General Electric Co. v. Allis-Chalmers Co.

[1] This suit in equity is for an injunction and accounting for alleged infringement of letters patent No. 726,391, for an improvement in means for varying delta-connected voltages, issued April 28, 1903, to Armstrong and Woodbridge, assignors to the complainant. The only claim involved is No. 1, which is as follows:

“The combination óf a plurality of translating devices connected in delta, interior taps from said devices, and a plurality of conductors constituting a multiphase system of distribution connected to said taps.”

The patentees describe the. object of their invention in the following language:

“This invention relates to electrical translating, devices which are connected in dplta and has for its object the provision of means whereby fractions of the maximum' voltage may readily be obtained without changing the connections between the several translating devices. We accomplish the desired result by bringing out interior taps from a plurality of translating devices permanently connected in delta and connecting the conductors of a multiphase system to the said taps or to certain of them. By an ‘interior’ tap is meant one ’which is not at an extremity of a winding. The devices may thus be connected in such manner as to produce a resultant voltage which is a fraction of the maximum, and by providing a number of taps from each device and furnishing switching mechanism to connect the conductors to different-sets of taps successively various values of voltage may be obtained. We have chosen to illustrate our invention in connection with transformers in a tEree-phase system; but it may be applied to. other translating devices and to a system of any number of phases. When we speak of windings as ■connected in delta, we do not intend to limit ourselves to three-phase connections, but mean to include any similar connection of any number of phase.”

It thus appears that the patent describes a combination of elements by which full and fractional voltages adapted to starting and operating rotary converters are obtained “without changing the connections between the several translating devices” or transformers, which serve to raise or lower the voltage of the current. The transformer consists of an iron core, a coil of wire around the iron core, called the “primary” coil, which receives the electrical energy to be transformed, and a second coil of wire also around the iron core, called the “secondary” coil, which delivers the transformed energy. An alternating current system involving but a single pair of wires is called a “single phase” system. When more than, one current is utilized, the system is called a “polyphase” system. The ordinary polyphase system is the three-phase system, in which three transmission wires are used, and in which there exists three phases of current in the system, which do not attain their maximum values at the same time, but are displaced, or out of the phase, so that their maximum values are successively attained. The Armstrong and Woodbridge patent .involves a polyphase system connected in delta as contradistinguished from the star or Y connection. In the delta connection the coils are connected end to end, forming a closed circuit, while in the star or Y connection three ends of the coils are connected to a common point, and the other ends of the coils to the conducting wires of the circuit. The complainant charges the defendant, Allis-Chalmers Company, with *147contributory infringement of the patent in suit, by reason of its man-iifacture and sale of an apparatus to the Ft. Wayne & Wabash Valley Traction Company, designed and intended by said defendant for use in the manner specified in the patent in suit, and which was installed at one of the substations of the traction company at or near Fogansport. In modern practice in long-distant railway systems, the electrical energy is generated in the form of an alternating current at high voltage at a central station and then delivered to substations along the line, where the alternating current is converted by appropriate apparatus into direct current of suitable voltage, for supply to the wire of a trolley car system for operative use. The method just indicated was that adopted and applied by the traction company above mentioned. Much might be added to what has been said with respect to the method above indicated, but it seems unnecessary, notwithstanding counsel of both parties have furnished elaborate and instructive briefs in relation thereto, for which the court is much indebted. As already indicated, the rotary converter is a machine used for changing an alternating current into a direct current, and consists of two parts, an alternating current motor driven by the alternating current supply, and a direct current dynamo, which, in turn, is driven by the alternating current motor, which generates a direct current to be supplied to the trolley wire. The alternating current motor of the rotary converter is a synchronous machine; that is, it must operate at the same speed as the generator at the distant generating station. This is necessary in order to avoid shock and possible breakage to the mechanism when connection is made between them; hence, the synchronous motor or converter must be started from rest and speeded up to the speed of the distant generator before being connected therewith. In other words, they must be in step when the connection is made. Since the rotary converter operates both as an alternating current motor and as a direct current dynamo, it is necessary that some means be provided to start the converters at the substation, and get them up to synchronous speed with the distant generator, and it is claimed for the patent in suit that it does this in a simple, cheap, safe, and automatic way “by bringing out interior taps from a plurality of translating devices, permanently connected in delta, and connecting the conductors of a multiphase system to the said taps or to certain of them. By an ‘interior tap’ is meant one which is not at an extremity of a winding.” The complainant’s expert, speaking on this point, says:

“That the interior taps of the patent in suit are never at the extremities of windings and by means of said taps, variable voltages are obtained without altering the number of effective turns; that is, without cutting in and out turns, and thus without changing the voltage conversion ratio.”

That .is to say, it is claimed for the Armstrong and Woodbridge combination that all of the turns of all of the secondaries are in effective use all of the time, whether half or full voltage is desired, and that this constitutes a radically new system. When half voltage is desired, the patentees combine one half of one secondary with the next succeeding half of the other secondary, but continue nevertheless *148to use all the turns of all the secondaries. This idea is expressed clearly, but somewhat more in detail, in the brief of counsel for the' complainant, as follows:

“Armstrong and Woodbridge’s new combination consists in (1) permanently ■connecting tlie secondaries of tire three transformers in delta and (2) providing ‘interior’ taps from each of these permanently connected secondaries leading to (3) a suitable switching apparatus. (4) The i>ermanently connected ends of the secondaries of the three transformers are connected to the same switching apparatus, and (5) the rotary converter or motor to be started is also connected to the switching apparatus.

“An operator by manipulating the switching apparatus can first allow a fraction of the full voltage to pass into the rotary converter or motor by merely moving his switch so that the contacts of the switch for the interior laps are in position.

“But in doing so all of the turns of all the secondaries are in use and are co-operating together to produce half voltage. When the rotary converter or motor has attained speed, the only thing that the operator has to do is to throw the switch so as to connect the contacts for the full voltage.

“Here again all of the turns of all the secondaries are in use but they have been combined together in a different way.

“The Armstrong and Woodbridge patent shows for the first time a system whereby lower voltages than the full voltages can be obtained from transformers or other translating devices, without changing the conversion ratio by cutting in or out a number of turns of either the primaries or the secón-, claries of each transformer.”

During the entire operation the ends of the secondary of the transformers are “permanently connected in delta,” and the circuit is not opened for obtaining either fractional or full voltages. This seems to be admitted by the defendant’s expert, as will appear from the following question and answer:

“Q. That is, they mean by permanently connected in delta the secondary ■coils shall be connected one to the other so as to form a ring at all times, whether the voltage on the consumption circuit be maximum secondary voltage or less than maximum; is that correct? A. I understand this to be the meaning of the specification, as interpreted by the drawings of the patent.”

In the prior art fractional voltages were secured either by increasing the number of turns of the primary of the transformer or decreasing the number of turns of the secondary of the transformer. This the patent in suit does not do, and hence avoids any change of ratio between the number of turns in the primary and secondary windings. By its method the circuit is not opened, nor is the number of effective turns varied.

It is contended, however, on behalf of the defendants that the patent in suit discloses nothing new, and'that in view of the prior art Armstrong and Woodbridge did no more than any one skilled in the art would instinctively have done, and to sustain their position have cited and discussed a number of prior patents, some of which belong to the single phase, some to the two phase and some to the three phase system. It is apparent that the apparatus of the single phase and two phase. systems can have but little relevancy, since the patent in suit relates to translating devices, permanently connected in delta, which implies at least a three-phase system. The patents of those groups do not therefore show the method of Armstrong and Wood-*149bridge; and, while interesting as showing the progress of the art, they are not deemed worthy of protracted discussion. It may, however, be said of them in general that they show means of obtaining variable voltages by cutting turns in and out and thereby changing the conversion ratio, but do not, as just stated, relate to a delta connected transformer. The defendants insist, moreover, that there was no invention involved in making the connection in delta after it had been for a long time adopted and used in the Y-conuection. This was likewise the opinion of the examiner in the Patent Office when the application for the patent in suit was under consideration by him, but, after argument by counsel and after a challenge by them to show a single instance where such a change had ever been made, he yielded the point and allowed the patent. This of itself is entitled to much weight in the resolution of the question now presented. One or two prior patents <licl, however, suggest that the connection might, if desired, be made in delta instead of in Y, but they neither did it, nor did they point out a way by which it might be done.

The first patent which will be specifically considered is a Swiss patent, No. 18,142, to Brown, Boveri & Co. In this case we have a Y-conneeted transformer, providing for a variation of secondary voltages, which, as the patentees in their specification say, “is best clone by cutting in or out turns of the secondary winding.” The defendant’s expert admits that this is the case, and that by means of the different taps a number of sections of the coils in the secondary circuit are varied. This is his language:

“The arrangement of the coils iuul their connections is shown in Fig. ."5 of tiie patent. It will he seen that a number of interior laps are brought out from each coil, and that contact devices operated by tiK' switch shown in the oilier figures of the drawings make contact with different taps, and thus vary the number of sections of the coil in the secondary circuits.”

lie claims, however, that this is done by the use of ‘‘interior” taps as described in the patent. The taps as used, however, are ai the extremities of the effective turns of the windings, whereas in the patent in suit the taps are inierior taps which are not placed or used at the extremities of windings, or of effective turns of windings.

A German patent, No. 75,361, of 1894, also shows a Y-connected system in which the voltage is varied by cutting in and out turns of the secondary. Mr. Woodbridge, one of the patentees of the patent in suit, speaking of this German patent, says that it shows a Y-connected compensator for starting- purposes exactly the same as the compensator which the complainant had used prior to the application for the patent in suit. It is, moreover, like the Swiss patent above mentioned, in that both change the conversion ratio of the windings cutting in and out turns thereof, which cannot be done if the transformers are permanently connected in delta. Of the hamme, British oatent No. 20,383. it is sufficient to say that this operates in the same manner as the Swiss and German patents just referred to. The Seitncns Bros. & Co. patent. No. 6,058, of 1898, consists of a Y-connected system which obtains a change in phase without changing the voltage, but does it by cutting out turns. The distinctions between *150it and the patent in suit are summed up by one of complainant’s experts in the following language:

“On the other hand, in the Seimens apparatus, the transformer secondary windings are Y-connected, and provided with taps located at suitable points thereon, from which three-phase secondary voltages, not of variable, but of invariable value or amplitude, are obtained, the location of the taps on the windings being so chosen, and the switching mechanism for connecting different sets of taps with the distributing conductors so designed, as purposely to obtain a shifting in phase of the three-phase voltages, as a whole, without changing their amplitude; and in accomplishing this result, the number;of effective turns in the windings in respect to the distributing conductors to which the secondary voltages are applied, is altered, and also the conversion ratio.”

The Rice patent, No. 508,839, operates as a single-phase apparatus. It obtains six independent single-phase currents from a three-phase system. The defendant’s expert admits that it differs from the patent-in suit, in that the “taps” are not connected to form a three-phase system of distribution, although he adds that the art at the time knew how such connection could be made. It divides one single-phase voltage into two equal half voltages in phase with each other. In the patent in suit fractional voltages, as shown by complainant’s expert, are obtained by combining the voltages as delivered from the transformers out of phase with each other to give a resultant voltage which is not the arithmetical sum or difference of the two component voltages. This the Rice patent does not do.

The Steimetz patent, No. 679,008, also shows a Y-connected transformer, and obtains half voltage by cutting in and out turns in the secondaries, and thereby altering the conversion ratio. This is disclosed by the specification, which reads:

“In order to vary tbe electromotive force of tbe direct-current terminals of tbe converter, I may, if desired, arrange the transformers so tbat tbeir ratios of conversion may be varied. For tbis purpose each of tbe secondaries is provided with suitable switching mechanism for varying tbe number of effective turns.”

Its inventor was an expert witness in the case on behalf of the complainant, and, speaking of his patent, says:

“Tbis patent shows novel features of systems of distribution by rotary converters comprising the operation of such converters as six-phase machines and the variation of tbeir voltage by changing tbeir supply voltage by means of varying tbe number of effective turns of tbe supply transformer; tbat is, using all tbeir turns or only a part thereof.”

This is substantially admitted by the defendant’s expert Dr. Duncan, as will appear from the following question and answer:

v “Q. And in going from low to high voltage in tbe connections shown in Fig. 1 of the Steinmetz patent tbe ratio of conversion and tbe number of effective turns is changed when tbe switches are moved from one position to the other; is that not a fact? A. Yes.”

The Spence patent, No. 6,774 of 1900, is claimed by the complainant to be too late, in that it was not sealed until some months after the date of the Armstrong and Woodbridge invention, and that it did not become effective until sealed. Ireson v. Pierce (C. C.) 39 *151Fed. 796, 798; Railway Register Co. v. Broadway, etc., Railway Co. (C. C.) 26 Fed. 522, 526. But, irrespective of the question of priority, it will be found upon examination that it does not show, means for varying secondary voltages. It does not show translating devices connected in delta and furnished with interior taps for obtaining variable-delta connected voltages. In so far as it does or could vary the secondary voltages, it would be done by cutting in and out turns as in other Y-connectcd transformers. The patent in suit clearly shows invention. The prior art led away from its method. That art obtained fractional voltages by cutting in and out turns in the secondary windings thereby changing the conversion ratio. This was true of both the single-phase and two-phase systems, and the same method was naturally adopted and followed in the three-phase system. Variable voltages in the secondaries of transformers were obtained in the prior art, but they were not obtained from secondaries “permanently connected in delta” through a combination which used all of the turns of all of the secondaries all of the time as does the patent in suit. What might naturally have been expected in the development of the art is well, and I think truly, expressed by one of the complainant’s experts as follows:

"What, then, would be the natural or obvious method pursued by an engineer in connection with a three-phase system prior to Armstrong, and Wood-bridge? The answer obviously is that he would va.ry the secondary voltage by cutting in and out turns of the secondary windings and changing the conversion ratio, as was done in the prior single-phase and two-phase systems. In order to do this, obviously the engineer would not naturally permanently connect the secondary windings in delta, but would choose or select a connection of the windings, such that the method of cutting in and out turns and changing the conversion ratio could be used. ■ This connection is the star or Y-eonneetion, because, if it were used, then the length of the secondary windings could be varied by cutting off more or less front the extremities of tbe Y, as T have previously explained. The engineer of .that period was obliged to say to himself, ‘If 1 wish to cut out turns from the end of a winding, obviously I must use a winding that has an end,’ which would, of course, preclude the possibility of the idea of using a winding permanently connected in delta, occurring to him. And this is exactly what the different engineers and inventors did.”

Application for the patent in suit was made December 16, 1901, but it is claimed on behalf of the complainant that the invention was made and reduced to practice several months prior thereto. , Both the patentees have so testified, and their evidence is supported by that of several others to whom they disclosed it in August, 1900. A sketch 'is in evidence which is shown to have been made by Mr. Armstrong as early as August 25, 1900. That drawing is signed by him as of that date, and his signature witnessed by one Barry, who testifies that the invention was disclosed to him on that date, and that he then witnessed Mr. Armstrong’s signature. Two days afterward, the invention was disclosed to a Mr. Davis of the General Electric Company* who swears that he then with the stamp of that company imprinted thereon the following inscription: “Received, General Electric Company, August 27, 1900, Albert G. Davis, Patent Department”— and, further, that he, Davis, wrote on it the following in*152scription, "Docket,” with the following words added,’, “scheme for Variable secondary for delta transformers, Armstrong and Wood-bridge.” He explains that the word “docket” was written as an instruction to one of his subordinates to docket the case, so that an application for a patent might be prepared. Furthermore, the docketing clerk or chief clerk, one Hull, swears that he wrote on the sketch the following, “D. 1839,' B. B. Hull, 8 — 28, 1900.” The date of invention is further shown by important contemporary correspondence. The testimony thus given, by several witnesses is so clear and explicit, and so well supported by documentary evidence, as to establish the fact that the invention was made and disclosed as early as August 25, 1900.

[2] The defendants claim, however, to have shown a prior use of the Armstrong and Woodbridge invention at Salt Lake City some time in 1898 or 1899 at the East side station of the Union Light & Power Company. This claim is sought to be established by the testimony of three witnesses. That of one of them is, however, quite inconsequential. He was in 1898 or 1899 little more than a laborer. In 1897 he was a groundman or laborer, and after that until well along in 1899 a lineman and lamp trimmer. Moreover, his testimony upon the point in dispute has but .little bearing, and that of an indirect nature. Another of the witnesses testifies that in 1898 he was foreman of the line gang, and subsequently superintendent of construction for the above company. The person who was at the time its chief engineer was not called as a witness by the defendant, although it had sought and obtained permission of the court for that purpose. This circumstance is worthy of mention, since he of all others would seem to be the person who could have best and most intelligently and accurately testified in relation to the matter. Standard Sanitary Co. v. J. L. Mott Iron Works (C. C.) 152 Fed. 635; s. c. on appeal, 159 Fed. 135, 86 C. C. A. 325.

But considering such evidence as appears, and even assuming that to some extent it discloses a use of the apparatus of the invention, which, however, is denied by several witnesses, it falls far short of that clear and convincing testimony which the law properly demands for the establishment of a prior use. It does not show with definiteness when or how long the apparatus in question was in use. One of the witnesses says nothing upon the point, another “through that winter (1897) I believe about a year,” and the third, “I can’t remember now, the exact time, but some months.” This testimony is consistent with, and might appropriately be referred to, an abandoned experiment; for, having once installed the apparatus in connection with a light and power plant, it is altogether unlikely that, if successful, it would have been so soon removed. Moreover, such facts as do appear in that connection are denied by the testimony of several witnesses, covering the entire period of the alleged use. This testimony is so at variance with that of the defendants’ witnesses as materially to weaken, if not destroy, its probative force. Considering the entire evidence, it is insufficient to warrant the court in finding the existence of the alleged prior use. The main fact is left in *153uncertainty. A prior use lias not been established beyond a reasonable doubt. Almost any patent could be defeated if such evidence a? here submitted were deemed adequate for the purpose. The conclusion thus arrived at renders it unnecessary to consider the alleged prior publication which appeared in the American Electrician in July, 1901. Upon the whole case I think the first claim of the patent in suit is valid.

Moreover, under the evidence, it is established that the installation at the Logansport substation of the above-mentioned traction company infringes the same claim. This could hardly be seriously disputed, and indeed is not, as appears from the following question and answer taken from the testimony of the defendant’s expert:

“Q. Does the combination of apparatus shown in complainant's exhibit sketch of Logansport substation contain the following elements in combination: A plurality of translating devices connected in delta, interior taps in said devices, ¡md a plurality of conductors a multiphase system of distribution connected to said taps? A. Although the apparatus of complainant's exhibit sketch Logansport substation differs in detail from the apparatus shown In the drawing of the Armstrong and Woodbridge patent, yet 1 consider that it contains the elements in combination recited in your question.”

[3] The only question remaining for consideration therefore is whether the defendant Allis-Chalmers Company is guilty of contributory infringement in furnishing- that apparatus with the intent that it should be installed and used in the manner above indicated. At this point it should be staled that for the purpose of this case, it was admitted by counsel of the respective parties that in March, 1904, the defendant Allis-Chalmers Company acquired the ownership of a majority ot the capital stock of the defendant Bullock Electric Manufacturing Company of Ohio; that it then acquired, and now has, control of all the property rights, assets, business, good will, plant, factories, tools, machinery, and stock on hand of that company, that since that time it has continuously controlled, operated, and conducted the plant and business of that company as the electric department of its own business, that in exercising its ownership and control oí that company it authorized, directed, and controlled the manufacture and sale of the apparatus, the manufacture and sale of which is charged to be an infringement herein, and that it controls and directs the defense of this suit and is paying the expenses thereof. At the request of the complainant’s counsel, counsel for the defendant also produced a contract subsequently offered in evidence between the Bullock Electric Manufacturing Company of Ohio, and the above-mentioned traction company, dated April 2, 1904. and it was thereupon further stipulated by counsel that in pursuance of that contract and between its dale and the time of the filing of the bill of complaint herein it supplied the Et. Wayne & Wabash Valley Traction Company with the apparatus called for by said contract, which apparatus was subsequently installed at the Eogansport substation of said traction company; that the rotary converter described in said contract bore two name plates, each of which, besides other matters, had stamped thereon the name of the Bullock Electric Manufacturing Company, *154Cincinnati, Ohio. The superintendent of power of the above-mentioned traction company was called as a witness for the complainant, and testified that when it was decided to install the apparatus in question bids therefor were solicited from the General Electric Company and the Bullock Electric Manufacturing Company of Ohio, that bids were submitted by those companies, and that of the Bullock Company was subsequently accepted.

The complainant also produced as a witness the chief engineer of the Bullock Electric Manufacturing Company, and who was the chief electrical engineer of the Allis-Chalmers Company, who testifed that the Bullock Electric Manufacturing Company of Ohio furnished a rotary converter and three transformers for the traction company for installation at its substation at Logansport, which rotary converter and transformers were designed in his department and under his direction. He also produced drawings showing the interior construction and connections of the transformers, and admitted that the. rotary converter and transformers were to be used together, and further in relation to such use gave the. following testimony:

“The high-tension connections of the three transformers, as well as the low-tension connections are made in delta. The low-tension connections are divided in the center and from those center points connections lead to the three terminals of the double throw three pole switch; the other three terminals of this switch being connected to the main terminals of the transformers. The joint terminals of the double throw three pole switch are connected to the slide rings of the rotary converter. Q. What was the purpose of such arrangement? A. The purpose of the arrangement is to reduce the voltage at starting by means of the intermediate terminals already referred to.”

He also produced under subpoena duces tecum a blueprint drawing taken from the miscellaneous files of the Bullock Company, which drawing was marked, “Diagram of Connections Railway, Shunt Wound Rotary Converters, Bullock Elec. Mfg. Co. Apr. 22, 1904.” A comparison of that diagram with one showing the installation at the Logansport substation shows that the apparatus and method of connection were substantially the same. Indeed, the superintendent of the traction company testified that the blueprint represented the installation made at the Log'ansport substation. The defendant’s superintendent, however, -disclaimed all knowledge of this drawing other than that it was not made under his direction, but, when asked the following question, “As an engineer you know that this drawing could only have been made for the purpose of instructing those who were wiring this plant (Logansport), how the connections should be made, do you not?” answered, “Yes,” but some time later said that he should have qualified his answer by saying, “This or any other plant.” He was manifestly an unwilling witness, and his answers were frequently captious, if not evasive, and lacking in that frankness which begets confidence in the sincerity of a witness. Moreover, the complainant has shown by expert testimony apparently uncontradicted that the apparatus called for by the contract between the Bullock Company and the traction company could be combined only by following the method of the patent in suit.

*155Without reference to other testimony, it is concluded that the defendant Allis-Chalmers Company, through the Bullock Company _ of Ohio, manufactured and sold to the traction company the infringing apparatus in question, with the intent and purpose that it should be connected and used at Logansport in the manner that it was, and that in so doing it became, and was, guilty of contributory infringement of claim 1 of the patent in suit.

No proofs have been offered against the Bullock Manufacturing Company of New Jersey, a defendant, and, since the Bullock Manufacturing Company of Ohio was not served with process and has not appeared herein, the bill of complaint will, as to those defendants, be dismissed, while as to the defendant Allis-Chalmers Company a decree in favor of the complainant will be entered pursuant to the prayer of the bill.