From Casetext: Smarter Legal Research

General Electric Co. v. Wabash Appliance Corp.

United States District Court, E.D. New York
Jan 16, 1937
17 F. Supp. 901 (E.D.N.Y. 1937)


No. 7723.

January 16, 1937.

Howson Howson, of New York City (Hubert Howson, Merrell E. Clark, Alexander C. Neave, and John H. Anderson, all of New York City, of counsel), for plaintiff.

Darby Darby, of New York City (Samuel E. Darby, Jr., and Paul Kolisch, both of New York City, of counsel), for defendants.

In Equity. Suit by the General Electric Company against the Wabash Appliance Corporation and others.

Decree for plaintiff.

This is a suit for infringement of the Pacz patent No. 1,410,499, filed February 20, 1917, and issued March 21, 1922.

The patent contains claims covering a process and a product. However, only product claims 25, 26, and 27 are here in suit.

These three claims cover an incandescent lamp filament and read as follows:

"25. A filament for electric incandescent lamps or other devices, composed substantially of tungsten and made up mainly of a number of comparatively large grains of such size and contour as to prevent substantial sagging and offsetting during a normal or commercially useful life for such a lamp or other device.

"26. A drawn filament for electric incandescent lamps or other devices, composed substantially of tungsten and made up mainly of a number of comparatively large grains of such size and contour as to prevent substantial sagging and off-setting during a normal or commercially useful life for such a lamp or other device.

"27. A filament for electric incandescent lamps or other devices, composed of tungsten containing less than three-fourths of one per cent of non-metallic material and made up mainly of comparatively large grains of such size and contour as to prevent substantial sagging or offsetting during a normal or commercially useful life for such a lamp or other device."

The plaintiff contends that the filament covered by these claims represents an important advance in the art. The plaintiff and its licensees have undoubtedly sold billions of incandescent lamps embodying that filament. The defendants, Wabash Appliance Corporation, Abe Adler, and Abe Parker, the latter two being the sole stockholders and officers of the corporate defendant, are not licensed under the Pacz patent.

The plaintiff charges the defendants with infringement, in that the claim is made that the defendants make and sell electric incandescent lamps containing filaments which are in fact Pacz filaments of the claims in suit.

The history of the incandescent lamp began about fifty years ago, when Edison brought out the first commercial lamp that was successful. The Edison lamp contained a carbon filament and remained the standard of the industry until the advent of tungsten as a filamentary material.

Tungsten was disclosed in 1906 by the invention of Just and Hanaman. These tungsten filaments were of the squirted or pressed type, tungsten powder being mixed with a binding material in a plastic mass and then squirted under pressure through a die to a fine thread. It was then dried and sintered to form a filament of pure tungsten. Loops of this filament were then assembled on supports inside a glass bulb to form an incandescent filament. Plaintiff's Exhibit 8. This squirted tungsten filament was an improvement over the carbon filament but, because this filament was a little brittle and weak, it could not be made into long lengths or of accurate gauge.

While this tungsten filament was an improvement over the carbon lamp of Edison it, nevertheless, presented a problem known as "offsetting," which is the bodily shifting of a portion of the filament from an adjacent portion during the burning of the filament. These early tungsten filaments consisted of comparatively large crystals, many of which were large enough to extend clear across the filament, but they shifted. Because of the bodily shifting of portions of the filament (Plaintiff's Exhibit 12), the available cross sectional area of the filament for carrying electric current was reduced at the point of the "offsetting." The filament then became overheated and burned out.

The next step in the history of the filament was the Coolidge patent 1,082,933. His method of making the drawn tungsten filament included means for preventing "offsetting." Since "offsetting" was known to occur when, as Coolidge put it Plaintiff's Exhibit 2), the crystals became "so large as to extend across the entire section of the filament," Coolidge made provision (by the addition of certain chemical materials at an early stage of his process) for arresting crystal growth in the final filament. His theory was the maintenance of a small crystal structure. He did not have large crystals so as to extend the entire section of the filament. As a result of his method his filament was a non-offsetting filament.

The Coolidge method was, of course, a great improvement over the weak squirted filament. Its grains were so small and numerous that none extended all the way across the filament, which prevented slipping and in turn this made a non-offsetting filament. However, there was another serious defect characteristice of a tungsten filament. This was the defect of "sagging."

The Coolidge patent did not take care of the problem of "sagging." The filament in a lamp is not stiff enough to support itself and consequently must be strung between supports. The Coolidge filament would sag between its supports after use for some time.

"Sagging" deformed the filament from the position it was supposed to maintain and it would also reduce the efficiency of the lamp in time, causing its failure. This was a definite problem in the art. It was not solved by Coolidge. A large number of supports were used as a remedy, but this did not eliminate "sagging."

The problem presented to Pacz was to make a practical filament that would neither "offset" nor "sag." Pacz knew, like the others in the art, if the grains were allowed to grow the filament would "offset." If grain growth were prevented, like the Coolidge method, the filament would not "offset" but would "sag." The art taught one to avoid large grains since that would allow "offsetting." The problem remained unsolved until Pacz made the invention of the patent in suit.

Until the advent of this Pacz filament, the state of the art, undoubtedly, was such that experiments were being made in order to produce a filament that would not "sag."

The defendants have cited many patents as a defense to the one in suit and make a further claim that the Pacz patent did not improve the art existing at the time of the granting of his patent. These patents cited by the defendants can readily be distinguished.

To determine whether the insoluble problem was in fact solved by Pacz can easily be determined. Pacz made a multi-crystal filament having grains of large size to prevent "sagging" and of such contour that, even though the boundary between grains extended all the way across the filament, no slippage or "offsetting" occurred. Pacz was able to control the contour or shape of the grains so that they did not slip. He had had 217 failures and was finally successful on his 218th attempt. For that reason the filament is frequently referred to by number 218. This 218 filament, like the Coolidge filament, was drawn tungsten and was made in much the same way as that described by Coolidge. The difference between the Pacz filament and the Coolidge filament was that Coolidge overcame "offsetting" by adding small amounts of certain chemical materials at an early stage of the process which insured the maintenance of small grains in the final filament, while Pacz employed different chemical materials which produced the opposite effect of promoting in the final filament the growth of large grains of such contour that not only did it prevent "offsetting," as in the Coolidge filament, but solved the problem of "sagging."

The theory of Pacz was directly contrary to the prior teachings. The Pacz filament never produced a crystal boundary joining large crystals which were either disposed in the right position to "offset" or which were smooth, easily permitting the movement of one crystal away from the other. He produced large crystals not late in the burning life of the lamp but very early, obtaining a coarse-grained filament which did not "offset" and, since the filament was coarse-grained, it did not "sag."

As the structure of the filament under the Pacz method is comparatively coarse-grained, this was the reason given by Pacz why the filament did not "sag." However, that would mean to the art that the filament would "offset" because it was common knowledge in the art that where grain boundaries, large enough to extend across the filament, were produced, there would be bound to be slippage. Under the Pacz patent it is claimed that a particular kind of coarse-grained filament is produced which does not "offset." These coarse-grained boundaries are of such a nature that they prevent slipping action. The contour of these grains by the method used by Pacz is a very important element. Pacz found that he produced such a filament by the method disclosed in his patent in which a material intimately associated with the tungsten particles has an effect on the shape and size of the tungsten grains. The material which Pacz used for producing the desired result was an alkali silicate.

It is conceded by the plaintiff that no one knows even at the present time just why the Pacz process should make grains of such a size that the filament will not "sag" and of such a shape that the grains will not slip and produce "offsetting." Plaintiff's Brief, p. 11.

The claims now in suit were held valid by the Circuit Court of Appeals, Ninth Circuit, Anraku v. General Electric Co., 80 F.2d 958, which affirmed the decision of the United States District Court for the Southern District of California, 10 F. Supp. 935, in the case of General Electric Co. v. Anraku. A petition for writ of certiorari to the United States Supreme Court was denied. 298 U.S. 678, 56 S.Ct. 942, 80 L.Ed. 1399.

The opinion of the Ninth Circuit Court of Appeals sustained the claims of the patent now in suit in spite of the defenses interposed in that action, many of which are before the court in this case.

The first proposition that the defendants advance is that the Pacz product does not constitute a patentable invention. The defendants further claim that Pacz did not describe his invention in such full, clear, and exact terms that any one skilled in the art, subsequent to the expiration of the patent grant, could make the patented device solely from the information contained in and disclosed by the patent, and cites the case of General Electric Company v. De Forest Radio Co. (D.C.) 17 F.2d 90, in support of this proposition. The cited case involved the Coolidge patent. In the Coolidge case, the product claims defined the article by characteristics and not by structure. Claim 25 of the Pacz patent specifies that there shall be a number of grains and that they shall be comparatively "large" and of such "size" and "contour" as to prevent substantial "sagging and offsetting." This made a new structure. It is patentable as a product quite apart from the process. Dunn Wire-Cut Brick Co. v. Toronto Co., 259 F. 258 (C.C.A.6).

There was no novelty alone in the product of a sagless filament, but there was in one of a particular structure having a number of large grains, multi-grained, and of a special contour. This was new and novel to the art and clearly distinguished the Schaller patent, which had a single crystal filament that was sagless but a failure.

Secondly, the defendants claim that a valid patent cannot be granted by the government unless the claims of the patent are in such clear and precise terms that the public, during the life of the patent, may know what is excluded from it and what is vested in the patentee, defining the boundary lines of the monopoly of the patent clearly.

There is ample proof that the plaintiff's present commercial process does use the Pacz method as defined in its patent and produces not only a non-offsetting filament but also a sagless filament. Millions of lamps containing a filament having the Pacz structure are being sold to the public and the structure is new and novel to the art and is therefore patentable. The defendants' witness, Mr. Laise, admitted that Defendants' Exhibit KK was commercially a non-sagging and non-offsetting filament as was Defendants' Exhibit LL. Rec. p. 375.

There is a sufficiency of disclosure and the claims are so clearly set forth that one skilled in the art would have no difficulty in making a Pacz filament. The defendants set forth the rule laid down by Judge Learned Hand, speaking for the Court of Appeals for the Second Circuit, in Health Products Corporation v. Ex-Lax Manufacturing Co., 22 F.2d 286, 287, in which he stated: "It is scarcely necessary to labor the principle that specifications must be more than a suggestion for promising experiment, hit or miss. They must contain complete directions, leading with certainty to the result."

The defendants' wire is undoubtedly a Pacz filament. Both the plaintiff's and the defendants' filaments are drawn, they are both made of tungsten, they both have less than three-fourths of one per cent. non-metallic material, they consist of a number of comparatively large grains so as to prevent sagging, and the grains are of such contour and shape so as to prevent offsetting.

Dr. Jeffries, called by the plaintiff, frankly and with no intention to deceive the court stated the formula of materials which made up the Pacz filament. On the other hand, the defendants' witness, Mr. Laise, when called upon to state the formula of materials used by the defendants, refused to reveal the secret of the manufacture of the defendants' filament. This could not be merely a coincidence or accidental, and in itself answers the argument of the defendants that the plaintiff's patent is invalid because of insufficiency of disclosure. The disclosure was certainly sufficient so that the defendants could copy same.

As to the defendants' claim that the plaintiff does not use the present commercial process of the Pacz patent, suffice it to say that the evidence clearly indicates and shows that the filament 218 is made directly pursuant to the claims. The materials used in making this Pacz filament are easily understandable to one skilled in the art. They were apparently imitated without trouble by the makers of the defendants' filament.

The defendants claim that the plaintiff's process will not produce a sagless filament. The evidence showed that what the manufacturers of filaments were trying to make was a commercially sagless filament. Many of the lamps containing the Pacz filament were burned for demonstration purposes before the court, and from all of these tests the court is convinced beyond any doubt that the Pacz 218 wire makes a commercially sagless filament. This is particularly obvious when a comparison is made between the Coolidge filament and the Pacz filament.

At the time these demonstrations were made, the Coolidge filament sagged to some extent while the Pacz filament and the defendants' filament did not sag.

The defendants' expert, Mr. Laise, admitted that the defendants' filaments were commercially non-sagging. Defendants' Exhibits KK and LL, supra. A great deal of finely spun language was used by the witnesses in describing the action of the different filaments in evidence but the demonstrations, as seen by the court, were proof that could not be questioned and the interpretation to be placed on the action of the filament as seen by the court could hardly be described as well by words or language.

That the burning of these filaments before the court indicated unquestionably that the Pacz patent differed from the prior art in a great degree could be clearly ascertained by these demonstrations. The lamps containing the Coolidge filaments as stated before could be seen to "sag" slowly but surely, while the Pacz and defendants' filaments did not react that way. Pacz must have, in 218 experiments, utilized a great deal of time and energy to gain the objective which his filament obtained. Undoubtedly, all other manufacturers of filaments were striving to gain the same end.

"Sagging" was a definite problem in the art up to the advent of the Pacz filament. It is conceded that "sagging" destroys the longevity of the filament which in turn would destroy the commercial aspects of the filament. The Pacz filament assisted materially in the commercial success of that filament and differs from the prior art to a great degree because of its structure. The making of the Pacz filament could hardly be said to be carrying forward a new or more extended application of the original thereof. It was not a change in form only to some degree but the tests showed a marked change in the "sagging" problem of the Pacz filament as against the "sagging" of the other filaments prior to the Pacz patent. Pacz accomplished by his invention something different in principle and in kind from the prior art, and did so in the face of the contrary teaching contained in the Coolidge patent.

The Schaller patent, as already indicated, taught the world to keep away from multi-grained filaments because they could not be controlled. The patent in suit teaches the opposite theory. It has been proved that the Schaller patent was not a success commercially or otherwise and taught the art nothing of a substantial nature. Certainly, it did not teach the theory contained in the Pacz patent.

The Scoular patent represents one of the squirted filaments and to make same one has to use some type of binder for the tungsten powder. There is no mention in that patent of "sagging" or "offsetting." The Scoular wire was swaged and drawn (Rec. p. 416) and the filament did "offset" (Rec. p. 415).

The Kruger patent did not solve the problem confronting the art. It does not mention "sagging" or "offsetting," or large crystals or the use of an alkali silicate.

The Kremenzky German patent, about which no testimony was offered by the defendants, seems particularly directed to the squirting method.

"Die Welt der Technik," a German publication (Defendants' Exhibit HHH), which was introduced without testimony, speaks of the problem of "offsetting" but contributes nothing of interest concerning its solution.

The defense that the Pacz patent is invalid because of double patenting is predicated on the prior Pacz patent 1,299,017, granted April 1, 1919, on application filed May 25, 1914, three years before the Pacz patent in suit was applied for. There is also cited a prior use and sale of the Pacz filament in suit. The Westinghouse Company did sell in 1911 or 1912 a filament which contained four different binders, but the testimony is not clear which one of the binders was used at any time, nor was any of this wire nor any of the lamps with the wire produced into evidence. Certain exhibits were offered in evidence during Mr. Cowles' testimony, but his indefiniteness as to the size of the filament and as to the magnification of the picture left the matter very hazy. This same witness stated that he was in the employ of the Independent Lamp Company from 1912 to 1920, during which time he was making experiments, but there is no testimony as to any prior sale or use by this Independent Lamp Company. If there were double patenting, it would be based on the proposition that a patentee can have but one patent for a single invention as found in Miller v. Eagle Manufacturing Company, 151 U.S. 186, 14 S.Ct. 310, 38 L.Ed. 121, but this has not been established here.

Pacz, in his earlier patent, followed the same theory as Coolidge, which indicated that the only way to prevent "offsetting" in a multi-grained filament was to avoid coarse grains and insure a finely granular structure. In his early patent, Pacz stated as follows: "If the cross-section is shiny or shows coarse grains the filament will offset. On the other hand, if the structure is finely granular, the filament will not offset. Rods produced from the metal, as above described, when subjected to this test had cross-sections which were finely granular."

It will be readily seen that the Pacz patent in suit is distinctly different from the earlier one. The structure obtained by the first patent of Pacz was nothing like the one obtained in the second Pacz patent. Therefore, there was no double patenting within the meaning of the law.

That the Pacz patent was infringed is clearly indicated by the evidence on the trial. The tests of the Pacz filament and the defendants' filament proved conclusively that the filaments were one and the same. It is unavoidable to arrive at the conviction that the defendants' filament was made exactly as was the plaintiff's. The microphotographs showed that the defendants' filament was made exactly like the plaintiff's. The qualities of the filaments were as alike as "two peas in a pod," and the burning tests in open court proved that the two filaments were twins. The testimony showed that they were both drawn, both were made of tungsten, both had less than three-fourths of one per cent. of nonmetallic materials, both consisted of a number of large grains, large enough to prevent sagging and offsetting in spite of the fact that they extended clear across the filament. Rec. pp. 249, 250.

In disposing of some of the defenses interposed by the defendants, the court can hardly refrain from stating that "imitation is the sincerest form of flattery." The defendants surely must think that the plaintiff's filament is an excellent one or they would not have made theirs exactly like it. When the defendants' expert, Mr. Laise, declined to divulge to the court the formula of the filament manufactured by the defendants, the conclusion reached by the court was that the defendants were neither sincere nor frank. No sensible explanation for this refusal was given in spite of the fact that testimony revealed that the plaintiff's wire and the defendants' wire were made exactly alike. The expert for the defendants testified that the defendants' filament was not made similar to the plaintiff's wire, but refused to tell in what particular or manner it was made different from that of the plaintiff. Rec. pp. 464, 465, ques. 281-289, inclusive.

The defendants Wabash Appliance Corporation and the individuals Adler and Parker are one and the same. The acts of the corporation are the acts of the individuals and the individuals are liable for the corporate acts of the corporation. This rule is clearly laid down by Judge Manton in Claude Neon Lights v. American Neon Light Corporation (C.C.A.2) 39 F.2d 548.

Accordingly, the plaintiff may have a decree for injunction and accounting.

Summaries of

General Electric Co. v. Wabash Appliance Corp.

United States District Court, E.D. New York
Jan 16, 1937
17 F. Supp. 901 (E.D.N.Y. 1937)
Case details for

General Electric Co. v. Wabash Appliance Corp.

Case Details


Court:United States District Court, E.D. New York

Date published: Jan 16, 1937


17 F. Supp. 901 (E.D.N.Y. 1937)

Citing Cases

Gen. Electric Co. v. Wabash Co.

The patent, issued March 21, 1922, on an application filed February 20, 1917, contains process and product…

General Electric Co. v. Wabash Appliance Corp.

Suit by the General Electric Company against the Wabash Appliance Corporation and others for infringement of…