for making Dynamos 
 and Motors ; also tells you all about them. Price, 50 cents, Cloth Bound. 
 
 "Armature and Field-Magnet Winding," by Edward Trevert. Fully 
 Illustrated with working drawings and fine engravings. All the information 
 contained in this book is of a practical character, and the formulas used is of the 
 highest authority, by which one can wind a dynamo or motor for any output. 
 
 Price, $1.50, Cloth Bound. 
 
 " How to Make a Dynamo," by Edward Trevert. Illustrated. Giving 
 complete working directions and drawings for making a two incandescent light 
 or small arc light machine. Price, 10 cents, Paper. 
 
 " Electric Railway Engineering," by Edward Trevert. Illustrated. Em- 
 bracing power house, dynamo, motor, and line construction up to date. 
 
 Price, $2.00, Cloth Bound. 
 
 " Electricity and Its Recent Applications," by Edward Trevert. 350 
 pages. 250 Illustrations. 12010. Cloth. The best and most complete book of 
 the present time, and is particularly adapted for the use of students. Contains 
 20 chapters of the best and latest experiments, also complete working directions 
 for building all kinds of electrical machinery. Price, $2.00, Cloth Bound.
 
 " A Practical Treatise on the Incandesctnt Lam}? by J. E. Randall, 
 Electrician of the Incandescent Lamp Department of the Thompson-Houston 
 Co. Illustrated. This is the only work that explains in a practical manner the 
 manufacture of the Incandescent Lamp, and should be owned by every Electric- 
 ian and student interested in the subject. Price, 50 cents, Cloth Bound. 
 
 " A Practical Treatise on Electro-Plating," by Edward Trevert. Just the 
 book for Amateurs. Fully Illustrated. Sent to any address on receipt of price. 
 
 Price, 50 cents, Cloth Bound. 
 
 " Electric Motor Construction for A ntateurs" by C. D. Parkhurst. Illus- 
 trated. Just the book for beginners or for anybody wishing to construct their 
 own electrical apparatus. Giving complete directions and working drawings for 
 making an electric motor for running sewing machines, small lathes, etc. Also 
 gives directions and drawings for building an electric battery to furnish current 
 for the motor. Price, $1.00, Cloth. 
 
 "A Hand Book of Wiring Tables," for Arc, Incandescent Lighting and 
 
 Motor Circuits, by A. E. Watson. This book gives a large number of Formula- , 
 
 Rules and Tables for Wiring, and is illustrated with numerous diagrams. It also 
 
 contains a large amount of practical information upon the subject, up to date. 
 
 Price, 75 cents, Cloth. 
 
 "How to Make and Use Induction Coils," by Edward Trevert. The 
 only American book upon Induction Coils. It is fully illustrated, and every 
 student of electricity should own one. Bound in a neat cloth binding. 
 
 Price, 50 cents, postpaid. 
 
 " Questions and Answers About Electricity." A first book for students. 
 Theory of Electricity and Magnetism. Edited by E. T. Bubier, 2 d. Authors: 
 T. O'Connor Sloane, A. M., E. M., PH. D. ; Caryl D. Haskins, M. I. E. E. ; 
 A. E. Watson; Edward Trevert. Illustrated. Contents: Chapter I. 
 Theory of Electricity. II. -Theory of Magnetism. III. -Voltaic Batteries. 
 IV. -Dynamos and Motors. V. - Electric 'Lamps. VI. -Miscellaneous 
 Electrical Apparatus. VII. Electrical Measurement. 
 
 Price, Cloth Bound, 50 cents. 
 
 All Books sent postpaid on receipt of price. 
 
 Send Money by P. O. order or registered letter at our risk. 
 
 Bubiet* Publishing Co., 
 
 LYNN, MASS.
 
 University of California 
 
 REGION; 
 
 SOUTHERN REGIONAL LIBRARY FACILITY 
 
 405 Hilgard Avenue, Los Angeles, CA 90024-1388 
 
 Return this material to the library 
 
 from which it was borrowed. 
 
 Libr 
 
 NOV 2 2 1993 
 
 RECtivtu 
 
 iry
 
 UC SOUTHERN REGION* LBRARY 
 
 A 001243241 
 
 TK 
 2798 
 
 l!27t