Seeing this, it must be thought that this gentleman either does or will not, understand the subject. Then if Gaulard has succeeded with his apparatus in obtaining some advantages as proposed in the above-mentioned clauses, Nos. 1 and 2, these advantages can be obtained to a much higher degree with non-polar transformers. This has been proven by Prof. Ferraris.[6]

The improvements mentioned under Nos. 3 and 4 are only to be attained with bi-polar transformers after difficult and otherwise disadvantageous arrangements; for instance, the combination of the primary and secondary wires in a common cable, or, when the coils consist of ribbon wire, by the winding of the one inside the other. With non-polar transformers these improvements are already inherent. The Fuller transformer was just as much without poles as two horseshoe magnets are, with their like poles laid together.

In all these systems with series connection of the transformers, the intensity of the current in the primary circuit must be held constant in order that it may be possible for the induction apparatus to maintain the secondary electromotive force constant. Notwithstanding this, constancy was not attained, but only one cause of the variations annulled. Another cause of the variations of the difference of potential at the secondary terminals of the coil still remained; this was the loss of potential due to resistance and self-induction, which increased with the load. The electromotive force of the secondary, and therefore of the primary coils, accordingly increases as the current in the secondary decreases. When no secondary current is flowing, the electromotive force in the primary and secondary coils is a maximum. We have consequently this disproportion that the smaller the output of the apparatus the greater the energy consumed. With the secondary circuit open and a constant exciting current, the energy used could be as much as ten times as great as under full load.

The disadvantages of this system are apparent; for, putting aside the loss of energy arising from the disproportion between produced and consumed energy, each change of load on the secondary circuit exerted a great influence on the primary circuit, and again on the secondary circuits of the other coils in the main circuit.

All the transformer systems already described were intended, as we see, for subdividing the current, and as fitting therefor we find the series method of connection universally brought forward. With this method, owing to a rise in electromotive force which was dangerous to the lamps, &c., when only a part of those in the secondary circuit were extinguished, it was compulsory either to run the induction coil fully loaded or quite empty. Thus, when the number of lamps or other devices in use was varied, a regulation of the current strength and uniform working was either quite impossible, or only partly possible by unreliable and incomplete mechanical means. On this account no one succeeded with this method in carrying out a rational distribution of current by means of induction coils such as are required by the widespread demands for electric current from a central station.

The first to point out the disadvantages of the series method of connection was Rankine Kennedy, who had devoted himself wholly to the study of induction apparatus. These disadvantages he published in an article in the “Electrical Review” of 9th June, 1883. At the end of this article we find the interesting statement that transformers, when not connected in the primary circuit in series, as had been usual till then, but in parallel, form a self-regulating system of current distribution. Rankine Kennedy expresses this in the following words:—“In parallel arc, however, the secondary generator is a beautiful self-governing system of distribution.” At the same time, however, his article affords proof that the author then possessed only a limited comprehension of the physical facts concerned, because he maintained, for instance, that the introduction of an induced counter electromotive-force in the circuit of an alternating current dynamo might constitute a means of regulation without loss of energy; however, it might be allowed, that he meant by these words one of these elements which must be present in a really rational system of distribution with the use of transformers, if it were not the case that at that time he was not aware both of the properties of transformers suiting them for such a connection as well as those which make them self-regulating in a system of distribution. Above all this he had at that time never thought of a transformer in the sense, the word is used to-day, that is, as an induction apparatus, which converts high into low tension currents. This is quite clear, as is seen from the end of the sentence before cited, as he says, “But what about the size of conductors for such a system? Prodigious!” Kennedy thought to all appearance that the parallel connection of transformers made possible self-regulation in the same manner as the simple direct parallel connection of incandescent lamps. While at the same time he imagined that on account of the small resistance of each coil the resistance of the net of leads must nearly vanish, therefore he concluded that the parallel connection of such induction apparatus as he had in his mind’s eye was impracticable.

The apprehension of Kennedy’s ideas, as we have here stated, finds direct confirmation from the leading article in the “Electrical Review” of 9th June, 1883. At the end of this leader the editors say, that “Mr. Kennedy’s apparatus is an induction coil pure and simple.” “Messrs. Gaulard and Gibbs will scarcely deny, nor can they deny, that the action of this particular construction of the coil is identical with that of his.” In this sentence it is distinctly stated that the construction of Kennedy’s induction apparatus is identical with that of Gaulard and Gibbs’. Kennedy accepted this statement in silence; if it had been otherwise, he would have protested in his next appearance in print.

In order to make possible the connection of transformers in parallel, the advantages of which it may be said Kennedy had augured, there was still much wanting. Above all there was wanting the idea of a transformer as meant at present, and an exact knowledge of its action. F. Geraldy has expressed himself very suitably upon this point in the introduction to his report upon the trials made with the system of Messrs. Gaulard and Gibbs.[7]

“La distribution de l’électricité comporte la solution d’un grand nombre de problèmes. Il ne suffit pas de se décider en principe et lorsqu’on a choisi la distribution en quantité (en supposant même, que l’un des procédés puisse être appliqué d’une façon exclusive, ce qui n’est pas certain), lorsqu’on a trouvé le moyen de régler le générateur et les recepteurs conformément au mode choisi, il reste encore à lever quantité de difficultés, a créer et disposer beaucoup d’organes auxiliaires.” Geraldy explained distinctly that it was not sufficient to determine only the method of connection, but there were still a considerable number of obstacles to be surmounted before the object could be attained.

It has been a costly lesson, before the properties of transformers were known, which make them form a self-regulating system. Even in the year 1884 do we still find Messrs. Gaulard and Gibbs on the same false track as previously. It was in the Turin Exhibition where Messrs. Gaulard and Gibbs carried out their system upon a large scale, and where they also succeeded in gaining the interest of technical circles, and arousing general attention.

The transformers installed by Messrs. Gaulard and Gibbs in the Turin Exhibition were protected by the German patent, No. 28947, and this time again their transformers were wound with equal primary and secondary coils. The construction of the apparatus, as already explained, made it a necessary condition that the transformers be connected in series, because only by this means could the high tension current be utilised. It was a necessary corollary of this method of connection that the converting of the high potential of the primary circuit into low potential, was performed, not by the ratio of the number of turns in the coils of the transformers, but in a certain manner by the subdivision of the electromotive force in the circuit.

The special construction of the transformers used in the Turin Exhibition differed from the older apparatus in so far that both coils were formed of stamped out circular copper discs, which were soldered together by projecting teeth. The insulation was made of stamped-out paper discs. Both spirals were wound between one another. The building up of such coils was effected in the following manner (see Fig. 22a):—A red copper disc was first placed on the core, then insulation, upon this a black copper disc, then again a red copper disc, and so on. Like colours of copper discs were then soldered together at the projecting teeth. In this manner there were produced two spirals running parallel with one another, there only being one layer of coils. The employment of such ribbon conductors had some advantages, namely, good use of the space at disposal for coils, and rapid cooling through the projecting teeth. They had, also, disadvantages, the chief of which was, that the conductors were of bare metal, so that a fault in insulation could easily occur.

In fact, several faults in the transformers in Turin did arise from this cause, the action of the coils being disturbed. Further attempts with similar coils were made, the station houses of Turin, Venaria, and Lanzo being lit for five consecutive hours. The circuit was about 80 kilometres long, the main lead being of chrombronze wire of 3·7 mm. diameter. At Turin there were 34 Edison lamps of 16 c.p. each, and a sun arc lamp; at Lanzo there were nine Bernstein lamps, 16 Swan lamps, a sun arc lamp, and two Siemens’ arc lamps. In the exhibition itself, there were nine Bernstein lamps, nine Swan lamps, and a sun arc lamp. In the Figaro Kiosk nine Swan lamps were fed from a small transformer.

As already related, the trials of Messrs. Gaulard and Gibbs’ system at Turin had aroused in the widest circles the liveliest interest, and, consequently, the errors of the system soon became public. Thus we find in the technical literature of that time influential voices raised against the system, and pointing out its disadvantages.

Among others, Prof. Colombo read a paper during the course of the National Exhibition at Turin, the subject being the system of Gaulard and Gibbs. While doing sufficient justice to the good points of the system, he also said that although it solved the problem of carrying the electric current to great distances, it was in no way what it was represented to be, and what it should be: a system of distribution allowing the electric current from a distant central station to be led to meet the demands of any kind of consumer without any one of these interfering with the supply of current to any other. He characterised these drawbacks sharply, and very suitably, by the remark, that in the Gaulard and Gibbs system, each consumer drew properly his supply of current from his transformer, and not from a common network of leads always self-regulating, as is the case in every large installation with continuous currents. Prof. Colombo satisfied himself with this reference to its disadvantages, mentioning also what should be striven after to make the system a perfect one, saying that the ideal electric lead system was one combining the advantages of the Edison central-station with that of Gaulard and Gibbs.

Prof. Colombo confined himself to these hints, and he must acknowledge that the means leading to the attainment of this purpose remained still to be found out.

The reproduction of this lecture by Prof. Colombo is placed before an article by Depréz in “La Lumière électrique,”[8] in which latter the system of Gaulard and Gibbs is strongly criticised. Depréz showed that that system can have no claim to be new. He points also to the wants of the system, especially that of self-regulation, stating that the means remain still to be discovered, which would make possible the self-regulation of a system of distribution with transformers. He also says that Gaulard’s system of distribution had not solved this problem, and therefore could not be held to be practically useful.

We find the same view represented in an article by H. Roux,[9] where he points to the enormous fluctuations which take place when the resistance in the secondary circuit is altered. Some of the figures vouching for his opinion we shall now reproduce. They were taken by M. Pietro Uzel, in Turin, in an observational way.[10]

The observations are only quoted so far that the Watts Δ I at the secondary terminals are still increasing; were they continued further the damning fact would reveal itself that as the power put in increased, the power given out would approach zero.

Taking account of these fluctuations, it is not possible to see how, as Mr. Roux says with justice, a distribution of current by this system can be made in an efficient manner. Mr. Gaulard in his reply, virtually assents to this article, but adds, that these variations could be prevented, if the cores of the transformers be shifted either by hand or automatically. Both methods would be expensive, and, besides, the automatic regulation would be unreliable.

It was at once recognised by all those interested in the subject, that this system made possible a subdivision, but by no means a distribution of current.

Before proceeding further with the history of the development of the transformer, let us for a little while take up the question, what conditions are necessary for a practical and rational system of current distribution by means of transformers. As we have already explained in another part of this paper, the method of parallel connection, i.e., a system in which the difference of potential is held constant, is alone suitable. Depréz maintained in his time that the difference of potential between the terminals of the source of current must be kept constant. Should the distribution be made on this principle, the resistance of the network of leads must be very small, in order that with full load only a very small loss of electromotive force may take place in the leads. In the indirect system of current distribution, consequently, the tension at the secondary terminals of the transformers must also be maintained constant.

The question is now before us, “In what manner must the primary electromotive force vary to effect this?” Consider an iron core, having on two different parts round it, two rings of wire. This iron core may now be magnetised by bringing near to it in the line of its axis a permanent magnet. On drawing the latter quickly away, an electromotive force will be momentarily produced in both the wire rings, and the electromotive force will be proportional to the number of the disappearing lines of force. This number, in consequence of the dispersion of the lines of force, will be very different at different parts of the magnetised core. The induced electromotive forces in the windings of the wire will also be different. The equality of these electromotive forces, which is so important, can only be attained if all the windings are in relatively the same position with regard to the magnetic field. The circuits of both coils being closed, the one having a current flowing through it, the other through a suitable resistance, besides the condition mentioned in the last sentence, another must be fulfilled; this is, the internal resistance must be practically zero, i.e. the difference of potential between the terminals shall equal to all intents and purposes the total electromotive force.

We have now to examine how far the already observed constructions of transformers fulfilled these demands. A transformer in which the windings lie relatively in the same position to the magnetic field can quite well be bi-polar. All that is necessary for this is that the coils be wound on to the core next to one another; this is most simply managed in a transformer having a ratio of 1:1. This law was first determined by Maxwell. The apparatus of Strumbo shows such a method of winding already carried out.

Thus it may be seen that of bi-polar transformers, those which, with regard to the constancy of the secondary tension, are most suitable, are quite useless on account of their ratio being 1:1, although they are destined for the series method of connection.

The connection of proper transformers in parallel can only be made with such apparatus as, notwithstanding their ratio of transformation, possess windings having the same relative position to the magnetic field—this is only the case with non-polar transformers. Besides this quality of non-polar transformers, their magnetic resistance is so low that the condition of very low internal resistance is easily fulfilled.

The following conditions of a self-regulating and economical system of current distribution with transformers result, therefore, from the foregoing explanations:—

1. The generator of current must give a great difference of potential as constant as possible at the terminals of the transformers, and also independent of the number fed.

2. The transformers must convert the current of high electromotive force into a current of such electromotive force as may be desired. The transformers must have a closed magnetic circuit (that is, they must be poleless), in order that all the primary and secondary turns shall possess, relatively to the magnetic field, a like position, also in order that the resistances of the primary and secondary coils shall be so small that they cause practically no loss of electromotive force.

Through the fulfilment of both these conditions, it is rendered possible to maintain the secondary tension constant by maintaining the primary tension constant, indifferently whether it is regulated automatically or by hand. To suit this, the transformers must also be arranged into distributive stations of the second order, and derived in parallel from the main leads.

In May, 1885, a system of current distribution meeting all the just-mentioned requirements was publicly brought out, giving an illustration of a truly self-regulating system of current distribution. This was the system of Zipernowsky, Déri, and Bláthy.

The first two patents concerning this system date from 18th February, 1885, and are entitled, “Improvements in the means for the regulation of alternating electric currents,” No. 34,649, by Carl Zipernowsky and Max Déri, of Budapest; “Improvements in the distribution of alternating currents,” No. 33,951, by Max Déri, of Vienna. The third patent is dated 6th March, 1885, and is entitled, “Improvements in induction apparatus for the purpose of transforming electric currents,” No. 40,414, by Carl Zipernowsky, Max Déri, and Otto Titus Bláthy, of Budapest.

The system described in these three patents was immediately afterwards brought forward in the three exhibitions of Budapest, Antwerp, and London (Inventions Exhibition), arousing in technical circles a general and well-earned attention.

In the patent documents as well as in the earliest[11] articles in the journals concerning the system, two special forms of transformers are described, viz. that consisting of an iron core with the wire outside, and, secondly, that consisting of copper coils surrounded by iron wire. The transformers shown in Figs. 24 to 28 belong to the last of these classes, that in Fig. 23 to the first. The fundamental principle upon which all these transformers are constructed is that the subdivisions of the iron core run perpendicularly to the copper wires. Transformers such as are shown in Fig. 25 having a ring-shaped iron core wound with copper wire at first employed, later the inventors used in preference the form represented in Fig. 23.

In all these forms the principle is generally adhered to, that the magnetic resistance and the exciting power possess for each part of the length of the magnetic circuit the same value, and thus the formation of poles with the resulting dispersion of the lines of force is avoided.

This system procured for itself universal recognition, but especially in the Budapest Exhibition. There several exhibits within a radius of 1,300 metres were lit from a common central station. The several circuits were quite independent of one another, and lamps could be extinguished or lit in any one of them without anywhere producing a change in the intensity of the light, which could be perceived.

It was, therefore, in the year 1885 that the problem of current distribution by means of transformers was solved in a truly practical manner. The ideas which led the inventors to this thoroughly successful solution were then so unknown to practical and theoretical electricians, that it was long ere they were understood and appreciated. Even in February, 1886, such an electrician as Prof. Forbes maintained in his Cantor Lectures that the parallel connection of transformers was quite impracticable. He believed, namely, that a connection such as shown in Fig. 29 was useless, because the difference of potential at the generator diminished from the machine outwards, but that a connection such as shown in Fig. 30 must be used. According to him, in a direct system of distribution each lamp should have a separate lead, and having regard to the great number of leads which would thus be necessary, he concluded that the series method of connection was the right one. One would suppose that Prof. Forbes was not aware of the weighty disadvantages of this method. However, that was not the case. He proposed, that with series connection the strength of current should be kept constant, and that each transformer should have an especial regulating apparatus—the raising or lowering of the core; which, by the way, is an arrangement impracticable in a well designed transformer. Such a regulating apparatus has lately been made automatic.

“This is,” says Prof. Forbes, “the last triumph, which after a series of troublesome experiments has brought us year after year nearer to the solution of the difficulties.” “I am not in a position to explain here the modus operandi,” he says further, “but I have seen the apparatus working very satisfactorily.”

This apparatus has up till now not become known. The assertion that the troublesome experiments had brought us year after year nearer to the solution of the difficulties, is quite inappropriate. Just the opposite is the case; they have taken us year after year further away from the solution, until at last all was thrown overboard and a new commencement made.

Profs. Rühlmann[12] and Esson[13] also gave vent to their opinions against the connection of transformers in parallel. In a like manner Messrs. Gaulard and Gibbs for some time after the Zipernowsky-Déri system was known pleaded for their own method of connection, until at last they were obliged, on account of the unpleasant experiences at the Grosvenor Gallery in London, to adopt the system of parallel connection, which they then at once employed at Tours.

There were, up till very lately, still many electricians who did not perceive the advantages of parallel connection, just for the simple reason that they were ignorant of the properties of the non-polar transformer, suiting the parallel system of connection for a rational system of distribution. Especially the one property of transformers remained unknown to the literature devoted to the subject up to the year 1885, namely, that in transformers properly constructed the relation between the primary electromotive force and that of the secondary, remains unaltered notwithstanding any variations in the current taken out; also that if the primary electromotive force be kept constant the secondary would likewise remain constant, provided the transformer be connected in parallel.

It had taken 30 years, until at last the way was found leading to the desired result. We have already superabundantly explained that this direction was essentially different from that taken by all electricians until after Gaulard’s time; that not only the methods of connection, disposition, and regulation of the system, but also the construction of the transformers themselves had to be quite departed from, and apparatus constructed which obeyed totally other laws to those of the earlier forms.

If indeed earlier inventors proposed for other purposes magnetically-closed induction coils, the fame due to the birth of proper non-polar transformers, in which the whole of the primary and secondary turns have a like position relatively to the magnetic-field, first invented, carried out, and combined into a self-regulating system of current distribution, belongs undoubtedly to Messrs. Zipernowsky, Déri, and Bláthy.

It would have been thought that after the direct distribution of current to glow-lamps had taken up a determined position, it would not have been difficult to discover a self-regulating system of distribution with transformers. However, the fact shows this was not the case, for after the Edison lighting system was long known, we find such electricians as Haitzema Enuma, Gaulard, and Kennedy, experimenting with the series system of connection; indeed the last of these even deters his colleagues from the attempt to run transformers in parallel, because he openly held the opinion that this method of connection was impracticable.

We have here the development of current distribution by means of transformers, as it completed itself in Europe. The American electricians however, made the matter somewhat easier. They quietly waited until the invention gave useful results in Europe, and then simply imported it.

The field to-day belongs to the parallel method of connection, and after the installation in the alkali works at Aschersleben was destroyed by flooding, there only remains a single installation with series connection, as far as we know; this is that which was fitted up in Tivoli near Rome in the year 1886. This installation however, serves only to feed an invariable number of street-lamps, and can therefore have no claim to the designation of an installation for the distribution of electric currents by means of transformers.

LONDON: PRINTED BY WILLIAM CLOWES AND SONS, LIMITED,
STAMFORD STREET AND CHARING CROSS.