Appeal from the District Court of the United States for the District of Connecticut.
Before MANTON, L. HAND, and SWAN, Circuit Judges.
This is a suit in equity for the alleged infringement of claims thirteen and fourteen of Solenberger's patent (1,670,082), for improvements in piston rings to be used in internal combustion engines. The piston of such an engine has a clearance between itself and the cylinder, which must in some way be filled to avoid the escape of oil into the combustion chamber, and to make a tight seal against the escape of the exploding gases. Moreover, as the top of the cylinder heats, its bore somewhat enlarges, and the ring must be able to expand, contracting again as it descends; this is called, "breathing." It had long been the practice, both in motor cars and stationary engines, to make the seal by a split elastic ring, fitting in a horizontal groove of the piston, and by its spring-like action expanding to take up the clearance. The outside of the ring was thus in direct contact with the inside of the cylinder, save as an extremely thin film of oil might separate them; and yet the friction was reduced to a minimum, for the ring was short axially. The seal depended upon the inherent expansion of the ring alone, which, being split, could be forced into the cylinder groove, and then spread itself. It had also long been the practice at times to set an inner ring, or "expander," between the ring and the base of the groove; ordinarily a piece of corrugated metal ribbon, bearing alternately, first on the bottom of the groove, and next on the inside of the ring. This was compressed when the ring was compressed, and its resulting potential energy reinforced the tension of the ring.
All this being old, the patentee's room for invention was correspondingly narrow. His disclosure is of two independent ideas, each embodied in separate claims, not necessarily used together. The first was to prevent the "pumping" of oil into the combustion chamber. If the ring be solid, except for a gap, it cannot be safely fitted so tightly into the piston groove that there will not be some vertical play. Continued use enlarges this, and makes the ring "pump" oil into the combustion chamber. To secure a tight fit the patentee prescribed that his ring should have two or more rows of horizontal slots, staggered, so that the metal separating the ends of two slots in one row should be above or below an open slot of the row below or above it. Thus the ring as a whole could be vertically compressed, and would hold itself elastically between the upper and lower edges of the groove, though not too tightly to prevent "breathing."
The slots had another purpose. The ring might be made thin radially, and the groove deep enough to allow an "expander" of the kind described to be used. This left alternate triangular spaces between the inner side of the ring and the outer side of the "expander," which allowed the circulation of oil to and from the cylinder. The suggested purpose of these was to reinforce the resistance of the ring to vertical compression, by the inertia of the oil filling the slots. Although the only other purpose ascribed to the slots in the specifications, was to increase the radial flexibility of the ring (page 3, lines 50-54), upon the trial the plaintiff's witnesses professed to find still more; the spaces behind the ring were said to form a lubricating trap or reservoir to distribute oil between the cylinder and the ring. Again, the "expander" acted as a baffle plate for the oil and decreased the "slap" of the piston. The second is probably true and may contribute to the plaintiff's success; perhaps the third is true too, though the conclusion is somewhat speculative. We may for argument accept both.
The specifications contained another invention, and it is this upon which the suit turns. Cylinders get out of true with use; there is always a lateral thrust, due to the revolution of the lower end of the connecting rod, and in time this affects the bore; at places its section becomes ellipsoid. To prevent leaks it is therefore necessary either that the cylinder be reground, or that the rings shall take up the added space left by abraded metal. This the ring may in part do by its own elasticity, especially when aided by an "expander"; but unless it be exceedingly flexible, the expansion will be uniform over the whole circumference, and if the ring fills the larger axis of the deformed cylinder, it will bear too heavily upon the shorter, thus setting up undue friction, and in the end scoring the metal. This the patentee sought to relieve by weakening the ring so much that it could no longer rely upon its own elasticity, but must be reinforced by an "expander"; which, bearing at separate points as it does, distorts the ring out of circular form only where necessary, without setting up too much friction elsewhere. The ring was made weak, both by reducing its radial thickness to less than three per cent. of its diameter and by the slots. Such a slotted flexible ring backed by an "expander," was the invention of claims thirteen and fourteen; it has nothing to do with axial compressibility.
The Ramsey Accessories Manufacturing Company, which conducted the defense, and which we shall speak of as the defendant, is the manufacturer of piston rings of various kinds, among which it sells a slotted flexible ring, used with an "expander." This has a single row of six slots, which lead from the cylinder to the spaces between the inside of the ring and the "expander." At the bottom of the ring groove is a hole through which the oil drains back into the crank-case, the space between the ring and "expander" never acting as a reservoir. However, the defendant's ring is flexible within the requirements of the disclosure, being below the prescribed limit of thinness, and having slots which weaken it. Moreover, contrary to the defendant's assertion, it is never sold or intended to be sold without an "expander"; obviously its inherent tension is too weak to serve alone, certainly in a worn cylinder. The defendant's advertisements declare, for example, that the "outer ring is made of soft gray iron without tension" (the last two words are italicized in the original) -- "It flexes readily to irregular cylinder shapes and seals the cylinder at every point. * * * The famous ramco Cushion Inner Ring used in this combination gives the outer ring the exact amount of tension needed to seal the cylinder without dragging."
There had been thin piston rings before, used with an "expander," better to conform to the cylinder; such were used by the Marmon and Overland Motor Companies. But these were unslotted. Again, there had been slotted rings used with an "expander." The best example of these is the Perfect Circle ring, slotted, but thicker than the prescribed limits of the patent, and sold for use with and without an "expander," presupposing therefore that it was elastic enough alone. Another was the Wel-Ever ring, which had a series of notches in the lower edge designed to allow the entrance of oil. A third was the De Luxe, like the Wel-Ever, except that the notches were replaced by a single row of holes. These were also too thick, and were sold with and without expanders. They too drained off the oil into the crank-case, like the defendant's. There were many patents for piston rings, some with corrugated spring "expanders" -- this going back to 1872 -- others slotted. So far, however, as the art shows, either in the office or outside, there had not, before Solenberger's application, been any slotted ring, purposely made too flexible to serve alone, and supplemented by an "expander." After its appearance the device went at once into very extensive use, more than two million a year being sold for several years before the trial. It was primarily meant for use in worn cylinders, and had proved itself most serviceable for that purpose.
On the first hearing the judge held that the claims were valid and infringed, but after the interlocutory decree was entered and an appeal taken, the defendant brought forward new evidence upon which another judge came to a different result. He found that the element of axial compressibility was implied in all the claims, which the defendant's ring concededly did not have. Further, that the combination was an aggregation; that is, that the slotting of the rings added nothing functionally to the thin ring reinforced by an "expander." Finally, that the new prior uses -- the Perfect Circle, Wel-Ever and De Luxe rings -- were good anticipations. The claims being invalid, he dismissed the bill. This appeal followed.
The claims in suit are certainly not limited to axially compressible rings. Other claims contain that feature and these, with companions not in suit, do not. The specifications make it clear that the patentee had two entirely independent ideas; first, an elastic seal with the edges of the groove, accomplished by axial compressibility, useful in new as well as old cylinders; second, extremely flexible rings, backed by "expanders," primarily important for worn cylinders. The latter part of the specifications (page 3, lines 30-90), is concerned with the second alone; it contains no suggestion of axial compressibility except the phrase "peculiar slotted construction" (page 3, line 51), which, though the defendant repeatedly presses it as incorporating several rows of staggered circumferential slots, we do not read so narrowly. The same distinction is also plainly foreshadowed in the statement of the objects of the invention (page 1, lines 73-90). Axial compressibility has nothing to do with the second invention, though the slots have; not only do hey serve to weaken the elasticity of the ring by removing the metal, but they probably also have an effect upon the lubrication by making possible an oil trap directly leading to the cylinder. Thus, we cannot accept the first ground of the decision below.
Nor can we the second; that is, that the patent was for an "aggregation." As we said in Sachs v. Hartford Electric Supply Co. (C.C.A.) 47 F.2d 743, 748, we do not understand that that word implies mechanical interaction of the parts, but only that all the elements must unite to realize a single purpose, a question not always easy to answer, perhaps in the end not distinguishable from whether the juxtaposition was fruitful as a whole. But quite aside from that, if here the slots were appropriate to produce a practicable piston ring for worn cylinders, their addition to the Marmon and Overland rings was "functional," and cannot be said to be "aggregation," whatever that catch-word may be thought to cover. The purpose being to make such a ring, it is only necessary that this feature should be, not necessary, but auxiliary, to its satisfaction; it was such, both as it weakened the ring, and as it was part of the oil trap.
The last question is of validity, which as usual, runs into that of infringement. We may assume that, so far as claims thirteen and fourteen include by implication a slot which leads into an oil trap in the groove, they may be valid; but they are not then infringed. To reverse the decree we must construe them as covering any slotted ring. It is true that literally they are new; there had never been the combination of an "expander" with a slotted ring which was so flexible as properly to accommodate itself to a worn cylinder. The new feature was the slots, on whose first function -- to weaken the ring -- the invention cannot however stand. Given that purpose, and given earlier rings so thin that they would serve, it was a mere detail whether to weaken them by slotting, by thinning, by narrowing the gap, or by all three. Slots were themselves old, and a means at hand, once the purpose was conceived.
The second function -- that of lubrication -- is, as we have said, a gloss upon the patent. We agree that this is not fatal; a patentee may invoke in support of his invention advantages of which he was not aware; Eames v. Andrews, 122 U.S. 40, 70, 7 S. Ct. 1073, 30 L. Ed. 1064; Diamond Rubber Co. v. Consolidated Rubber Tire Co., 220 U.S. 428, 435, 436, 31 S. Ct. 444, 55 L. Ed. 527; Novadel Process Corp'n v. J. P. Meyer & Co., 35 F.2d 697, 702 (C.C.A. 2). However, the structure must disclose them; the patentee may have builded better than he knew, but he is confined to what he built. The difficulty we find here is that all that he showed was an oil trap, and this the defendant does not use. However such a trap helps in lubrication, it is the merest speculation to suppose that the drained groove does ...