Ruben Condenser Co. v. Aerovox Corp.

CAMPBELL, District Judge. This is a patent suit. Plaintiffs are seeking relief by injunction and money damages for the alleged infringement by the defendant of patent No. 1,891,-207, issued to Samuel Ruben, assignor by mesne assignments to Ruben Condenser Company, for electrolytic condenser, dated December 13,1032, on an application filed June 19, 1930. The plaintiff Ruben Condenser Company is the owner of said patent, and the plaintiff P. R. Mallory & Co., Incorporated, is the exclusive licensee under said patent. The plaintiff Ruben Condenser Company is the owner of other patents of Samuel Ruben, relating to dry electrolytic condensers, and the plaintiff P. R. Mallory & Co., Incorporated, is the exclusive licensee under such Ruben patents. For convenience I will hereinafter refer to the Ruben Condenser Company as the Ruben Company, and to P. R. Mallory & Co., Incorporated, as the Mallory Company. The Mallory Company manufactures and sells dry electrolytic condensers under the Ruben patents, and has issued sublieenses to others to manufacture and sell dry electrolytic condensers under said patents. The defendant, Aerovox Corporation, manufactures dry electrolytic condensers under the two Ruben patents, Nos. 1,710',073 and 1,714,191, being licensed by the Mallory Com-' pany, but has no license rights under the Ruben patent, No. 1,891,207. Plaintiffs’ suit is based on claims 1, 5, 7, 9,13,16,19, and 22 of the Ruben patent, No. 1,891,207, in suit, and specimens of the alleged infringing electrolytic condensers man-ufaetured and sold by defendant are in evidence as Exhibits 4, 5, and 6. ■ By amendment to the bill of complaint, the plaintiffs also charge the defendant with unfair competition. The defendant is the owner of the two following described patents: Patent No. 1,789,949, issued to Alexander, Georgiev, assignor to Aerovox Wireless Corporation, for electrolytic cell, dated January 29, 1931, on an application filed October 18, 1939; and Patent No. 1,815,768, issued to Alexander Georgiev, assignor to Aerovox Wireless Corporation, for electrolyte, dated July 21,1931, on an application filed December 9, 1930. To the bill of complaint and the amendment to the bill of complaint, the defendant interposed by answer the defenses of invalidity and noninfringement, and denied unfair competition, and in addition set up a counterclaim against the plaintiff Mallory Company for the alleged infringement of claims 11, .18, and 19 of said Georgiev patent No. 1,789',949 (specimens of the alleged infringing dry electrolytic condensers manufactured and sold by plaintiffs arc in evidence as Plaintiffs’ Exhibits 8 and 10 and Defendant’s Exhibits P-2, Q-2) and for the alleged infringement of claims 1, 2, 3, 4, 8, 10, 11, 12, and 13 of the said Georgiev patent No. 1,815',768 (specimens of the alleged infringing dry electrolytic condensers manufactured and sold by plaintiffs are in evidence as Plaintiffs’ Exhibits 7 to 14, both inclusive, and Defendant’s Exhibits P-2 and Q-2). The defendant also pleaded in its counterclaim a cause of action for alleged unfair competition. The plaintiffs by their reply denied validity and infringement of the said Georgiev patents, asserted a right of immunity from suit, and denied unfair competition. The defendant offered no evidence on the trial in support of its charge of unfair competition. All of the three patents involved in this suit relate to electrical condensers of the type known as “dry electrolytic condensers.” An electrical condenser is a device for storing a charge of electricity, consisting essentially of two conducting electrodes separated by an insulating dielectric. In the modern types the electrodes are metallic plates or foils, and the dielectric is a sheet of mica or paper. The sapacity of such condensers, that is, the amount of electrical energy whieh they are capable of storing, varies inversely as the thickness of the — 170 — dielectric, and, as glass, mica, and paper have very tangible thickness, the capacity which can be obtained per unit area of electrode when they are used is small in comparison with the capacity of a condenser in which the dielectric has only an immeasurable thickness, such as one in which use is made of the property of certain metals, aluminum being the common commercial example, for forming a dielectric film on the surface, which is of exceedingly minute thickness, probably not much thicker than a molecule. The so-called “wet electrolytic condensers” were the first type to go into commercial use, and in them an aluminum electrode, coated with a thin dielectric film, is immersed in a liquid electrolyte solution, which serves the function of the complementary .electrode of the condenser by conducting the electric current up to the dielectric film. In 1928 and 1920-, wet electrolytic condensers on the market were made to stand about 425 volts. Work was being done in Georgiev’s laboratory on wet condensers in 192.9', and wet condensers sold by other concerns at that time were rated at 450 volts. Sprague’s wet electrolytic condensers were rated at 400 volts around the end of 1929v Patent No. 1,714,191, issued to Samuel Ruben, for electrical condenser, dated May 21,1929', on an application filed December 23, 10216, discloses a dry electrolytic condenser composed .of two strips of aluminum foil, one of which has a dielectric film formed on it, the two foils being separated by a gauze spacer which contains a viscous electrolytic composition of glycerin, sodium bicarbonate, and boric acid. Patent No.' 1,710,073, issued to Samuel Ruben, for electrical condenser, dated April 23, 1929, on an application filed March 21, 1927, discloses a similar construction with a viscous electrolyte composition of .glycerin, borax (sodium borate), and boric acid. , Suit was brought against the present defendant in this court, on' June 3,1931, for infringement of these two patents, and on February 25,1932, defendant settled the case and took a license under them, and has since paid royalty on all dry electrolytic condensers manufactured by it. As a part of that settlement the defendant was given an option for a license, at an additional royalty, on what eventuated into Ruben patent, No, 1,891,207, but it elected not to exercise such option for a license. The Ruben patent, No. 1,891,207 disclosed •the making of a dry electrolytic condenser with an electrolyte comprising ethylene glycol, ammonium borate, and boric acid, compounded in such a way that the finished electrolyte contains crystals suspended in a viscous liquid. The superiority of such condensers had been determined by Ruben, and he filed his application for the patent in suit prior to any activity by the defendant or Georgiev with respect to such an electrolyte. Defendant’s chief attack on the Ruben invention appears to be based on the alleged commercial priority of the defendant’s electrolyte, but that is not a' determining factor, as under our laws (Rev. St. § 4886', title 35, § 31, U. S. C. [35 USCA § 31]) priority of invention is controlling unless the first inventor abandons his invention. Ruberi was the prior inventor of the invention of the patent in suit and did not abandon it. Priority in marketing a so-called 500-volt dry electrolytic condenser does not furnish the criterion by which to determine the validity of all of the claims of the Georgiev and the Ruben patents here involved, as not one of the claims of any of those patents says anything about a SOO-volt condenser. In fact, it seems doubtful to me that the defendant in 1930 or 1931 made better than a 400-volt operating 500 peak voltage condenser. The patent in suit, which is the result of the work of Samuel Ruben, carries forward the work represented in and is an improvement of his patents Nos. 1,710,073 and 1,714,-191. That it was the work of Ruben is shown by the following facts in evidence: The work of Lowenstein under Ruben’s instruction as far back as October, 1928, when Georgiev was working on 6 and 12 volt condensers at Aerovox, and there is no evidence to show that Georgiev then knew anything about the desirability of boiling an electrolyte: The delivery by Ruben, under date of March 29,1929, to “Doe” Shoemaker of rather complete specifications for KMPvolt condensers. The notes of Lowenstein, of May 18,1929, regarding an electrolyte for high voltage condensers. The reproduction of this electrolyte in 1933, by Knowles. The sending by Ruben in August, 1928, to Grigsby-Grunow Company, of specifications for production of high-voltage dry electrolytic condensers. .. • — 171 — That Georgiev had made no approach! to any such voltage at that time. The comparisons made by Raines in May, 19'30, under instructions given by Ruben, of condensers with glycerin and ethylene glycol electrolytes which showed the superiority of condensers made with ethylene glycol electrolytes over those made with glycerin electrolytes, the sending by Ruben of his specifications to the Mallory Company and to the Grigsby-Grunow Company, the history of the subsequent work at the Mallory Company and the Grigsby-Grunow Company, which shows that when they followed Ruben’s instructions the results were excellent, and that, when they could not or would not, they ran into trouble. Ruben did not testify that his electrolyte becomes chilled at 85° C. Dr. Ffankol gave that figure, but did not use the word “chilled.” Dr. Ffankel says that he did not obtain a clear solution called for by the patent, but this was evidently because he failed to follow its instructions correctly, as Raines described the process. That the invention of the patent in suit was the work of Ruben is not, in my opinion, shaken by the testimony of Garstang, a former employee of the Mallory Company, and a novice in the art of dry electrolytic condensers, that he did not succeed in making a good condenser before January 1,1931. A patent is addressed to those skilled in the art. The determination of the proper degree of heat to be used is fairly within that skill. American Stainless Steel Co. v. Lud-lum Steel Co. (C. C. A.) 290 F. 103, 108.' I am convinced that, if he had followed instructions, he would have succeeded generally, and he himself admits that he made a Ruben condenser which stood 500 volts, on September 10,1930. Assuming that Garstang is correct in disclaiming responsibility for the flyleaf, the fact remains that he Credits himself with writing, on the day before Christmas, 1930, the text of a booklet on Elkon condensers, in which he claimed 575 volts for the Ruben condensers. An ethylene glycol electrolyte is superior to a glycerin electrolyte in dry electrolytic condensers, in a real practical way, in that it has a lower power factor, which is important in A. C. motor starting condensers, where the power loss in the condenser must be kept at a minimum; changes capacity less at low temperatures, which is of importance where apparatus employing condensers is to be used out of doors; and has a greater capacity per unit area of foil than a condenser with a glycerin electrolyte; and therefore- a condenser of a given capacity can be made in a smaller volume with a glycol electrolyte than with a glycerine electrolyte. The ethylene glycol electrolytes give lower leakage than the glycerin electrolytes, and this is a positive advantage. While the cost of ethylene glycol is greater than the cost of glycerin, and the cost per condenser with ethylene glycol electrolyte is greater than the cost per condenser with glycerin electrolyte, the advantages are such as to warrant the difference in cost to the users. Ruben completed an ethylene glycol electrolyte as early as May 23,1930', and filed his application covering it on June 19,1930-, and Georgiev was not working with ethylene glycol before July 1, 1930, when Siegel bought a gallon of ethylene glycol because Ffankel suggested that it be tried along with the other things, such as glucose. Glycol and glucose were among the things which Georgiev, in November, 1930; listed to be tried as experiments, and did not, on December 9, 1930, know that ethylene glycol would malee a superior electrolyte. Ruben has an earlier date of invention and reduction to practice of an ethylene glycol electrolyte than Georgiev, and the Ruben patent here in suit is not anticipated by the Georgiev patent, No. 1,815,768. Ethylene glycol stands midway between ethyl alcohol and glycerin. The natural supposition was that it would not make as good an electrolyte as glycerin, as ethyl alcohol is not used for condenser electrolytes. It was impossible, despite the similarities between the two substances, to predict that, because glycerin had made a good electrolyte in the past, ethylene glycol would make a good electrolyte in the future. The discovery, which one would not naturally suspect, that glycol makes an excellent electrolyte, was Ruben’s and not Geor-gieVs, and the discovery was such a good oiio that even the defendant has found it necessary to discard the old glycerin electrolyte entirely. The first limitation which the defendant seeks to impose on the patent in suit is based on Ruben patent, No. 1,710',073: Ruben patent, No. 1,710',073, discloses a diy electrolytic condenser with a viscous electrolyte of borax, boric .acid, and glycerin. Such an electrolyte the defendant is entitled to make under its license. The defendant — 172 — Las not the right under the license to make an ethylene glycol ammonium borate boric acid electrolyte, which is not disclosed in Ruben's patents, Nos. 1,710',073 and 1,714,191, but is covered by Ruben patent, No. 1,891,207. The license in question, which relates to Ruben patents, Nos. 1,710,073 and 1,714,191, provides in part: “No license is granted under any other patent now or hereaf ter owned or controlled by Mallory claiming subject matter not disclosed in said patents or either of them.” Glycerin is named in patent No. 1,710,073, and glycerin is a polyhydrie alcohol, but all polyhydrie alcohols, including ethylene glycol, are not disclosed in Ruben patent, No. 1,710,073. The claims of Ruben patent, No. 1,891,207, which are involved in this suit, call for a glycol or specifically for ethylene glycol, and every one of such claims definitely excludes glycerin from its scope, as glycerin cannot be said to be a glycol. The defendant has since December 13, 1932, the date of the issuance of Ruben patent, No. 1,891,207, in suit, sold condensers that have been made with an ethylene glycol electrolyte containing crystals in suspension. The second limitation which defendant urges against the Ruben patent, No: 1,891,-207, in suit is based on an interference between Ruben and Edelman. In that proceeding Mr. Ruben stated over his signature, which he identified, as follows: “I, Samuel Ruben, being a party to the above entitled interference, in pursuance of an agreement entered into the 26th day of December, 1929, hereby agree that no testimony will be filed in my behalf in the above entitled interference, and request that this statement be taken as the basis for an immediate judgment on the record in favor of said Philip E. Edelman' in said interference.” I do not see the relevance of the interference proceedings to any of the issues before this court. The Edelman application which was involved in this interference related to an electrolytic condenser in which the electrolyte comprised a mixture of one part of ammonium phosphate and eight parts of sodium phosphate, heated with only enough distilled water to make up for evaporation loss occurring during the process. When cooled, the salts crystallize out as a solid interlinked mass of crystals, resembling a rock in hardness and containing no fineombined solvent liquid. ■ That there was a marked difference between the paste electrolyte of the Ruben patent, No. 1,714,191, and the Edelman electrolyte of rocklike hardness, containing no uncombined solvent liquid, was recognized by the Patent 'Office. Nothing in the Ruben-Edelman interference affects either the validity or the scope of the claims of the Ruben patent, No. 1,891,-207, in suit, as such patent does not relate to any such composition, and neither the Mallory Company nor the Aero vox Corporation uses any such electrolyte. The third limitation which the defendant seeks to impose is based on a statement in the specification as filed, that Ruben's ethylene glycol ammonium borate boric acid electrolyte, prepared as he describes, forms a clear solution when hot and crystallizes as the mixture cools. Defendant contends that the invention and claims are limited to a dense, immobile, or putty-like electrolyte. This contention is not sustained. When the directions of the patent are followed exactly as to proportions and procedure, a viscous mass is obtained with crystals in suspension. Claims 1 and 5 as filed with the original application describe the product accurately. Defendant cannot escape liability because of the use of the word “crystallize” in the patent in suit. When it is said that* a solution crystallizes, it commonly refers to what happens to the electrolyte of the Ruben patent, No. 1,891,207, in suit, to the Mallory electrolytes, and to the present day Aerovox electrolytes; one of the components present in the solution becomes more insoluble and crystallizes. From that standpoint a “solution which crystallizes,” and a “solution in which crystals form,” are the same thing. The bill of particulars limits the date which Ruben can claim for the making of his invention. It does not limit the scope of his claims, as that is dependent entirely on the prior art. It hardly seems necessary to refer specifically to the prior art patents and publications, which defendant refers to under the heading “Equivalence of Glycerol and Glycol,” because they obviously taught no one the superiority of the ethylene glycol electrolyte invented by Ruben, nor bow to make such an electrolyte. Even the defendant.does not use any electrolyte described in any of the prior patents which it cites, nor in the Georgiev electrolyte patent, but pays to the patent in suit the sineerest form of flattery, imitation, at the risk of an action for infringement. ■ — 173 — The commercial success of the patent in suit is impressive. The Mallory Company has nine licensees. The Mallory Company, the Sprague Specialties Company,, and Electro Formation, Inc., sold during the year 1933 condensers with ethylene glycol electrolytes, under Ruben patent, No. 1,891,307, the patent in suit, amounting to $1,738,000. And it must not be forgotten that the defendant showed its high opinion of the patent in suit by making sales of $653,000' during 1933', which is not included in the last-mentioned sum; all of the dry electrolytic condensers sold by the defendant since the middle of 1933 having contained an ethylene glycol electrolyte. The Ruben patent, No. 1,891,307, is valid. The claims of the Ruben patent, No. 1,891,207, in suit, are Nos. 1, 5, 7, 9, 13. 16, 19', and 32, and they relate to ethylene glycol electrolytes and not to glycerin electrolytes. Claims 1, 5, 7, and 9 of the patent in suit derive from the claims bearing the same numbers in the application as originally filed. Claim 1 reads as follows: “1. In an electrolytic condenser the combination, with spaced electrode plates, one of which is elee-trolytieally film-formed, of a conductive composition therebetween, containing ethylene glycol and a film-maintaining electrolyte of one of the weak acids, boric, citric, malic, lactic, tartaric and phosphoric with a salt of one of said acids suspended therein, the whole forming a crystallized electrolyte.” Claim 5 reads as follows: “5. In an electrolytic condenser, the combination with spaced electrode plates one of which is elee-trolytieally film-formed, of a conductive composition therebetween containing ethylene glycol and a film-maintaining super-saturated electrolyte of a weak acid and a salt thereof suspended therein, the proportions of ethylene glycol and electrolyte being such that a mixture of paste consistency is formed when heated to 60° C.” Claim 7 reads as follows: “7. In an electrical condenser, an electrolyte composed of crystalline compounds of ammonium borate, boric acid and ethylene glycol.” Claim 9 reads as follows: “9. In an electrical condenser, an electrolyte composed of crystalline compounds of ammonium borate, boric acid and a glycol.” These claims read on the defendant’s condensers. The raw materials of the defendant’s electrolytes are ethylene glycol, ammonia and boric add, and the ammonia reacts with some of the boric acid to form ammonium borate. The raw materials being mixed, compounds are formed, and in the defendant’s condensers which were opened in court crystals were present. The contention of some of the defendant’s witnesses that crystals are not present in defendant’s electrolyte prepared according to the stipulation, in which the boiling is carried to 280° F. (138° C.), is not sustained, even if the contents of the bottle offered in evidence as Exhibit X were clear and apparently free from crystals, as the condition of this bottle, with the cork waxed, was most favorable to inhibiting the precipitation of crystals, and it is worthy of note that no such precautions were taken with the bottle containing the defendant’s electrolyte boiled only to 370° F. (132° C.). Claim 13 derives from a claim inserted by amendment under date of July 30', 1931. Claims 16, 19, and 32: derive from claims entered by amendment dated November 17, 1931. These four claims describe in more detailed language the physical nature of the electrolyte composition disclosed in the specification as- originally filed, using the phrase “said electrolyte in appearance being a viscous mass of a syrupy material in which solid particles are suspended, and of at least molasses-like consistency.” Claims 16’, 19, and 23 add the fact that the composition includes the products of reacting the glycol component with a borate or with ammonium borate and boric acid, at substantially the boiling temperature. It is stipulated that the defendant’s electrolytes are made by mixing ethylene glycol, boric acid, and ammonia water, heating until there is a clear solution, and boiling with the condenser sections submerged in the solution, until the boiling pomt reaches 379° F. or 380'° F. When the ammonia water and boric acid are mixed, the ammonia combines with some of the boric acid to form ammonium borate, so that this compound is present in the mixture to react with the ethylene glycol compound. The defendant’s condensers opened in court established the viscous and syrupy nature of the defendant’s electrolyte and its molasses-like consistency, and, I have hereinbe-fore said, the presence of solid particles suspended in the syrupy material was also established. — 174 — All of the claims in suit are infringed. The charge of unfair competition made by plaintiffs against the defendant is based upon letters written by the defendant to competitors and others,- giving them notice of defendant’s claim to the invention of patent No. 1,815,768, I can find nothing to condemn in this action ; on the contrary, it seems to have been a proper' and considerate way of giving notice. Art Metal Works v. Auto Match Corp. (D. C.) 36 F.(2d) 964, 956. This conclusion is strengthened by the fact that such notices wore not sent out until after the adjudication in Aerovox Corp. v. Concourse Electric Co., Inc. (C. C. A.) 65 F.(2d) 386. The letters with reference to the institution of suits were true and warned against infringement. The correspondence with R. C. A. Victor Company, charging it with infringement, seems to have been in good faith, as it was followed by a suit on said patent against the metropolitan distributor of its radio sets. The charge of unfair competition made by the plaintiffs against the defendant is not sustained. This brings us to the counterclaim of the defendant. This is based on patent No. 1,815,768, issued to Alexander Georgiev, assignor to Aerovox Wireless Corporation, for electrolyte, dated July 21, 1931, and patent No. 1,789,949, issued to Alexander. Georgiev, assignor to Aerovox Wireless Corporation, for electrolytic cell, dated January 20, 1931. Both of these patents are owned by defendant, and the patent for electrolytic cell is hereinafter designated as the structure patent. The defendant charges that all of the Mallory condensers infringe claims 8,10', 11, 12, and 13 of the Aerovox Electrolyte patent, No. 1,815,768, and that the Mallory condensers' that infringe claims 1, 2, 3, and 4 are'Exhibits 7, 8, and 9, and P-2 and Q-2, respectively. Defendant is licensed by the Mallory Company under Ruben patents, Nos. 1,710,-073 and 1,714,191. Both of the Aerovox patents were adjudicated valid and infringed, by the Circuit Court of Appeals for the Second Circuit, in Aerovox Corporation v. Concourse Electric Co., 65 F.(2'd) 386, but claim 18 of Georgiev patent, No. 1,780,949, and claim 13 of Geor-giev patent, No. 1,815,768, which are involved in the present suit, were not involved in that suit. The electrolyte'patent, No. 1,815',768, discloses-the preparation of an electrolyte for dry electrolytic condensers involving the selection of raw ingredients, which, except glycol or ethylene glycol and glucose, were long known in the patent literature as useful for electrolyte purposes in condensers of the type under consideration: The patentee in his specification says: “I have also found that while glycerol is to be preferred, it may be wholly or partly replaced by any alcohol with two or more hy-droxyl radicals, such for instance as glycol, glucose, etc., all of which I have designated as a class, by the term, polyhydroxy alcohol. * * *» The patentee never made, nor did he invent, an electrolyte using glycol or glucose; on the contrary, the use of ethylene glycol in making dry electrolytic condensers was the invention of Samuel Ruben, who completed an ethylene glycol electrolyte as early as May 22,1930, and filed his application covering it on June 19,1930; Avhereas Georgiev was not working with ethylene gfycol before July 1, 1930', which was subsequent to the filing of the Ruben application, when Siegel bought a gallon of ethylene glycol because Frankel, not Georgiev, suggested that it be tried along with other things, such as glucose. Georgiev did not experiment with glycol or glucose, and did not know on December 9,1930, when he filed his patent application, that glucose would not make a satisfactory electrolyte, and that ethylene glycol would make a superior electrolyte. Ethylene glycol stands midway between, ethyl alcohol and glycerin. The natural supposition was that it would not make as good an electrolyte as glycerin, as ethyl alcohol is not used for condenser electrolytes. It was impossible, despite the similarities between the two substances, to predict that, because glycerin had made a good electrolyte in the past, ethylene glycol would make a good electrolyte in the future. No1 one had used ethylene glycol in any electrolytic condenser before Ruben. The patentee in his specification says: “An important feature of the invention is the use in the active principle of the electrolyte of a boron radical suitable in the forming of aluminum electrodes, but in a compound having a greatly enhanced dissociation capacity, whereby the chemical reaction is expedited, the resulting film has a high spe175 ciñe inductive capacity and greatly enhanced capability of withstanding high voltages.” “The compound used is preferably the ammonium salt of glyceryl borate alone or admixed with glyceryl borate itself, or an alkali metal salt of glyceryl Borate may be substituted for part or all of the said' ammonium salt.” “The above electrolyte is normally an oleaginous liquid, and it is preferably used with enough added water or other thinning fluid, such as methyl alcohol to bring its viscosity at a temperature of 200° F. down to between 3 and 4.5, as compared with the viscosity of water at 60 degrees F.” “The ingredients used are preferably boric acid, .glycerol and ammonia, either gaseous or as ammonia water. For producing the electrolyte, the ingredients may be combined in the proportion of one thousand (1,000) grammes of glycerol, six hundred and twenty (620) grammes of boraeie acid (substantially stochiometrie proportions) and fifty (50) c. c. or more of twenty-six (26%) per cent, ammonia water or the equivalent amount of ammonia gas.” “While no less than the minimum proportion of ammonia water or ammonia above specified is to be used, to attain the best results, much larger proportions of this ingredient may be employed, if desired. The ■greater the proportion of this ingredient, the greater the percentage of ammonium glyceryl borate in the resultant electrolyte, and the less the proportion of glyceryl borate.” “By the simple expedient of heating the mixture as above set forth to boiling and continuing until the boiling point at atmospheric pressure is above 130° C., and holding the solution at this temperature for preferably at least five minutés, the preparation of the electrolyte is completed.” “Prolonged ebullition of the electrolyte, .which is maintained heated in the tank in which the condenser sections are impregnated during part or all of the time of impregnation, causes some or all of the free water .released according to the reactions above stated to evaporate and the electrolyte becomes thicker and more viscous, and its boiling point rises, without change however in the stable chemical composition of the glyceryl borate and its ammonium salt.” “Such thickened electrolyte when its viscosity becomes considerably in excess of 7, for instance, is undesirable for a number of reasons among which are: a drop in the capacity,. an increase in the power .factor and inferiority of the filtering action of the resulting condenser, retardation in the impregnation, and greater thickness of the electrolyte in the finished condenser roll or stack, with consequent increase in volume of the resultant product.” “My electrolyte is readily restored to the proper degree of viscosity without impairing its. properties for the present purposes, by the addition thereto of appropriate quantities, of a neutral thinning liquid, such as methanol or acetone, but ordinary faucet water is quite satisfactory and is ordinarily preferred. The viscosity of the electrolyte may thus readily be maintained between the desired limits of 3 and 7, and preferably between 3 and. 4.5 and the eleetro&te may be used and re-used for impregnation of successive batches of condensers, and -without waste or impairment of the efficacy of the electrolyte. “Where water in excess of that here specified is added, or is originally present, the forming process is retarded, and corrosion of the plates or foils more likely, to occur, but reasonably satisfactory results are obtained though the electrolyte in original preparation is boiled for a much less period of time than according to the preferred practice above set forth.” The plaintiffs argued and offered much evidence in support of their contention that the Georgiev electrolyte patent in suit was invalid, but it does not seem to me to be necessary to discuss that phase of the case, as it appears clear to me that the Mallory condensers do not infringe. Claims 1, 2, 3, and 4, of patent No. 1,815',-768 are for the electrolyte, and read as follows : Claim 1: “1. As an electrolyte for an electrolytic cell, a liquid including more than five percent of an alkali salt of glyceryl bo-rate and having at 200'° F. a viscosity of three (3) to seven (7) inclusive as compared with the viscosity of water of 60° F. Claim 2: “2. As an electrolyte for an electrolytic cell, a liquid including more than five percent of the ammonium salt of glyceryl borate and having at 200° F. a viscosity of three (3) to seven (7) inclusive as compared with the viscosity of water of 60 ° F. Claim 3: “3. As an electrolyte for an electrolytic cell, á liquid including a mixture of glyceryl borate with the ammonium salt of glyceryl borate and having at 200° F. a viscosity of three (3) to seven (7) inclusive as compared with the viscosity of ■ water of 60'° F. . . ■ .. Claim 4: “4. As an electrolyte- for an electrolytic cell, a liquid including a mixture — 176 — of glyceryl borate with an alkali salt of glyc-eryl borate and ¡having at 200° F. a viscosity of three (3) to seven (7) inclusive as compared with the viscosity of water of 60° F.” Claims 8, 10’, 11, 12, and 13 are for the method, and read as follows: Claim 8: “8. The method of preparing an electrolyte for electrolytic cells, which comprises admixing alkali metal salt having a boron acid radical with a polyhydroxy alcohol and boiling until the free water and some of the water split off by the reaction is vaporized, with a resultant mixture of glyc-eryl borate with an alkali metal salt of glyc-eryl borate. ■ Claim 10: “10. The method of preparing an electrolyte for electrolytic cells, which consists in admixing boric acid, glycerol and ammonia water and boiling the solution. Claim 11: “11. The method of preparing an electrolyte for electrolytic cells which comprises subjecting a mixture of boric acid and polyhydroxy alcohol in the presence of an alkali to boiling and continuing until the boiling point at atmospheric pressure is about ISO1 degrees centigrade. Claim 12: “12. The method of preparing an electrolyte for electrolytic cells which comprises subjecting a mixture of boric acid and polyhydroxy alcohol in the presence of an alkali to boiling and continuing until the boiling point at atmospheric pressure is about 130 degrees centigrade, and maintaining the solution at this temperature for at least five minutes. Claim 13: “13. The method of preparing an electrolyte for electrolytic cells, which consist in admixing glycerol and boric acid with ammonia water in the proportion of 100 to 62 to 5, heating it up' to 130 degrees centigrade, and then boiling said mixture for at least five more minutes at atmospheric pressure.” The three Mallory electrolytes are described in paragraph 8 of Exhibit 1. I will refer to them as the Mallory A, B, and C electrolytes, in the order in which they are described in that exhibit. All three of them contain ethylene glycol. , None of the plaintiffs’ condensers contains any glyceryl borate or any ammonium salt of glyceryl borate, because there is no glycerol compound present. Glycol borate is not referred to in any of the claims in suit. Claims 1, 2, 3, 4, 8, 10-, and' 13 all call for glycerol or glyceryl borate in some form, and, as none is found in the Mallory electrolytes, they do not infringe. Claims 8, 11, 12, and 13 reeite polyhydroxy alcohol, but the specifications and Claims are to be considered together and in such, a way as to give a consistent meaning to each of the words and phrases used, and, so considered, none of the Mallory electrolytes infringe. The defendant seeks to show infringement of the method claims 8, 10, 11, 12, and 13 by ethylene glycol electrolytes, by invoking the doctrine of equivalents. This it seems to me cannot be done, as the effect of it would be to write a new set of claims for the patent. The patent was not a pioneer, and has not been used by the defendant since May or June, 1932, at which time the defendant abandoned the glycerol electrolyte in favor of the ethylene glycol electrolyte. Comparison must be had with the electrolyte of the patent in suit, and not with defendant’s commercial structure. As I have hereinbefore shown, Georgiev was not the inventor of an ethylene glycol electrolyte, nor a glucose electrolyte, and the most that can be said as to Georgiev was that as to them he expressed a hope, which was really not his but Flrankel’s, but such hope cannot deprive the plaintiffs of the invention of Ruben, which was claimed in his application filed about six months before Georgiev filed his application. A resort to the doetrine of equivalents would do violence to the well-settled rules for the interpretation of a patent, which is to be interpreted in accordance with the general rules governing the interpretation of contracts. I. T. S. Rubber Co. v. Essex Rubber Co. (D. C.) 279 F. 593, affirmed 272 U. S. 429, 47 S. Ct. 136, 71 L. Ed. 335. The Georgiev patent says that, when glycerin, glycol, and glucose are intended, the term “polyhydroxy alcohol” is used. Therefore, when the term “glycerol” is used, it and not ethylene glycol is intended. The term “polyhydroxy alcohol” cannot be made to cover ethylene glycol without also covering glucose. Defendant contends that the patent should be construed in such a way that the term “glyceryl borate” means just what it says in claim 9, but in claim 8 means a borate of any polyhydroxy alcohol. Such an interpretation would be inconsistent as between claims 8 and 9, and also between the specification and claim 8. The specification says that the — 177 — term “polyhydroxy alcohol” is used when the patentee intends to refer to the class of alcohols having two or more hydroxyl radicals. Defendant’s contention, “substantially the only difference between claim 8 and claim 9 is that the latter specifically recites glycerol as an ingredient, while the former more broadly recites polyhydroxy alcohol,” is inaccurate. Claim 8 specifies an alkali metal salt, whereas claim 9' specifies ammonium borate. Claim 8 therefore includes the salts of sodium, potassium, lithium, rubidium, eassium, and, by analogy, ammonium, which, though it is not a metal, is commonly classed in its chemical reaction as an alkali metal. Claim 8 is therefore much broader than claim 9, not only in its recitation of polyhydroxy alcohol as a starting material, but also in its recitation of the entire class of alkali metal salts. Automatic Recording Safe Co. v. Burns Co. (C. C. A.) 281 F. 985, 987, cited by the defendant, seems to have been misconstrued by it as applied to this case, as it seems to me that a correct application of the principle of that decision to the present case is that the phrases “polyhydroxy alcohol” and “glycerol” do not mean the same thing, and the pat-entee appreciates the distinction. Also, that the term “glyceryl borate” does' not mean the borate of any polyhydroxy alcohol, but is specific to the borate of glycerol. To construe it otherwise would do violence to the language of the specification and the claims of the Georgiev patent. The Mallory electrolytes B and C have vis-cosities which are less than 3 at 200'° F., as compared with the viseosity of water at 60'° F. As to Mallory electrolyte B, the defendant’s witness Waterman gave viscosities of 1.6, 1.75, and 1.75. Barsky, plaintiffs’ witness, gave 1.9, and Booe, called by the plaintiffs, testified to a value of 1.81 and 1.91 measured by a Stormer viscosimeter. This electrolyte is in the Mallory Nos. 4 and 5 condensers, Exhibits 10 and 11.' • As to Mallory electrolyte C, Waterman, the defendant’s witness, gave viscosities of 1.5.1.6, and 1.7. Barsky, plaintiffs’ witness, gave a value of 1.5. This electrolyte is in the Mallory Nos. 6, 7, and 8 condensers, Exhibits 12, 13, and 14. On all the evidence it thus appears that the viseosity of the Mallory B and C electrolytes is outside of the viscosities 3 to 7 specified and claimed by Georgiev. As to Mallory electrolyte A (which has been m use by the Mallory Company since February, 1931), there is some conflict in the evidence. At 200° F. this electrolyte contains crystals. Plaintiffs’ witness Barsky testified-that ho tried to determine the viseosity with the Scott viscosimeter, but without success, as the crystals plugged up the orifice before half the required amount of electrolyte had passed through, and that he attempted to estimate its viscosity by plotting the viscosity of a series of electrolytes similarly prepared, but containing smaller amounts of boric acid. By this method he estimated the viscosity at 2.1. The witness Booe, called by the plaintiffs, used a Stormer viscosimeter, an instrument that measures the viseosity by determining the drag exerted on a revolving cylinder, and has no orifice to become plugged up with crystals, and by it he determined the viseosity of electrolyte A as 2.73. The defendant’s witness Waterman testified that he tried to measure the viscosity of this electrolyte with a Scott viscosimeter. He said that the electrolyte contained crystals at 200° F., but did not know that this had any effect on the results. The values he obtained —3.7, 4.3, 3.5, 5.1 — are so irregular, that it seems clear to me that there must have been some cause for that irregularity, and I believe it to have been the crystals in the electrolyte. The measurements made by Booe are entitled to the greater weight and axe accepted by me as correct. The viseosity determination on the Scott viscosimeter is based on the number of seconds required for 56 e. e. of liquid to pass through a given orifice. It is therefore apparent that any obstruction of the orifice by crystals or anything else will require a greater number of seconds to allow the 50 e. c. of liquid to pass through, and the viseosity will appear to be higher than it is in fact. Convinced as I am that, in the Waterman readings, the condition of the orifice was different on each reading, due to. crystals in the electrolyte, I can find no basis on which to apply a correction on each of those readings, and therefore I must find that they are not accurate, but .high, in an amount that can not be determined. The patent calls for an electrolyte having a viseosity between 3 and 7 at 200° F., as compared with the viscosity of water at 60° F., but the witness Waterman, in making his measurements, which were reported to Mr. Plirseh by letter of October 31,1933, did not make his comparison with the viseosity of water at 60° F., but compared it with the viseosi — 178 — ty of water at room temperature, without making any record of the room temperature at the time. The viscosity of water at 60° F. is greater than its specific viscosity at higher temperature, the difference for the 10° between 60° F. and 70'° F. being in the neighborhood of 10 to- 13 per cent. This is not the standard described in the Georgiev claims, but one which would result in making the viscosity of the Mallory electrolyte appear .higher than it would be if the comparison had been made with the prescribed standard. The Mallory A electrolyte is in Mallory Nos. -1, 2, 3, 9, and 10' condensers, Exhibits 7, 8, and 9, and Exhibits P-2 and Q-2, respectively. The defendant has failed to bear the burden of establishing infringement by showing that Mallory A, B, and .C electrolytes have viscosities within the range specified in the claims. The evidence is uncontradicted that Mallory B and C electrolytes have viscosities outside the range specified in the claims. The evidence offeree! on behalf of the defendant to show that the viscosity of the Mallory A electrolyte was within the range specified in the claims is inaccurate and therefore without probative force, but the evidence offered by plaintiffs of the determination of the witness Booe on the Stormer viscosimeter, corroborated by the estimates of the witness Bar-sky, establishes the fact that the Mallory A electrolyte has a viscosity below the range .of 3 to 7, specified and claimed by Georgiev. Therefore, claims 1, 2, 3, and 4 are not infringed. Claims 1, 2, 3, and.4 of the Georgiev electrolyte patart in suit each- call for a liquid electrolyte, whereas in Mallory there is a paste electrolyte, and this obvious distinction is not avoided by the specious argument of defendant’s counsel, that the paste is a liquid at 209° F., because what the patent specifies is that the electrolyte is normally an oleaginous liquid, and is preferably used with enough added water to bring its viscosity down. The ordinary reading of the claims, especially in the light of the specification, calls for an electrolyte which is liquid at normal temperature, and which has a specified viscqsity at 200° F. Mallory, with its paste electrolyte, does not infringe claims 1, 2, 3, and 4 of the patent in suit which call for a liquid electrolyte. The admission of defendant’s counsel, that the Mallory B and C electrolytes, which are somewhat more fluid than the Mallory' A electrolytes, do not infringe claims 1, 2, 3, and 4, because they are too liquid, that is, they have too low a viscosity at 200° F., makes it impossible for me to understand his argument that the Mallory A poste electrolyte infringes the claims for a liquid electrolyte. For the reasons last stated, claims 1, 2, 3, and 4 are not infringed. Claims 11 and 12 call for a boiling point of 130° C., and claim 13 for a heating point of 139°. C., and that is not met by a boiling point of 129° C.. or 126° C., in view of the record in this case, which shows that the formation of glyceryl borate and an alkali salt of g’lyceryl borate takes place when borax and glycerol are mixed at room temperatures, and that subsequent heating or boiling simply drives off more water. Inventive novelty, if there be any (which I do not find) in boiling until a temperature of 130° C. is reached, cannot extend to all boiling-, nor can it extend to any boiling substantially less than 130° C. Claim 13 was not in issue in Aerovox Corporation v. Concourse Electric Co. (C. C. A.) 65 F. (2d) 386, 389, 390, but claims 11 and 12 were, and the temperature of 130° C. was regarded as an essential limitation. On the record in this case, it is even more plainly shown that, if claims 11, 12, and 13, which contain the same limitation, have any validity, it depends on the temperature specified. The Ma.llory electrolytes, which go only to temperatures of 120° and 125° C., do not infringe any of these claims. The Mallory electrolytes do not infringe .any of the claims in suit of the Georgiev electrolyte patent, No. 1,815,768. The Georgiev patent in suit, No. 1,789,-949, for electrolytic cell, herein ealled the structure patent, discloses a dry electrolytic condenser, in which the dielectric is formed in a thin film upon the surface of the anode by electro chemical action, the dry electrolytic condenser roll is placed loosely in an aluminum can with a cathode tab of the condenser roll connected both electrically and mechanically to the can and to an outside terminal lug, and the anode tab connected to an aluminum terminal post in an insulating cover, which closes the open end of the can. The patentee in the patent said: “An- > other feature is the use of the identical metal, preferably aluminum for the cathode, the anode, the can, the rivets and any and all other metal elements in- the completed structure, thus precluding galvanic action.” This was not a discovery of Georgiev’s, as it was’ at the time good engineering to know — 179 — that dissimilar metals in contact with one another are likely to give rise to corrosion if they come in contact with anything containing salts in solution, and this knowledge went back to Volta’s day. It was well-known engineering practice to connect one plate of the condenser to the can to equalize the potential. The practice of the wax condenser art was followed in early dry electrolytic condensers. A vacuum tank was used for impregnation, the impregnation unit was encased in a tin can, and all voids were filled with pitch or wax. The patentee of the patent in suit in his specification said: “By the elimination of paraffin, wax or pitch, there is avoided, on the one hand, any injury to the electrolytic film, due to the heat of such composition, when poured in, and, on the other hand, any heat generated in use of the condenser is more readily dissipated due to the absence from the structure of any such poor heat conductors as paper, oil-cloth, wax or pitch. Substantial voids between the condenser roll and can are substantially devoid of filling composition, thereby to facilitate convection currents of air and heat dissipation.” The Circuit Court of Appeals of this circuit in Aerovox Corporation v. Concourse Electric Co., 65 F.(2d) 386, 387, said with reference to the prior art: “The art of dry condensers had before the application been familiar with a gauze between the electrodes soaked like a sponge in the electrolyte; this would cause the film to be replaced, as it wore out, or was ruptured. To insure against leakage of the electrolyte the condenser was covered with pitch or wax, filling the spaces between the sides of the box and the condenser proper. The electrodes with intervening gauze were of* large area and were rolled upon themselves into a cartridge, one terminal being fastened to the edge of the anode and the other to that of the cathode, to which conducting wires might be attached. Thus the box, sometimes of metal, sometimes not, was insulated from the condenser and the electrolyte had no means of escape. In operation heat is generated by electrolysis and with it gas. As a result the gas forms a path through the pitch, or wax, and the electrolyte follows through the passages so made. The connection so established by the electrolyte causes a galvanie action between the metal parts, electrodes and box, if these are of different metals, and a loss of current between the cathode and the box when that is of any metal. One or both - of these difficulties had existed in all dry condensers before Georgievr whenever the box was of metal.” The invention of the structure patent is defined as follows, by the Circuit Court of Appeals of this circuit, in Aerovox v. Concourse, supra: “He proceeded .upon a new theory. His condenser was, indeed, made up of the same parts as those which preceded,, but instead of filling the spaces between the cartridge and the walls of the box with pitch or wax, he left them open, thus allowing any leakage of electrolyte to drain to the bottom and any gas produced to escape. To neutralize corrosion by galvanie action and loss of current, he directly connected the cathode with the box, which he made of the same metal, aluminum. The result was that there could never be any difference in potential between them and no current would leak, the negative terminal being on the box itself.” The importance of having all parts which dipped into the electrolyte of the same metal, and aluminum at that, during the forming process, had been disclosed, but the condenser must be made, so that this process shall continue during operation. It was an obvious means of securing equal potential between the cathode and the wall, and avoiding current losses, to connect the cathode to the wall. After stating those facts, the Circuit Court of Appeals said: “We do not therefore rest upon these. * * * But we think that they were mere incidents of a new, a quite new, idea for such a condenser; that is, not to insulate it at all, but to allow it to be set loosely in the box. This had theretofore been supposed to be objectionable for the reasons we have given. Georgiev thought otherr wise and abandoned the uniform course of the ext.” The only alleged prior art not in the record in Aerovox v. Concourse, supra, and offered in evidence in this case, are British patent 3STo. 314,160, issued to the Telegraph Condenser Company, Limited, and William Joseph Cole, for improvements in or relating to electrolytic condensers, completely accepted June 27, 1909', bn an application filed April 18, 1923; United States patents Ho. 1,726,-744 to Sullivan, Hx 1,891,345 to Davidson, Ho. 1,912,223 to Ruben, and Ruben specification of March 29,1929. The British patent states: “According to this invention, an electrolytic condenser is produced by interleaving with one or more metal sheets a corresponding number of sheets of absorbent fibrous material, rolling the whole spirally to form a rolled condenser, — 180 — and then'impregnating the fibrous material with a suitable electrolyte. The whole is preferably fitted into a suitable container which may form or be connected with one of the electrodes of the condenser." It describes the making of terminal tabs by slitting the electrode foils and folding the cutting portion outward at right angles analogously to the transverse slitting of the foil described by Georgiev. It then states that the rolled condenser unit may be placed in a container and suitably spaced from the sides thereof, the intervening space being then filled by hot electrolyte which solidifies on cooling, or by an absorbent material such as cotton wool or the like. Two connecting terminals are provided, one of which is insulated from the casei and the insulated terminal is made of aluminum, as regards the part which lies inside the case. This is no better as a reference than Ruben patent, No. 1,714,191, and what the Circuit Court of Appeals in this circuit says in Aerovox v. Concourse, supra, about that patent is applicable to the British patent in the instant suit. In the British patent all the parts exposed to electrolyte are not made of aluminum, nor is the can or the terminals necessarily made of aluminum. No voids in the unit are clearly shown. The manner in which the terminals from the condensers are connected to the tabs from the sections is set forth, in detail, but there are specified the use of connecting members, which are not made of aluminum. The permissive disclosure of that patent, about pouring electrolyte into the container around the outside of the condenser section, relates to the wet electrolytic condenser embodied in that patent. If the voids between the anode and the can are filled with electrolyte as proposed rather than with pitch, the result would be a substantial duplication of the structure of the wet electrolytic condenser of Mershom patent, No. 1,773',492. Although the Mershon patent is referred to in the record in Aerovox v. Concourse, supra, it was not commented on by the Circuit Court of Appeals, and I see no reason for comment here. The plaintiffs contend that in the said British patent “the electrolyte serves to establish electrical connection between the can and the cathode foil.” This, however, is not a bonding of the.cathode to effect grounding of the can; on the contrary, it increases the electrical leakage seriously, which is precluded in the Georgiev structure patent in suit. The British patentee showed no appreciation of the importance of keeping the electrolyte away from the nonaluminum parts. United States patent No. 1,723,744, issued to Raymond H. Sullivan, for condenser, dated September 3,19291, on an application filed November 17, 1926, is not for an electrolytic condenser, and does not present the problem of electrical leakage. It was introduced, I assume, to show a condenser in which there are provided means for grounding one element of the condenser upon the metal case in which it is inclosed. The application for patent No. 1,891,345, issued to John C. Davidson, for improvements in electrical condensers, granted December 20', 1932, on an application filed May 16, 1928', shows a dry electrolytic condenser in a metal container, spaced from the wall of the container by sand, gravel, pulverized cork, or the like, but as I view it, this inhibits convection currents of air, without preventing electrolyte seepage to the can and other non-aluminum metal parts that results in corrosion. The Ruben specification delivered to “Doc” Shoemaker under date of March 29', 1909', shows an experimental insulated dry electrolytic condenser, set in what is described as a “metal case,” but aluminum is not specified, with air spaces around the condenser section except at the top, which is sealed with black halowax, and with terminal wires or condenser leads described as of copper, both leads; being insulated from the can. Patent No. 1,912,223, issued to Ruben, for improvements in electric condensers, granted May SO1,1933, on an application filed May 15, 1928, shows a cylindrical condenser roll inserted loosely in a rectangular easing, without an, insulating material around the roll, and with spaces. Both this patent and the specification of March 219, 1929, have less pertinence than the Ruben patent, No. 1,714,191, which was disposed of in Aerovox v. Concourse, supra. Plaintiffs offered considerable evidence to show that the feature of the air spaces is devoid of utility, and in fact positively harmful, but I am not impressed with it. Plaintiffs also contend that, in its commercial structure, defendant does not in all respects fallow the teaching of the patent in suit, but that is not of importance, nor does it show lack of novelty or utility of the patent in suit, and comparison is to be made of the plaintiffs5 structure with the defendant’s — 181 — structure patent in suit, and not with, defendant’s commercial structure. The patent has been adjudicated, and the invention has been held to reside in not insulating the condenser at all, and in allowing it to set loosely in the box, or as was said in Aerovox v. Concourse, supra: “ * * * A cartridge (condenser roll), inserted from above, cannot touch the walls, and the space left gives opportunity for the escape of gases, and incidentally lets out any electrolyte that may leak. This was1 the essence of the invention.” It is true that the structure patent in suit has not supplanted all forms: of condensers, and that large quantities of other forms are being marketed, but the structure patent in suit has had considerable commercial success. The structure patent in suit of Georgiev, No. 1, 781),£49, claims 11,18, and 19', is valid. This brings us to the question of infringement. Claims 11, 18, and 19', of the Georgiev structure patent read as follows: “11. An electric condenser comprising a conductive container, a wound electrolytic condenser roll therein and extending longitudinally therec, , a terminal strip protruding from one erm,of said roll and mechanically joined direcjpLto the wall of said container, said eonta:', ^: having an electrical terminal cooperati • with said joint, a cover of insulating matei.al for said container, and a binding post through said cover and affixed to the other terminal of said condenser.” “18. An electrolytic condenser comprising anode and cathode foils of aluminum, fibrous sheet means between said foils and impregnated with electrolyte, a closely adhering film of dielectric on the anode area, a can of aluminum enclosing said condenser roll and presenting air spaces thereabout, an aluminum part unitary with said cathode foil and in, direct mechanical contact with the lateral wall of said can for terminal connection thereto, the anode having a terminal insulated from the can, the voids between the condenser roll and the can being substantially devoid of filling composition thereby to facilitate convection currents of air.” “19‘. An electrolytic condenser including an active condenser element having electrodes of identical metal, a dielectric film formed on one of said electrodes, an enclosing metallic can of the same metal as said electrodes, the wall of which determines air spaces about the active condenser element, a terminal for said latter electrode insulated from the can, a terminal for the other electrode directly affixed to said can mechanically and electrically, metallic connecting members electrically and mechanically associating said electrodes with their respective terminals, each of said metallic connecting members being of the same metal as said electrodes, the. voids between the active condenser element and the can being substantially devoid of filling composition, thereby to facilitate convection currents of air.” The Mallory condensers 2, 4, 9, and 10, which are alleged to infringe, are aluminum can condensers, and comprise two elongated sheets of aluminum, an anode and a cathode, with an interposed layer of gauze and a superposed layer of gauze. The anode has a dielectric film preformed thereon, appearing as a white coating. The gauze being coated with the plaintiffs’ electrolyte paste, is rolled together. The complete rolled unit is lodged in a cylindrical aluminum can, with spaces or air voids which are ample. The aluminum anode tab terminal of the unit integral with the anode foil is clamped under the head of the aluminum anode terminal screw, while the cathode terminal tab integral with the cathode foil is clamped against the wall of the aluminum can by the insulating cover which rests against the bead of the can and is held in place by rolling over the rim of the can against a rubber gasket, said rim also serving for a grounding terminal connection of the can to the metal mounting plate against which it is forced, by tightening of the mounting nut and lock washer. The Mallory condensers in question each present air spaces between the roll and the can; all metal parts thereof exposed within the can, including the can itself, are of aluminum; and the cathode tab is grounded to the can. This was admitted by plaintiffs’ expert. It seems to me that plaintiffs’ contention as to noninfringement of the structure patent in this ease is practically the same as that of the defendant in Aerovox v. Concourse, supra. I am not impressed, however, with the contention that results just as good, or even better could be accomplished with complete pitch filling of cans made of tin, and by the use of brass terminals insulated from the can. If this contention was correct, with all the prior aut on which to draw, why do plaintiffs find it necessary to imitate the Georgiev structure patent in suit? Every element of claim 11 of the patent in suit is found in Mallory condenser Nos. — 182 — 2,. 4, 9', and 10, subject to a trivial qualification; that is, plaintiffs argue that no condenser would infringe claim 11, unless its cathode tab were connected by rivets with a terminal lug at the exterior of the can, substantially .exactly as shown in the drawings of the Aerovox structure patent in suit. In the Mallory condenser the cathode tab is just as securely and effectively connected electrically and mechanically to the wall of the can, as if rivets were used for the purpose ; the rolled over rim of the can, by which the security of this contact is maintained, serves just as effectively for terminal grounding connection of the can as would a separate terminal lug. The Mallory construction accomplishes the same result in identically the manner recited in elaim 11. Plaintiffs’ argument of noninfringement is based on an attempt to read into the claim details not recited therein, but illustratively shown in the drawings of the patent. No such limitation is necessary to save the validity of the patent. If such an interpretation as plaintiffs contend for in this case be the proper one, then the Circuit Court of Appeals would have been precluded from finding infringement of the Aerovox condenser in Aerovox v. Concourse, supra. I find nothing requiring extended consideration in the file-wrapper estoppel argument against elaim 11, ■ as that claim distinguishes from the elaim which it replaced in the original application (original claim 12) which was of a scope to cover a construction in which the flimsy protruding tab of the cathode would be used for external connection. Claims 18 and 19' read on the Mallory condensers 2, 4, 91, and 10. Plaintiffs’ expert says, “There are some air spaces, no doubt about that,” but as to these claims plaintiffs advance an additional argument about air spaces. It is true that stagnant air is one of the poorest heat conductors known, and yet the comparative tests of Mr. Knowles prove that the difference in the heat dissipation between air-spaced and pitch-filled condensers is negligible. This paradox is explained by the existence of “convection currents of air,” as recited in claims 18 and 19. Regardless of the theoretical explanation of the paradox, Georgiev’s discovery that the pitch always used theretofore, with its cost and weight, and the mess, heat and smell of melting and pouring it in place to fill all voids-in the can, could be eliminated, constituted invention, even standing alone, but in addition is the fact that the air spaces will serve as a cushion to absorb and limit the rise of gas pressure during operation of the condenser, as compared to a structure, all of the voids of which are filled with pitch. The use of only enough pitch in the Mallory can to keep its contents from shaking does not relieve from infringement, as the claims do not recite the can as devoid of filling composition, but as “substantially devoid of filling composition.” The amount of pitch that is used affords the air spaces which give opportunity for the escape of gases during operation, as found in Aerovox v. Concourse, supra. The Mallory electrodes 2, 4, 9, and 10 infringe the Georgiev structure patent, No. 1,789,949, in suit. This leaves for consideration only the immunity granted by the defendant to the Mallory Company, by the letter of May 2, 1933. A previous suit between the parties was settled, and, as a part of the settlement of the controversy, the defendant gave to the plaintiff P'. R. Mallory & Co., Incorporated, a letter dated May 2, 1932!, addressed to the plaintiff P. R. Mallory & G®., Incorporated, which read as follows: “iMBonneetion with the settlement of the pending suit by Ruben Condenser Company and yourselves against us, in the United States District Court for the Eastern District of New York, for infringement of the Ruben patents Nos. 1,719,073 and 1,714,191, and in consideration of your executing the settlement papers which have been prepared, we hereby agree for ourselves and our successors and assigns, that we will not hereafter assert any elaim or institute or prosecute any suit against you, your customers or your present licensees, (1) for infringement of any United States patent now owned or •controlled by us, including specifically the Georgiev United States patents Nos. 1,789',-949 and 1,815',768, or which may hereafter issue upon any application now owned or controlled by us and pending in the United States Patent Office, or (2.) for infringement of any United States patents which may hereafter issue for any invention heretofore made and now owned or controlled by us or any of our employees, by reason of the manufacture, use or sale of condensers under the Ruben patents Nos. 1,710,073 and 1,714,191, in accordance with the practices followed by you or them at any time from January 1, 1929 to date, in the manufacture of such condensers as you or they have sold in substantial quantities.” — 183 — As I have found the Georgiev electrolyte patent No. 1,815,768 was not infringed by the plaintiffs, I will consider the foregoing immunity agreement solely with reference to the Georgiev structure patent, No. 1,789,949. The defendant conceded that Mallory No. 1 condenser is entitled to immunity, and, under the strictest interpretation of the agreed ment, Mallory No. 2 and No. 3 condensers are entitled to immunity, as substantial quantities of them were made by Mallory Company before May 2, 1932. In fact, Mallory No. 3 condenser is not an aluminum can condenser, but is in a bard-board container, and is not covered by the structure patent in suit. This is likewise true of Mallory No. 5 condenser, which is not an aluminum can, but is in a cardboard tube, and is not covered by the structure patent in suit. Defendant makes no charge of infringement of the structure patent by Mallory condensers 6, 7, and 8, but limits itself solely to Mallory condensers 2, 4, 9', and 10'. As I have found, Mallory No. 2 condenser is immune. ' This brings us to a consideration of what did the parties mean or understand by the words “in the manufacture of such condensers as you or they have sold in substantial quantities.” As we are dealing only with the structure patent, it is of no moment what were the elements of the electrolyte inclosed, so long as it was an ethylene glycol electrolyte and not a glycerol electrolyte, but we are concerned only with the structure. I need no evidence to" explain or make eledr any ambiguity, because I see none in the words used, and I have disregarded the testimony of the witnesses Mallory and Bristol., The motion of the defendant to strike out that testimony is denied, with an exception to the deféndant. ' Obviously what was intended was to grant immunity in the manufacture of the types of condensers covered by the structure patent, which they had manufactured in substantial quantities within the period named. Of course, they were not tied down to the exact sizes, capacities, and voltages, which of necessity would vary because of the differing conditions under which they were to be used, but they were limited to the types of condensers they had mahufaettired. Wide variances would deprive them cf immunity. Of course, immunity was intended to be given to the aluminum can structures, otherwise it would have been meaningless as to the structure patent, as the cardboard containers were not covered by [the structure patent. All of the structures of the Mallory condensers 2, 4, 91, and Í0 seem to me to be of the same type; all of tSem being aluminum cans but differing somewhat in size, capacity, and voltage. None of the electrolytes contained therein were glycerol electrolytes, but, on the contrary, all the electrolytes contained therein were ethylene glycol electrolytes; therefore, under the terms of the agreement in the form of the letter, supra, the said Mallory condensers 2, 4, 9', and 10 are all immune under the agreement in question. The plaintiffs are entitled to a decree against the defendant, holding claims 1, 5, 7, 9; 13, 16, 19, and 23 of the Ruben patent,' No. 1,891,207, valid and infringed, with in-' junction, damages, accounting, and costs; and the usual order of reference; and the defendant to a dismissal of the' cause of ae-: tion for unfair competition, on the merits, but without costs; and in favor of the plaintiffs dismissing the counterclaim as to the Georgiev electrolyte patent, No. "1,815,768, claims 1, 2, 3, 4, 8,'10, 11, 12, and 13 on the merits, and dismissing the counterclaim of unfair competition, on the merits, with costs against the defendant; and in favor of the defendant holding claims 11, 18, and 19' of the Georgiev structure patent, No. 1,789',949', valid and infringed, ■ but without injunction, damages, accounting, and costs;, and in fa? vor of the plaintiffs against the defendant granting to the plaintiffs immunity of, from, .and because of, past, present, and future infringement of said structure patent, Nx 1,-789,949, by the manufacture and/or sale of condensers of the'type of Mallory condensers 'Nos. 2,4, 9', and NX A decree may be entered in accordance with this opinion. Settle decree on notice. Submit proposed findings of fact and conclusions of law in accordance with this opinion, for the assistance of the court, as provided by rule 70% of the Equity Rules (28 DSC A § 723) and rule 11 of the Equity Rules of this court.