Opinion
No. 9711.
February 7, 1945.
Appeal from the District Court of the United States for the Eastern District of Michigan, Southern Division; Arthur F. Lederle, Judge.
Action by Universal Products Company, Inc., against Montgomery Ward Company for infringement of two patents. From a judgment for the defendant, 50 F. Supp. 798, the plaintiff appeals.
Judgment amended, and as so amended affirmed.
John A. Blair, of Detroit, Mich. (Harness, Dickey Pierce, all of Detroit, Mich., on the brief), for appellant.
Virgil E. Woodcock, of Philadelphia, Pa. (Edwin J. Balluff, of Detroit, Mich., on the brief), for appellee.
Before HICKS, ALLEN, and McALLISTER, Circuit Judges.
Appellant, Universal Products Company, assignee, brought suit against Montgomery Ward Company, appellee, for infringement of two patents, in the selling and offering for sale of universal joints and parts thereof. The claims in suit were No. 4 of Patent No. 1,921,274, August 8, 1933, to Warner, for a "Universal Joint" and the sole claim of Patent No. 1,987,678, January 15, 1935, to Goddard and Ketcham for a "Universal Joint and Cover." The District Court found the claims in suit invalid for lack of invention. It declined to pass upon the defenses of noninfringement and unclean hands.
Universal joints are adapted for the transmission of torque between shafts mounted at an angle to each other and the overall length of which alters in operation. In an automobile there are usually two universal joints in the driving assembly between the engine and the rear axle. In the type of joint in suit the joint itself must absorb the shifting angle and shifting length of the assembly caused in part by the up and down motion of the rear axle. There must be provision within the joint for movement and adjustment of its connecting parts with reference to each other, without interruption in the flow of power from one of these parts to the other.
The principal members of this type of joint are a drive shaft with a head, and a body or casing attached to a second shaft into which the first shaft fits. Earlier patents had a ball-shaped head adapted to fit into the central bore of the body as in Emerson, No. 1,183,064, — 1914; or, into a spring-held socket, as in Flick, No. 1,512,840, — 1924; and the axis of each member of the assembly was centered with reference to each other by these connections. The motion of the drive shaft was transmitted to the casing of the second shaft by a pin or cross-head which was inserted crosswise through an opening in the ball-head and which had trunnions at either end Balls were mounted on each trunnion, and as the shaft rotated, the balls or rollers impinged upon cylindrical recesses oppositely disposed in the casing on each side of the central bore, and transmitted the rotative motion or torque of the drive shaft to the casing. The ball and socket type joint, best exemplified in Flick, had the weakness that the centrifugal force developed by rotation tended to throw the lubricant in the joint away from the ball and spring guide at the center of the assembly, causing wear of those parts. In automobiles this weakness was accentuated by high speeds.
This joint was known as the "sliding trunnion" type, since the ends of the trunnions slid back and forth in the cylindrical recesses as the shaft rotated. In Emerson and Flick the trunnions and balls were driving media only. No centering function was claimed for them, although the balls were designed to fit the recesses snugly, to eliminate a clicking sound as the car started.
Centering is the maintenance of the point of intersection of the axes of the two shafts at the center of the ball-head or shaft head. Centering was important because any eccentricity at high speeds resulted in a violent and irregular motion called "whip" which tended to rack the joint and damage the car.
The Lipe joint, patent No. 942,087, December 7, 1909, was a sliding trunnion type but utilized a different principle for the centering. The casing was boxlike with two flat sides connected by two flatly cylindrical sides, the inside arcs of which were parts of a circle which had its center in the head of the shaft. The shaft head carried a cross pin and trunnions similar to Emerson but it had flat sided rollers for driving members which engaged the flat sides of the casing. The heads of these rollers were sections of a sphere, radii of which would converge at the center of the shaft head. There was no ball head on the shaft and no socket. Centering was attained through a rolling engagement of the rollers with the flat sides of the casing and by sliding engagement of the spherical heads of the rollers with the cylindrical portions of the casing.
In addition to the above, Lipe sets forth, in his specification, another construction of his joint. He says: "In Figure 4 I have shown a slightly different construction of my universal joint in which spherically curved surfaces for one of the joint-members 22, are provided on screws 23 which secure to said joint-members engaging parts 24 having peripheral engaging surfaces 25. * * *"
The McGee patent, No. 1,582,997, May 4, 1926, had similarities to Lipe, and in addition disclosed spring pressed plungers mounted in recesses in the ends of the cross head, which bore against a machined surface in the "coupling box" or casing.
The Lipe joint had drawbacks in its size, cost and torque capacity. Its parts had to be assembled and bolted together, which added to cost, weight and bulk, but Dunn, appellant's chief engineer, said of it: "* * * the torque is taken through the flat sided driving members. The centering is done by spherical buttons which are centered on the cylindrical portion of the body. It is a very good method of centering. * * * The motion in one plane is rotary motion of the trunnion and the button. In the other plane, it is rotary motion about the center, the axis of the joint, as the sphere produced by the buttons, — that is, these buttons actually act as part of a total sphere and can be slid to any portion without play. * * *" Thus, the inside facings of the Lipe casing were constructed to control the motion of the drive head and its parts, so that any oscillation in either plane centered at the same point in the drive head, as it would have in the ball socket type.
Warner used the type of body disclosed by Brush, No. 966,519, August 9, 1910, which could be manufactured more simply, and weighed less than Lipe. It could be forged or pressed, and the central bore and the two oppositely disposed cylindrical recesses, since they were parallel, could be machined in a boring operation, thus simplifying greatly the process of manufacture. But because of the size of the cylindrical recesses or guides, the trunnion mounted balls which operated therein were necessarily of such small diameter that in their sliding action, they centered the cross head only when it was in a position perpendicular to the long axis of the guides. WARNER PATENT NO. 1,921,274:
We shall describe this patent with somewhat more particularity. It eliminated ball head socket centering, utilized the small compact casing of Brush, Emerson and Flick, and achieved centering between the guides by sliding contact between the fittings at the end of the cross head and the cylindrical guides. Warner mounted on the trunnions truncated spherical bearing members whose radii were substantially equal to the radius of curvature of the cylindrical guides or raceways. These spherical members were rotatable upon the trunnions. At the ends of the trunnions, centering buttons, somewhat like the plungers of McGee, or the buttons of Lipe with the screw feature eliminated (Figure 4 of the Lipe drawings), were provided for engagement with the outer portions of the cylindrical guides. These buttons resembled mushrooms in that they had a central stem, fitting recesses in the ends of the trunnions, and a flared head which overlay the ends of the truncated spherical members. Around the stem and between the flared head and the end of the trunnion was a bowed spring washer which tended to maintain the buttons in contact with the cylindrical guides. (This spring washer was not claimed.) The heads of these buttons had their outer surfaces curved "substantially about the axis of fulcruming movement of the shaft member" in the plane of the raceways. That is, in this plane, the surfaces of the buttons were shaped to the curve of a circle with radii focusing at the center of movement of the head of the shaft, as in Lipe. Transversely, the surface of the buttons had a radius of curvature substantially equal to that of the cylindrical guides, thus the surface of the centering button had a double curve, although not spherical as in Lipe, with the center of one curve at the fulcruming point of the shaft head and of the other coincident with the center of a cross section of the raceway.
Claim 4 is as follows: "4. A universal joint comprising a casing having opposed and parallel cylindrical guides, a shaft member projecting endwise into the casing between the guides, trunnions projecting transversely from opposite sides of the member and into the guides, truncated spherical members having openings receiving the trunnions for slidable movement thereon, each spherical member contacting with one guide at opposite sides of the trunnion, and a member on the end of each trunnion for engaging the guide beyond the end of the trunnion, said last mentioned member having its surface next to the wall of the guide curved in the direction of the guide about an axis passing transversely through the shaft member substantially at its center and curved transversely of the guide substantially with a radius of curvature equal to that of the guide."
We fail to find anything in this combination of an inventive nature. Most of its elements are grouped together in Brush, Emerson and Flick. Warner has added another element, which may be described in short as the button with the double curve. In so far as these buttons were "curved in the direction of the guide about an axis passing transversely through the shaft member substantially at its center" they are found in Lipe. In Lipe these spherical buttons engage and fit the spherical curvature of the guides. So far nothing new is found in claim 4 but appellant claims that in addition to the spherical curvature of his buttons, longitudinally parallel with the guides, they curve "transversely of the guide substantially with a radius of curvature equal to that of the guide."
With reference to this feature of his claim, Warner, in the specification, says: "Preferably also it has been found desirable to curve such centering buttons transversely of the plane of such fulcruming movement, with a radius of curvature substantially equaling that of the side cylindrical guides, so that greater contact will be had between the buttons and the walls of the side guide during the fulcruming movement."
Dunn further explains that: "The Lipe button has a spherical contour. In other words, it is just part of a sphere. If that were carried on, it would generate a sphere. That is not true of the Warner patent. The Warner button has a compound curvature * * * it would have the same curvature as the Lipe in one plane, but in the opposite plane it would have the same curvature as the raceway, in which it fits. * * * About this axis, if we turn this button it acts precisely as the Lipe button does, but in order to be able to clear itself, as it were, when you get off center, that is the curvature, say, at the edge of the button, or out here, is not the same as the curvature in the center. Therefore, you can't have the same radius. In other words, it cannot be a sphere, it has to be of some other curvature, whereby it will clear and yet contact at all points. And it works out, then, that the fact it is given the same radius in this direction as the radius of the raceway, that it will clear at all points." (Italics ours.)
Again, in referring to a model of the Lipe patent in evidence, Dunn testified in response to the question, "Functionally, it is centered exactly as the pin is centered in the Warner type of joint," replied, "Yes, that is true."
We think that except for the double curve button Warner utilized well known constructions. He adapted the Lipe elements of centering to the compact bodies of Brush, Emerson and Flick. There is nothing new in a double curve. It is as commonplace as a doughnut. Because the Warner button had to function transversely in the side cylinder guides, it required a radius of curvature equal to that of the guides. In other words, it had to be curved to fit the space. To accomplish this was not invention. Nothing more was involved than mechanical skill. It was a matter of fitting rather than invention and when the fitting was completed the centering was still attained "precisely" as in Lipe and the contact of Lipe was still present. Adjustability only was involved and adjustability is not invention. Paquette v. Potter Mfg. Co., 6 Cir., 46 F.2d 271. See also United States Gypsum Co. v. Consolidated Expanded Metal Co., 6 Cir., 130 F.2d 888, 893.
The Warner joint was smaller, cost less to manufacture and had commercial success, but we are not convinced that this was due to the claimed combination. We note that in the commercial model needle rollers were substituted for plain bearings in the drive balls, and Dunn admitted that an increase of efficiency of from 10 to 1 to 50 to 1 might be expected in shifting from plain bearings to needle rollers, depending upon the application in which they were used. Evidence that the commercial success of the Warner combination was due to the torus-shaped buttons rather than to the needle bearings, is notably lacking.
GODDARD AND KETCHAM, PATENT NO. 1,987,678:
According to the specification, the purpose of this patent was "to provide an improved assembly of joint in which the component parts are permanently retained within the housing; to provide an improved cover which seals lubricant within the joint, excludes dust and dirt prior to final assembly on the vehicle, and which serves as a spring seat for holding the base of the spring in definite position."
It had been the practice to furnish automobile manufacturers with propeller shaft assemblies for universal joints without a permanent cover on that end of the housing abutting the flanges, and opposite to the end in which the propeller shaft is inserted. Except for an intervening gasket, the housing was bolted directly to the flange and the temporary cover which held the various parts in place during shipment was removed before assembly. When the cover was removed some times the centering buttons or the balls or needle rollers fell out while the union was being made and sometimes were not replaced. Sometimes the workmen failed to put in enough grease, or had difficulty with the compression spring interposed between the shaft head and seat in the flange portion, thus slowing the assembly line.
The specifications continue: "In the present construction, the flanged end of the universal joint housing is provided with a permanent cover plate with a gasket interposed for sealing the housing and at the same time providing a permanent seat for the compression spring. The proper amount of suitable lubricant may be placed into the universal joint housing at the time of its manufacture and assembly and the spring properly positioned therein so that the propeller shaft assembly with joints thereon may reach the automobile assembly line as a single complete unit ready for bolting to the respective companion flanges of the transmission shaft and axle pinion shaft."
The cover plate was provided with a plurality of tabs bent over the flange of the housing and permanently securing it thereto. The flange was recessed so that, when the tabs were bent around it, they fit flush with it, thus permitting snug assembly into the companion flange. The plate also had an annualar groove at its center to hold the spring in place. The specification states that: "Such a construction provides a unitary assembly which may be readily placed in position against its companion flanges and bolted thereto on the assembly line without the assembly workmen bothering with the insertion of lubricant, springs or other loose parts."
The claim was: "In a universal joint, a housing having an open end surrounded by a radially extending attaching flange, a cover plate substantially co-extensive with said flanged end, and a plurality of tabs on said plate for engaging said housing flange, said flange being recessed in registry with and to a depth substantially equal to the thickness of said tabs."
Appellant's counsel limited the claim in a colloquy with the court, as follows:
"Mr. Blair: It is a universal practice to fasten things on by tabs.
"The Court: Then the only step that you are claiming for this patent is that the designer, * * * decided not to have those tabs overhang over the outside diameter of the body, that you made a recess there to fit them in.
"Mr. Blair: That is right. You had to slot the body and get the tabs inside.
* * * * * *
"I am perfectly prepared to stand on the claim as it exists without any further limitations. It specifically is limited to the very meat of the inventive thought. * *"
The question thus left with the court was, whether there was invention in recessing the outside diameter of the casing to receive the tabs. The court held that it was not because it was well within the skill of the art and the decision is obviously correct.
We need not pass upon the questions of infringement and unclean hands.
The decree went beyond the pleadings in that it held patent No. 1,921,274 void in its entirety. Claim 4 was the only claim in issue. The decree will therefore be amended to declare claim 4, only, to be void for want of invention; and so amended will be affirmed.
Affirmed.