General Electric Co. v. Wagner Electric Mfg. Co.

TOWNSEND, Circuit Judge. The patent in suit relates to transformers with a voltage of twenty to thirty thousand volts. Its objects, as stated by the patentee, are “to so arrange the energizing coils that they are well insulated from the surrounding core and from each other, at the same time providing suitable means for ventilation,” and “to provide a transformer having two complete and separately controlled systems of cooling, one being for the coils and the other for the laminated core.” The only issue herein is anticipation. The patented transformer comprises an outer inclosing case mounted on a base provided with a chamber open at the bottom for the admission of air or other insulating medium. Within the said — 773 — case are primary and secondary coils, vertically placed, and the laminated core, horizontally placed, and divided into sections, separated by passages through which the insulating and _ cooling medium can flow. The fundamental principle of construction consists in so inclosing the coils and their ventilating passages in an inner inclosing case, extending from a point at a considerable distance above the coils to a point equally below the coils, as to insure complete electrical and mechanical insulation of the core and its ventilating system from the coils, and adapt it to independent regulation. It further consists in insulating the primary coils from the secondary coils and providing ventilating spaces between and around the coils in such a way as to materially reduce the quantity of heat-inclosing wrappings and still preserve sufficient electrical insulation. It is sufficient for the purpose of this inquiry to say that the patentee substituted for the large quantity of insulating covering required by the prior art a small amount of such mechanical insulation, and also ventilating air spaces, and thereby overcame the objections of overheating attendant upon the use of insulating material alone. One current of air enters at the base, .passes up through the vertical passages between the coils; another current passes up at the side, and through the horizontal passages between the laminations of the core, and to the outside of the case. “It will be seen that there are two separate ventilating systems for cooling the transformer, one for the coils, the other for the iron, and that each of them is controlled independently of the other.” The claim of complainant as to the prior art, and'the nature of the patented improvement thereon, is shown by the following extract from the deposition of the patentee, Moody: “The arrangement differs from the earlier forms on which our experiments above mentioned were made, and in accordance with which our first two or three years’ commercial' production was carried out, in that there is no connection between the spaces between the coils and the spaces between the sections of the iron. We found it necessary to adopt this arrangement for the following reasons: I am referring to transformers wound for quite high potentials, say from 10,000 to 40,000 volts. To insulate such potentials with the best-known insulating materials, one must have a considerable thickness of insulation — so great a thickness as to greatly retard the dissipation of the heat from the windings if the insulation is wrapped upon the coil itself. Having found this to be so, we attempted to arrange a considerable portion of the insulation exterior to the coils. This was at first accomplished by the use of various forms of spacing strips, which held the coils some distance from each other and from the iron; also by the use of sheets of solid insulation placed between the coils. “As we attempted to design for higher and higher potentials, however, we soon found that, in an air-cooled transformer at least, this construction soon reached its limit, due to the fact that the surfaces on which we depended for this insulation could not be kept clean, and that a surface which is dusty and dirty is a very poor insulator. For instance, such a surface under a strain of 20,000 volts might break down when the distance between the parts between which such difference of potential existed was 8 or 10 inches. . “In the form in which we originally arranged our cooling ducts, the air first passed through a section of the iron core, then through one side of the coils, then to the center of the core, next through the other side of the coils, and finally escaped through the outer section of the core on the opposite side from where it entered. We had therefore four places between the coils and the iron in which it was necessary to have insulating spacing strips, and the — 774 — width of these strips had to be sufficient to stand the pressure against which they were supposed to insulate, and then the spacing strips were dirty. In other words, if we were building a 20,000-volt transformer, and expected the insulation on the coils to take one-half of the total strain, the dirty spacing strips must take the remaining 10,000-volt strain. For this a distance of three inches might, perhaps, be necessary; consequently six inches of the space available for winding within the transformer’s iron core was sacrificed for insulation. “Due to these facts, we found it commercially impossible to design transformers along such lines, and intended for air cooling, for potentials much above 10,000 volts. I began to see that it was necessary that we should have nothing but solid and continuous insulation between the winding and the iron core and between the high-pressure winding and the low-pressure winding, and consequently that all openings for the entrance and exit of the air must be outside of the transformer’s core, where the necessary projections could be made of the solid insulation to obtain safe distances against the leakage of the current, without increasing in any way the dimensions of the coils or core in order to give space for such insulation. “The arrangement as shown in the patent is the outcome of this thought. In it we have a construction where high potential coils are completely incased in a solid insulating box, separated from the coils by the necessary air passages, and projecting beyond the core at either end a sufficient distance to safely allow of the opening of the box for the entrance and egress of the air. The cooling-of the iron is taken care of separately, and solid insulations completely separate the windings as a whole from the iron core.” The claims in suit are as follows: “(4) In a transformer, the combination of primary and secondary windings with passages extending between the windings for the circulation of a cooling medium, and a laminated iron core with a second set of passages, through which circulates an insulating cooling medium, said sets of passages forming independent cooling and ventilating systems, one for the coils and one for the core. • “(5) In a transformer, the combination of primary and secondary windings, a laminated iron core, passages extending between the windings through which an insulating medium circulates to cool the windings, a second set of passages extending through the iron core, independent of the first, in which an insulating medium circulates to cool the iron, and means for regulating the circulation through both sets of passages. “(6) In a transformer, the combination of primary and secondary windings, a laminated core, means for maintaining an up-and-down circulation of air through the coils, and means for maintaining a transverse circulation of air through the iron core, both circulations being independent. “(7) In a transformer, the combination of primary and secondary windings separated by air passages, a laminated core divided into sections, and air passages independent of those between the windings formed between the sections, a common source of air supply, and means for regulating the flow of air through each set of passages independently. * * * “(11) In a high-potential transformer, the combination of primary and secondary windings divided into sections, a laminated core, an insulating case between the core and the windings, air passages between the case and the windings, insulating eases surrounding the primary windings and separating it from the secondary, the cases being provided with open ends which project above and below the ends of the coils so as to increase the creeping or leakage surface, air passages between the primary coils and the casings, and a common chamber supplying the air for both sets of passages.” The closest approximation to the patent in suit is shown in defendants’ exhibit “Model of Ferranti Transformer.” It is claimed that this model accurately represents the construction disclosed in the Ferranti British patent of 1891, and certain details shown in the Ferranti patent of 1885 and referred to in said later patent. It — 775 — comprises two vertically disposed oval core plates in sections or strips, having open spaces between said sections, and high and low potential coils, horizontally disposed, which link with the core plates, and are separated from each other and from the iron core by “cylinders of insulation,” between which and the coils are air spaces. The coils are subdivided into parallel rings or sections, with air spaces between them. The position of defendants in regard to this model and the patent in suit is shown by the following statement by their expert: “In both the Ferranti patent and in the patent in suit, the devices shown in the drawings reveal a transformer mounted upon a base, the transformer being so mounted that the air of the room rises up through the base or into the base from the side, passing through between the outside case next to the iron and the secondary winding; passing also between the secondary winding and the insulating cases within; it passes also upward through all of the channels between the insulating cases within and the primary winding, and also between the sections of the primary windings. The air passing through all of these channels is in each case and in the same sense supplied from a common source or chamber.” This statement may be accepted as correct. In view of the language used in the Ferranti specification, and the admissions of complainant’s expert, it may be assumed that Ferranti extended the ends of the insulating cylinders above and below the coils. On these grounds it is forcibly argued that the Ferranti device of 1891 fully meets claim 11 of the patent in suit. These Ferranti patents are chiefly relied on to defeat the alleged novelty of the plan of insulation covered by claim 11. The two insulators are strikingly similar in construction and arrangement. Upon mere inspection, irrespective of questions of function or result, the Ferranti patent closely corresponds with the following description in the specification of the patent in suit: “The insulation thus applied forms a sort of openwork structure having a number of channels or air passages, preferably closed or substantially closed from one another, and open top and bottom to allow a free circulation of air between the different windings and between the coils and insulation. The insulation forming the walls of these channels preferably extends beyond both ends of the coils, so as to present an extended creeping surface over which the current must pass before it can reach the secondary winding.” It is only by a consideration of the objects sought and a comparison of the results accomplished by the two devices that the materiality of the differences can be determined. Generally speaking, Ferranti’s device does not appear to be so constructed as to provide for or promote circulatory ventilation. His air spaces “not only serve to keep the coils cool, but also to secure improved insulation.” Moody’s object is “providing suitable means for ventilation * * * by means of a current of air or other insulating medium.” But even if the idea of ventilation is sufficiently disclosed (and it is mentioned in Ferranti’s specification), his construction precludes the possibility of ventilation through separate and distinct passages about the primary and secondary coils insulated from each other and independent of the transverse passages of the core, which is an essential element of the Moody invention. If a current of air were passed through the Ferranti air spaces, it would flow indis — 776 — criminately back and forth through core and coil spaces. The passages, therefore, are neither mechanically nor electrically separate. That Ferranti’s air spaces were rather in the nature of mechanical insulators than of means for positive ventilation is further indicated by Ferranti’s suggestion to put the coils in jars of insulating material, sealed over at the top to prevent leakage of current or entrance of moisture. Complainant’s expert states that the Ferranti construction would be impracticable for high power transformers, and gives substantial reasons therefor. He testifies that he attempted to use such a construction as is proposed by Ferranti, and was obliged to abandon it. That the Ferranti transformer was not used during the years when inventive genius had been stimulated by the development of the demand for high-power transformers indicates that, at best, it was only an imperfect, undeveloped theory of reduced coverings and air spaces. That it did not sufficiently disclose the patented invention is demonstrated; that it might by the genius 'of an inventor, or upon some afterthought theory of an infringer, be adapted for practical use, is insufficient. The answer to this latter contention is that these defendants did not see fit to use it or to improve upon it, but have appropriated complainant’s construction. While this claim 11 is especially directed to the insulating feature of the patent in suit, it sufficiently covers the novel means infringed by defendants, by which the air passages through the coils are secured. For the foregoing reasons, we conclude that it is not anticipated by the Ferranti patents. The date of the patent in suit is October 19, 1897. Defendants have introduced evidence that in 1896, no date being given, and in 1894, they constructed transformers which, it is claimed, embody the construction covered by claim 11. The complainant has introduced evidence to carry the date of invention back to August, 1896. While this evidence is not as definite or certain as is desirable, it is thought to be sufficient to establish priority over defendants’ 1896 transformer. The claim of anticipation by the 1894 transformers is not pressed in defendants’ brief. They were intended to be immersed in oil, and, when thus immersed, were never used at a higher pressure than 4,000 volts. When used with air, their capacity was reduced between 25 and 33 per cent. They did not embody the independent, separate systems of cooling of the patent in suit, and other details of its construction are lacking. Defendants’ 1896 transformers were similar to those of 1894, except that they were designed for a higher potential and therefore show wider air spaces between the coils, and mica insulation surrounding the high and low tension coils. These differences merely involve the degree of insulation and ventilation. It does not appear that there are any other substantial differences of construction; in fact, the testimony of defendants’ expert indicates that in other respects the transformers of 1894 and 1896 are practically the same. The defense of anticipation by defendants is not sustained. The remaining claims cover the combination of primary and secondary windings with passages between for the circulation of air or other cooling medium, with the core and its passages so con — 777 — structed and arranged as to secure independent ventilation systems for coil and core. The patentable novelty of said combination is first challenged on the ground that the patent is for a transformer cooled either by air or oil, and is anticipated by prior oil-cooled transformers. The repeated references in the specification to “ventilation” or “air cooling” or “circulation,” the construction of the patented device which negatives the idea of its use with oil, because if oil were to be used various elements would be unnecessary or modified in form or arrangement, in connection with the references to “air or other insulating medium” to be used for ventilation, are strongly suggestive, if not conclusive, that the patentee conceived his invention to consist, inter alia, in an air-cooled and ventilated construction, and merely wished to protect himself against equivalents of such medium. The Thomson patent, No. 516,850, shows a casing surrounding the coils filled with oil, and the core located outside of said casing. Between said casing and the core are air spaces, provided in order to insulate against the escape of electricity and heat between the case and the core. The heated oil is conveyed by pipes to a water-jacket above the casing, where it is cooled, and it then returns to the coil case. The air spaces are separate from and independent of the oil case, but there is no suggestion of their use for purposes of ventilation. The language of the specification as to the passage between the core and the case which surrounds the coils and the oil, relied on by defendants’ counsel as a description of a means of ventilation, falls far short of disclosing or describing the system of ventilation of the patent in suit. The objections to this patent as an anticipation are well stated by complainant’s expert as follows: “Tlie patent of Thomson, No. 516,850, does not tel] the reader skilled in the art how to make use of air for the independent and efficient cooling of the core so as to co-operate with the cooling medium which is sent through the spaces between the windings; but, more important than this, he does not tell how to make use of air for cooling the windings in any way which would be practically valuable. * * * Nor does it teach the desirability of the use of air as a cooling medium in the passages between the windings; nor does it teach the desirability of the combination of any system of coil passages coacting with any system of core passages for the purposes of cooling and ventilation, these two systems being mutually independent.” The Forbes patents and publications show and describe means for oil-cooling stationary armature coils, and are only remotely relevant to the issues herein. In the prior patent, No. 558,090, granted to Moody, the patentee herein, he described a transformer designed to be cooled by oil, although it was so constructed that air might be substituted for oil. There is no separation of the coils from each other, the insulation is applied directly to the coils, and the channel filled with oil serves both as an insulating and cooling medium. There is no sectional division of the core; there are no separate sets of independently regulated air passages. There is but one continuous set of passages, and but one single system of distribution, with a single regulation. The transformer was designed and adapted for moderate potentials, and the evidence indicates that it could not, as thus constructed, withstand any such voltages as those for which the transformer in suit is adapted. It is unneces — 778 — sary to discuss the other patents introduced by defendants. Oil-cooled transformers with circulating systems were old, and air passages, more or less remotely suggesting ventilation, were shown in the prior art. The discussion of the Ferranti patent shows how closely prior inventors had approximated to the construction of the patent in suit. We have been unable to find in the prior art any single device, or any sufficiently definite suggestions derivable from the various devices, which sustain the contention of defendants that the patented improvement is merely the result of mechanical skill. The reasons for the conclusion that the patented device involved invention sufficiently appear from a comparison of its construction, adapted to attain the objects stated in the specification and the practical results thereby secured, with the impracticability or insufficiency of the devices of the prior art. The failure of defendants to avail themselves of said earlier devices' or improve them, and their bodily appropriation of the patented construction, is most persuasive upon the question of invention. The decree is affirmed, with costs.