Dubilier Condenser & Radio Corp. v. Aerovox Wireless Corp.

MACK, Circuit Judge.

Appeal from a decree dismissing a bill for infringement of claims 3 and 8 of United States letters patent No. 1,497,095, issued June 10, 1924, to William Dubilier, and assigned by him to plaintiff, because of invalidity of the claims, and finding, moreover, noninfringement by one of defendant’s devices, type 1350. The District Judge was of the opinion that the other type, No. 1250, would be an infringement, if the claims were valid. The issues of validity and infringement are thus before us on this appeal.

In the early days of radio broadcasting, beginning about 1920, many radio receiving sets were assembled by amateurs, using condensers, transformers, and other individual parts purchased separately; other sets were assembled by manufacturers from separately purchased parts and sold as complete units. With the rapid growth of the business, there grew up from both sources a demand for a fixed condenser of small capacity, ranging from about .0001 to .006 microfarads, which was required for a variety of purposes, unnecessary to be detailed here, in the standard circuits. The device, known as a receiving set condenser, consists principally of tin foil sheets, which serve as the plates or armatures, interleaved with thin mica sheets, which constitute the dielectric or insulating material; the sheets are built up into a stack.

Because of the relatively small capacity required, these stacks are ordinarily quite thin. The amount of electrical energy which a condenser will store, called its capacity and measured in microfarads, depends primarily upon the juxtaposed area of the tin foil plates, the spacing of such plates from each other, and the nature of the insulating medium or dieléetrie interposed between them. If these factors are constant, the capacity of the condenser will remain constant; if the distance between the plates be varied even microscopically, or if a foreign substance, such as moisture or air, gets into the dielectric between the plates, the capacity will vary as much as 40 per cent. Consequently the electrical efficiency of the condenser depends upon the maintenance of the initial spacing and the exclusion of foreign matter.

Mica has long been known as one of the best solid dielectric materials, obtainable in very thin sheets, and is universally used. In ¡order to exclude moisture or air from between the laminas of the condenser, it has been the custom for many years to impregnate the sheets with an insulating compound, which became part of the dielectric. Varnish has frequently been used for this purpose, as it yields a rigid stack after drying; but, where a condenser of low electrical losses is desired, varnish is not suitable because of its poor di.eleetric qualities; therefore paraffin wax is commonly used as an impregnating material, even though it does not impart the same rigidity as varnish. To secure and maintain the rigidity essential both to hold the condenser stack against even microscopic internal movement, and for other purposes, some form of pressure is usually applied; this may be done in the course of the baking during manufacture, or by mechanical means thereafter, or by both.

In addition to the requirement of permanence in capacity and low electrical losses, a receiving set condenser must be small, light weight, and capable of being readily mounted and connected in a circuit. Such condensers are installed between larger instruments, being suspended between wires, or hold against a panel or the terminal posts of other instruments by screws. It is important that the condenser be capable of being interchangeably mounted in any one of the vari*658ous locations in which it might be needed in any particular type of set. Moreover, since electrical connections are usually made by soldering wires or screws to the condenser terminals, and since heat is generated by the vacuum tubes, the condenser must be designed so as not to be affected by this heat; and, for commercial success, it should, of course, be marketable at a low price — an experienced witness said 25 to 30 cents.

The Dubilier patent in suit, No. 1,497,095, does not purport to include, and indeed it expressly disclaims, the structure of the condenser stack as part of the invention; it merely recommends that the condenser be “compressed, evacuated of all air between the plates, and impregnated with melted paraffin, as may be required,” referring, by way of example, to a prior Dubilier patent purporting to cover this impregnating process. The patent ia essentially a structural patent, relating to improvements independent of the stack construction, and specifies a condenser “provided with means whieh will keep the elements constituting the stack or body of the condenser pressed together with firmness, and prevent any loosening or separation of the plates, by whieh the performance of the condenser would be seriously impaired.”

In order to secure this desired pressure, there are applied to opposite faces of the stack, parts whieh the claims define as “relatively heavy and stiff insulating bearing members” (claim 3), and “relatively stiff and thick bearing plates of insulating material disposed on opposite sides of the stack, to build up the condenser to substantial thickness and rigidity (claim 8)”; bakelite being recommended for this purpose. To apply mechanical pressure to these insulated bearing members, the patent prescribes the use of hollow rivets or tubular elements, whieh pass through the stack and insulating bearing members. These rivets are also described as making contact with the two sets of tin foil plates of different polarity, so as to permit terminal connections to be made directly through the rivets. The patent further specifies the interposition of stiff clips or plates between the heads of the hollow rivets and the insulating bearing members, whieh clips are supposed to exert pressure on the bearing members, similar to that exerted by the head of an ordinary carpet taek, over the juxtaposed foil in that portion of the stack known as the “active area.”

The resulting condenser very thoroughly met the requirements of the trgdo; it possessed a constant capacity, a high electrical efficiency, was compact, could be readily and interchangeably mounted in any of the standard sets, and could be supplied within tho price range which satisfied the demand. The record discloses that it enjoyed a huge commercial success, and practically became a trade standard.

The claims relied on are as follows:

“3. An electrical condenser comprising a stack of conducting and insulating sheets, relatively heavy and stiff insulating bearing members overlying opposite faces of said stack, members of relatively stiff material at the opposite ends of said condenser and its stack, said members being of substantially the same width as the width of the condenser, with their opposing edges separated by only a narrow space adjacent the middle of the condenser, whereby uniform pressure is exerted over substantially the entire area of the condenser, each of said last named members having a portion engaging one of the bearing members, and a portion extending over substantially into engagement with the other bearing member, tubular elements passing through said stack, said bearing members and said last named members adjacent the opposite ends of the condenser and having parts for causing said bearing members and said members engaging tho bearing members to be secured together to clamp the active area of the sheets making up the stack, said tubular members being electrically connected with conducting sheets of different polarities to serve as terminal connectors and to enable conducting members to be inserted into such tubular members to make electrical connection thereto.”
“8. An electrical condenser comprising conducting plates and insulating sheets interleaved to form a stack, relatively thick and stiff bearing plates of insulating material disposed on opposite sides of the stack to build up the condenser to substantial thickness and rigidity, and clamping members extending through said stack and bearing plates, to exert compression upon the stack through said bearing plates, said clamping members comprising rivets headed over to exert compression as aforesaid and said clamping members being hollow and being respectively in circuit with conducting plates of different polarity to enable the members to serve as terminals.”

Claim 8 differs from claim 3, in that it does not require the use of additional clamps, but specifies that the hollow rivets directly engage the insulated bearing members.

1. When reduced to its essentials, the Dubilier patent purported to add to the existent *659art the following: Heavy insulated bearing members on opposite sides of the condenser staek; hollow rivets penetrating both bearing members and stack at points well toward the center of the staek, and as near as possible to the active area thereof, contact existing between the plates and such rivets; and (in claim 3) clamps of stiff material inserted between the head of the rivets and the bearing members, so as better to distribute pressure on the latter and on the stack. Plaintiff concedes that Dubilier was not the first to appreciate the need of applying distributed mechanical pressure to a condenser staek, and admits that many of the individual elements were old in the art. It contends, however, that the patent contributed to the art a new combination of these old elements, which eoacted in such manner as to satisfy both the technical and commercial requirements of a receiving set condenser.

With this contention we cannot agree, and we are of the opinion that the District Court was right in holding the claims in suit invalid for anticipation. The problem here turns chiefly upon whether or not the art of making radio condensers was entirely new, or may be considered a development of the art of manufacturing magneto condensers, which, save for the matter of size and resulting capacity, are almost identical with the former.

Plaintiff insists that, although magneto condensers comprise a staek of mica and tin foil, usually impregnated and clamped together, they cannot be considered part of the prior art here, because varnish was used in their manufacture, the staek was compressed while this varnish was still fluid, and then the whole was baked in an oven. It is argued that these condensers rely on pressure applied during manufacture, as contrasted with pressure during use, which plaintiff secures by means of bearing members, metal plates, and rivets. The answer to this contention is to be found in the type 22 magneto condenser, manufactured in 1915 by the Robert Bosch Magneto Company and sold throughout the United States. This condenser, except in one respect, is almost a complete anticipation of Dubilier-; it shows a laminated stack of tin foil and mica; at opposite ends of the stack are placed from five to ten extra layers of mica, which, when varnished and pressed together, are practically equivalent to plaintiff’s bearing plates; and, finally, encircling the staek and extending over a considerable portion of the two flat surfaces arc to be found metal bands, through which run solid and hollow rivets, binding the whole condenser together.

This Bosch product differs from Dubilier condenser only in that the rivets are disposed well toward the edges on a thick staek, and the bearing members are composed of laminated multiple layers of mica solidified by baked varnish, rather than bakelite. A patent monopoly cannot be founded on the mere moving of the rivets for the purpose of obtaining what is said to be more essential in a radio than in a magneto condenser, a better distribution of pressure over the “active area,” and the substitution of a comparatively new and better material in the bearing members.

Turning directly to the prior art of radio condensers, we find that British patent No. 23,821, to Siemens, accepted March 24.1910, discloses the use of pressure plates, clearly the equivalent of Dubilier’s bearing members, held together by a C-spring. These pressure plates “are sufficiently thick to withstand” the pressure of the springs “without distortion,” and permit the pressure to be “evenly distributed over the whole area of the condenser plate.” There is nothing in the patent to support plaintiff’s contention that theso bearing plates were to be of metal, a conducting rather than an insulating medium, and in our judgment Siemens thus clearly disclosed the use of pressure plates or “bearing members.” United States patent No. 1,440,142, issued December 26, 1922, to Euegel, shows a condenser, not limited to magnetos, in which the staek is bound together by solid rivets, disposed well toward the center of the staek, in much the same position as Dubilier’s hollow rivets. While no stiffening plates are shown, metal plates extending over more than two-thirds of the stack are shown between the rivets and the staek.

In reaching its finding of invalidity, the District Court relied principally on United States patent No. 1,181,623, issued May 2, 1916, to Yan Devanter. This patent disclosed the use of insulating cover plates heavy and stiff enough to perform the function of distributing sufficient pressure to insure constancy of electrical capacity on the stack, the pressure being exerted by means of metal and clamps which also served, as in Dubilier, as terminals. Indeed, the present plaintiff, which manufactures under the Dubilier patent in suit, was held to infringe this very Yan Devanter patent in Splitdorf Electrical Co. v. Dubilier Condenser & Radio Corporation (D. C. S. D. New York, April 18, 1928) 37 F.(2d) 745. While it is true that Yan Devanter did not specify the use of bakelite for his stiffening end plates, the substitution of this material clearly did not constitute inven*660tion. Moreover, the metal plates overlaying the bearing members, within claim 3 of the patent in suit, are clearly disclosed by Yan Devanter, and have been held to constitute an infringement of that patent.

It is extremely difficult from the record before us to ascertain the extent and importance of the pressure which is supposed to be exerted on the impregnated stack. Nevertheless,it is apparent that this pressure, such as there is, is exerted through the mechanically fastened bearing members. These' pressure plates or bearing members are the principal feature of the device, and were, as we have seen, old in the art. That they were known as early as 1910 is shown by Siemens; that a newly developed insulating material, bakelite, was substituted for the laminated mica plates shown by Bosch and Yan Devanter, is not sufficient for validity. Neither is the substitution of hollow rivets, for the solid rivets shown by Fuegel or the C-elamps shown by Siemens, of patentable significance, in view of the use of similar hollow rivets in the Bosch magneto condenser.

Ultimately the question resolves itself into a determination of whether or not, in view of the state of the prior art, the condenser evolved by Dubilier embodied merely mechanical adaptations or constituted a sufficiently new combination to merit patent protection. As cogent evidence that the Dubilier device was a new and valuable contribution to the art, plaintiff has shown that the condensers manufactured under the patent in suit rapidly achieved a wide market and were extensively imitated, and in some four instances competitors accepted consent decrees and were licensed under the patent. While commercial success often has an important bearing on the question of invention, it is far from conclusive. Dilg v. George Borgfeldt & Co. (C. C. A.) 189 F. 588, 591.

The District Court attributed much of the success of plaintiff’s condenser to the teaching of the Yan Devanter patent, which it infringed. Not only is this true, but the phenomenal growth of the radio industry, coincidental with the marketing of plaintiff’s device, certainly contributed largely to this result. Accordingly, mueh of plaintiff’s success was, in our judgment, circumstantial, and not dependent upon the teaching of the Dubilier patent. Moreover, the licenses and consent deerees are not convincing. The majority of these acquiescing competitors were relatively small concerns. One of them is no longer in the condenser business; the license to another has the usual provision that payments of royalties shall cease if the patent is held invalid. As pointed out in John E. Thropp’s Sons’ Co. v. Seiberling, 264 U. S. 320, 329, 330, 44 S. Ct. 346, 68 L. Ed. 708, the purchase of peace, when it is known that the patentee or assignee will have to proceed against a large competitor, is often a wise course for the small manufacturer. In these circumstances, we cannot predicate validity of the Dubilier patent upon mere commercial success. See Harvey Hubbell, Inc., v. General Electric Co. (C. C. A.) 267 F. 564; Wire Wheel Corp. v. C. T. Silver, Inc. (D. C.) 266 F. 221, 229.

Inasmuch as we are of the opinion that the District Court was right in dismissing the bill for invalidity of the claims in suit, there is no occasion to discuss infringement.

Decree affirmed.