Argued February 26, 1981.
Decided October 27, 1981.
Robert W. Gutenkauf, Minneapolis, Minn., for plaintiff-appellant.
Glenn O. Starke, Milwaukee, Wis., for defendant-appellee.
Appeal from the United States District Court for the Eastern District of Wisconsin.
Plaintiff-appellant George M. Mooney, the inventor and owner of the two patents which formed the basis of this suit, seeks review of a district court order holding one patent invalid and the other not infringed. U.S. Patent No. 3,245,280 (the "gear structure patent") involves a one piece ratchet clutch and bevel gear used primarily in outboard and stern drive motors. U.S. Patent No. 3,248,782 (the "method patent") discloses a method of production of the one piece gear unit.
On July 31, 1974, Mooney sued the Mercury Marine Division of Brunswick Corporation, a well-known manufacturer of outboard and stern drive motors, alleging infringement of Claim 4 of the structure patent and Claim 1 of the method patent. Brunswick answered and counterclaimed for a declaratory judgment of patent invalidity and non-infringement. An eight-day bench trial was conducted in February 1979. On March 21, 1980, the district court issued its decision and order concluding that Claim 4 of the gear structure patent was invalid for obviousness in light of the prior art under 35 U.S.C. § 103, and that Claim 1 of the method patent had not been infringed. Mooney filed a timely appeal. We affirm.
This litigation focuses on the ratchet clutch and bevel gear units found in both outboard and stern drive boat motors. The engine in both the outboard and stern drive systems is located above water level. In the outboard system, to which for the sake of simplification we shall confine our comments, the engine is attached to a vertical drive shaft. With this system, power must be transmitted from the vertical drive shaft to the underwater horizontal propeller shaft (which drives the propeller) by means of a single bevel gear set.
In the stern drive system, the above-water engine is attached to a horizontal drive shaft which is connected by bevel gears to a vertical drive shaft. This vertical drive shaft in turn is connected by an additional set of bevel gears to the horizontal propeller shaft. Power flows from the horizontal drive shaft, down the vertical drive shaft to the horizontal propeller shaft.
A bevel gear set incorporates two parallel beveled gears which both mesh with a central pinion gear attached at the bottom of the vertical drive shaft. Each of these parallel beveled gears then turns in a direction opposite to the other (clockwise and counterclockwise) as they are driven by the pinion gear. One of these parallel gears (having a clockwise motion) serves as a "forward" gear to drive the propeller in a forward direction while the other parallel gear (having a counterclockwise motion) serves as a "reverse" gear to turn the propeller in the opposite direction.
In order to shift the power applied to the propeller from "forward" to "neutral" to "reverse," a clutch must be used in connection with the beveled gears. A clutch member (between the two parallel beveled gears) is keyed to the propeller shaft; the clutch member can engage either the "forward" or the "reverse" beveled gear, or can be disengaged from both. To accomplish this engagement, each of the beveled gears has firmly attached to it, and recessed concentrically within it, a ratchet clutch which directly engages the clutch member. Thus, the clutch member moves between the beveled gears from engagement with the ratchet clutch attached to the "forward" gear, to disengagement (neutral), to engagement with the ratchet clutch attached to the "reverse" gear. As the shift lever moves the clutch member from "forward" to "neutral" to "reverse" positions, the movements of the propeller and the boat respond accordingly.
Figures 1-3 which are based on plaintiff's exhibit 4 (with certain additional information which we have provided) illustrate a clutch and bevel gear unit and how it operates by movement of the clutch member from engagement with one ratchet clutch (Figure 1), to disengagement (Figure 2), to engagement with the other ratchet clutch (Figure 3).
The means of binding together the respective recessed ratchet clutches with the two beveled gears is essentially the subject of this controversy. Specifically, the ratchet clutch concentric with, and recessed within, each of the beveled gears was originally brazed and pinned to the outer bevel gear ring (thus constituting a two piece unit). The gear structure patent here contemplated the fabrication and production of a recessed ratchet clutch and an outer concentric beveled gear as one unit from a single piece of metal. The method patent described how such a fabrication could be accomplished and the bevel gear unit produced.
Plaintiff's company, Capital Gears, began to supply ratchet clutch and bevel gear units to Kiekhaefer Corporation, the predecessor of defendant Brunswick, in the late 1950's. At that time, as indicated, the clutch and bevel gear unit was manufactured in two pieces which were subsequently brazed together. But as Kiekhaefer began to manufacture more powerful and sophisticated motors, more stress was placed on the gears, and the brazings did not hold. To strengthen the unit, Kiekhaefer began to pin the brazed joints, but this method of fabrication also proved inadequate when the corporation introduced even more powerful motors which required correspondingly stronger gears.
Kiekhaefer contacted plaintiff, the president of Capitol Gears, in early 1962 in an effort to obtain a solution to the problem. Mooney subsequently conceived the notion of producing a one piece gear and immediately informed Kiekhaefer of his idea. Within approximately eight weeks, Mooney designed and completed a machine to produce a one piece clutch and bevel gear unit for commercial production. Samples of the new gear were sent to Kiekhaefer for tests, and the results demonstrated that Mooney's one piece unit was superior to the two piece construction.
One piece gears had not been used in commercial production prior to Mooney's invention apparently because pre-existing machine tools could not cut the ratchet clutch teeth which were recessed from the plane of the bevel gear and therefore inaccessible. See Patent No. 3,245,280, Col. 2, lines 9-20.
On January 13, 1963, Mooney applied for a patent on both the one piece ratchet clutch and bevel unit and the method of production. After the Patent Office determined that Mooney's claim covered two separate inventions (i.e., the gear structure and the method), he was required to restrict the application to only one invention. Mooney elected to pursue his claim on the method of production and filed a second divisional application for a patent on the gear structure. The Patent Office issued the gear structure patent (U.S. Patent No. 3,245,280) to Mooney on April 12, 1966. The method patent (U.S. Patent No. 3,248,782) was granted on May 3, 1966.
At the time of issuance, both patents were assigned to Capitol Gears, Inc. On May 15, 1974, Capitol Gears reassigned the patents to Mooney, who filed this suit several weeks later.
Although the gear structure patent is composed of several claims, only Claim 4 is at issue here. Claim 4 essentially describes a one piece beveled gear with a recessed ratchet toothed clutch.
4. As a new article of manufacture, a one piece integral composite gear and clutch adapted for rotation about an axis, said gear and clutch including an unobstructed annular ring of a plurality of gear teeth disposed about said axis of rotation and an annular ring of a plurality of actuate sloping ratchet teeth concentric with and within the inner periphery of said ring of gear teeth, said ratchet teeth being recessed relative to said gear teeth, each of said ratchet teeth having a generally radial engaging face extending generally along a line passing through said axis of rotation, the engaging face of each of said ratchet teeth being inclined inwardly from the ridge of each of said teeth by up to 8 ° . (PX 28). App. p. A-22.
The method patent consists of two claims, but only Claim 1 is at issue here. Claim 1 describes the four step procedure developed by Mooney to manufacture a one piece ratchet clutch and bevel gear unit. First, a cylindrical piece of steel bar stock is machined to outline shape. Second, the work piece is placed in a jig and the sloping ramp surface of a ratchet tooth is cut by bringing a rotating cutting tool with an end cutting face into contact with the work piece, which also rotates relative to the cutting tool (producing a cut "in one direction"). Third, the engaging face of the ratchet tooth is cut by rotating the cutting tool "in another direction" and bringing a side edge cutting face into contact with the work piece, which again rotates relative to the cutting tool. Finally, the bevel gear teeth are cut into the outer ring of the work piece, and the one piece gear unit is complete.
Claim 1 provides:
1. A method of making a one piece integral/composite polyfunctional toothed working part adapted for rotation about an axis and having at least two concentric annular rings of teeth disposed about that axis, one of said rings of teeth being unobstructed by any portion of said working part and another of said rings of teeth being recessed with respect to said first ring of teeth, said method comprising forming a work piece by machining metal stock to outline shape, cutting said recessed teeth by rotating a rotatable cutting tool having an end cutting face in contact with the blank work piece and moving said work piece relative to the end cutting tool in one direction and rotating a rotatable cutting tool having side edges cutting faces in contact with said work piece and moving said work piece relative to the side cutting tool in another direction to delineate said teeth, and cutting said unobstructed teeth by revolving edge cutters of a gear cutter in contact with the blank work piece and moving said work piece relative to the gear cutter to delineate said teeth. (PX 29). App. p. A-23.
At the time Mooney developed the one piece gear unit, Kiekhaefer did not have the facilities to manufacture its own gear units and purchased all the necessary units from Capitol Gears or other suppliers. In 1965, however, Kiekhaefer hired Roger Shine, a gear development engineer, to establish a small production machinery department. By 1966, Brunswick, the successor to Kiekhaefer, had begun to produce a number of components for outboard and stern drive motors, including two piece ratchet clutch and bevel gear units. Brunswick then hired Precision Design, Inc., to develop a method for producing one piece gear units similar to the one invented by Mooney. In 1968, Brunswick began to manufacture its own one piece ratchet clutch and bevel gear unit although it still purchased both one and two piece units from Capitol Gears.
The one piece gear unit produced by Brunswick does not materially differ in appearance from the Mooney one piece unit. However, Brunswick uses a somewhat different four-step method to manufacture the unit. First, Brunswick forges a work piece with pre-delineated clutch teeth. Second, the engaging faces of the ratchet clutch teeth are cut by a rotating cutting tooth with a side cutting face which moves radially from the center of the work piece. Third, the sloping ramps of the ratchet teeth are cut by a rotating cutting tool with an end cutting surface which moves in the same radial direction. Finally, the bevel gear teeth are cut into the outer ring of the work piece.
The parties agree that defendant's one piece gear:
includes an unobstructed annular ring of a plurality of gear teeth disposed about an axis of rotation and a ring of ratchet clutch teeth located within the periphery of the ring of gear teeth. The ratchet teeth are recessed relative to the gear teeth. Each ratchet tooth has a generally radially engaging face which extends generally along the line passing through the axis of rotation of the gear.
Some of the ratchet clutch and bevel gears made by the Defendant have ratchet teeth with engaging faces that are inclined inwardly from the ridge of the teeth by an angle of between 3 ° — 8 °
App. p. A-25-6.
Brunswick's success in producing a one piece ratchet clutch and bevel gear unit prompted Mooney to institute this action. Specifically, Mooney charged that Brunswick wilfully and deliberately infringed Claim 4 of his gear structure patent and Claim 1 of his method patent through the manufacture, sale and use of its own one piece ratchet clutch and bevel gear unit. Brunswick denied any infringement and challenged the validity of both patents on grounds of obviousness ( 35 U.S.C. § 103), and inadequate description ( 35 U.S.C. § 112). In a commendably thorough opinion, the district court concluded that Claim 4 of gear structure patent was obvious to one skilled in the art and therefore invalid. Mooney v. Brunswick Corp., 489 F. Supp. 544 (E.D.Wis. 1980). The court also determined that defendant's method of producing its own one piece ratchet clutch and bevel gear units was sufficiently distinct to preclude infringement of Claim 1 of Mooney's method patent. On appeal, Mooney seeks review of the district court's rulings both as to validity and to infringement.
Mooney initially asserts that the district court erred in concluding that Claim 4 of the gear structure patent would have been obvious to one skilled in the art at the time the patent was issued and therefore is invalid under 35 U.S.C. § 103. This allegation of error is based on two contentions. First, Mooney argues that the district court did not give proper weight to the presumption of validity accorded to a patent which has received the approval of the Patent Office through reissue proceedings. Second, Mooney maintains that the district court incorrectly determined that the gear structure patent would have been obvious to one skilled in the art.
A. The Reissue Proceedings.
Reissue proceedings were originally designed to enable a patent holder to remedy inadvertant mistakes in patent drawings and specifications. Sontag Chain Stores Co. v. National Nut Co., 310 U.S. 281, 60 S.Ct. 961, 84 L.Ed. 1204 (1940). The procedure was amended in March of 1977, however, to permit patent owners to obtain a ruling from the Patent Office on the pertinence of additional prior art which was not cited in the original patent application. See 37 C.F.R. § 1.175(a)(4) (1980). If the Patent Office determines that the patent is not defective or invalid in light of the additional prior art, the reissue application is rejected. If, however, the patent is defective, the reissue application is allowed and the owner of the patent must surrender it to the Patent Office in return for a corrected Reissue Patent. 35 U.S.C. § 251.
In an attempt to introduce an adversarial element not found in routine patent applications, the reissue process also permits third parties to file opposition statements challenging the validity of the patent in issue. Once a reissue application has been filed, notice of the filing is published in the Official Gazette of the United States Patent and Trademark Office. 37 C.F.R. § 1.11(b) (1980). Members of the general public are then allowed to submit statements in opposition to the reissue application. 37 C.F.R. § 1.291 (1980). Although the rules specify no definite time period during which protests must be filed, no action will be taken on the reissue application for two months following publication in the Official Gazette. 37 C.F.R. § 1.176 (1980). In addition, patent examiners are not permitted to discuss questions of patentability involving the patent at issue with applicants until after the first official action of the Office. 37 C.F.R. § 1.133 (1980). These rules were designed to enhance the reliability of reissue procedures by permitting the examiner to study the written record (which includes both supporting and opposing statements) and make an independent determination on patentability without being subject to outside influence.
The plaintiff in the instant case was one of the first patent owners to file a reissue application under the amended procedures. In his application, Mooney cited 56 prior art references including 51 separate patents and 5 general items which he thought the examiner should review in connection with the application. Each of these items of prior art had been uncovered during discovery in the present litigation. Plaintiff apparently filed the reissue application in order to strengthen his position in this case, and properly informed the Patent Office that the lawsuit was pending.
On August 9, 1977, notice of Mooney's reissue application was published in the Official Gazette. An attorney for Brunswick subsequently filed an opposition statement with the Patent Office on November 18, 1977. A second opposition statement was filed by Robert E. Clemency on November 23, 1977. Both these statements also noted prior art relevant to Mooney's reissue application.
The Patent Office denied the reissue application on May 23, 1977. The examiner stated that he had considered the prior art provided by the applicant and concluded that the claims of the gear structure patent were allowable over the prior art. Since the specifications of the patent needed no correction, the application lacked a statutory basis for reissue and was therefore rejected.
All patents enjoy a general presumption of validity which can be overcome only by clear and convincing evidence to the contrary. 35 U.S.C. § 282 (1979). See Chicago Rawhide Manufacturing Co. v. Crane Packing Co., 523 F.2d 452 (7th Cir. 1975), cert. denied, 423 U.S. 1091, 96 S.Ct. 887, 47 L.Ed.2d 103 (1976). On this appeal, Mooney maintains that this presumption of validity was strengthened when the Patent Office approved his patent through rejection of a reissue application. After examining the record, we conclude, as did the district court, that irregularities in the reissue proceedings in the instant case diminished rather than enhanced the presumption of validity of Mooney's gear structure patent.
Reissue proceedings do not necessarily strengthen the presumption of validity of a patent in every case. Johnson Johnson, Inc. v. Wallace A. Erickson Co., 627 F.2d 57 (7th Cir. 1980).
The district court found several serious irregularities in the reissue procedures involved in the present case. Most importantly, the court concluded that the opposition statements filed with the Patent Office were not properly considered by the examiner. Although the statements were never entered in the record and acknowledged by the Patent Office in accordance with applicable regulations, testimony at trial revealed that the examiner did have the opposition statements in his possession. Mooney argues that we must therefore assume the examiner considered the merits of these statements. Since the examiner never cited the statements in his final evaluation, however, we must conclude, as did the district court, "that they were not considered or alternatively they were not considered relevant by the examiner." Mooney, 489 F. Supp. at 555. In either case, the statements were not apparently accorded their proper weight and this irregularity diminishes the effect of the reissue procedure.
The district court also concluded that the patent examiner failed to consider two of the most relevant pieces of prior art, the Kiekhaefer drawing (No. 43-31886) and the Gale Products drawing (No. 302,486). Since both of these drawings were disclosed to the examiner in the reissue application, Mooney claims that it can be assumed that the patent examiner gave them proper consideration. Under the circumstances of this case, this argument is not persuasive.
Mooney's reissue application cited 56 prior art references (including the Kiekhaefer drawing and the Gale Products drawing) but in no purported order of importance. The letter which accompanied the application similarly failed to disclose which references were most relevant to the patentability of the gear structure. Ordinarily, an examiner specifies the pertinent prior art references in the Office Action, but neither the Kiekhaefer drawing nor the Gale Products drawing were listed by the examiner in the Office Action in this case. Considering the large number of references which were included on the patent examiner's list, we do not think it unreasonable for the district court to have concluded that the examiner failed to consider these drawings as relevant prior art. As a result, the statutory presumption of validity which may be enhanced when prior art has been considered and rejected by the Patent Office is dissipated in Mooney's case with respect to the Kiekhaefer drawing and the Gale Products drawing. Compare Ortho Pharmaceutical Corp. v. American Hospital Supply Corp., 534 F.2d 89 (7th Cir. 1976) and Tracor, Inc. v. Hewlett-Packard Co., 519 F.2d 1288 (7th Cir. 1975) with Chicago Rawhide Manufacturing Co. v. Crane Packing Co., supra.
Finally, the court found that an associate of Mooney's attorney conducted interviews with the patent examiner, relating in part to patentability, before the first Office Action on Mooney's reissue application in violation of Patent Office rules. While Mooney does not deny the interviews, he maintains that they were conducted exclusively for the purpose of providing prior art to the examiner. The district court founded that most of the interviews were only for the purpose of making full disclosure of the prior art, but that some of the material discussed involved patentability.
Although the interviews in the instant case may have been to a considerable degree innocent in purpose, we believe that the proceeding may have been somewhat tainted by them. The purpose of the rule prohibiting interviews between the examiner and the parties involving discussion of patentability is to make the procedure more reliable and to permit the examiner to make an independent evaluation of the merits of the application. Any information which the examiner needs to consider should be submitted with the application or included in supplemental filings. It is not appropriate for a representative of the patentholder (or for that matter, of an opposing party) to discuss patentability with the examiner before the first Office Action. We therefore conclude that the district court properly determined that the "effect of this defect as well as the other[s] raised by defendant is to diminish any presumption of validity that would attach in the consideration of the prior art revealed under the reissuance procedure." Mooney, 489 F. Supp. at 554.
Even in the absence of a fortified presumption of validity, Mooney claims that the district court erred in concluding that the one piece ratchet clutch and bevel gear unit would have been obvious to one skilled in the art at the time the patent was issued. In reviewing a district court's determination of obviousness, we must examine (1) the scope and content of the prior art; (2) the differences between the prior art and the claim at issue; and (3) the level of ordinary skill in the pertinent art. Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Republic Industries, Inc. v. Schlage Lock Co., 592 F.2d 963 (7th Cir. 1979); Centsable Products, Inc. v. Lemelson, 591 F.2d 400 (7th Cir.), cert. denied, 444 U.S. 840, 100 S.Ct. 79, 62 L.Ed.2d 52 (1979). As to these matters, the findings of the district court should be sustained unless clearly erroneous. See La Salle Street Press, Inc. v. McCormick Henderson, Inc., 445 F.2d 84 (7th Cir. 1971).
(1) The Scope and Content of the Prior Art
The first step in resolving a question of obviousness involves a consideration of all the prior art. See New York Scaffolding Co. v. Liebel-Binney Construction Co., 254 U.S. 24, 41 S.Ct. 18, 65 L.Ed. 112 (1920). The prior art includes any relevant knowledge, acts, descriptions and patents which pertain to, but pre-date, the invention in question. 35 U.S.C. § 102.
In the instant case, Mooney's original patent application for the gear structure unit referenced only one piece of prior art, the Kaufmann patent (U.S. Patent No. 2,289,288). The Kaufmann patent described a clutch and bevel gear with recessed sloping ratchet teeth and radially engaging faces originally used in washing machines. Although the gear unit was actually manufactured in two pieces, the diagram accompanying the patent discloses a one piece unit. Regardless whether this disclosure occurred by design or by accident, we agree with the district court that the Kaufmann patent suggests a gear unit sufficiently similar to the unit designed by Mooney that the Kaufmann patent must be considered relevant prior art. See Application of Meng, 492 F.2d 843 (C.C.P.A. 1974); Application of Seid, 161 F.2d 229 (C.C.P.A. 1974).
The district court determined that the prior art also included two other drawings of clutch and bevel gear units. The Kiekhaefer drawing describes the two piece gear unit used by the defendant prior to Mooney's invention. The Gale Products drawing (No. 302,486), however, discloses a one piece bevel gear with two recessed ratchet teeth, which was used in outboard motors in the late 1940s.
Both plaintiff and defendant referenced numerous other prior art patents before the patent examiner and the district court. The examiner considered those prior art patents as part of the plaintiff's reissue application. The district court considered those prior art patents which the examiner failed to reference on the reissue application, finding that they were either irrelevant or relevant only to a very limited extent. Mooney, 489 F. Supp. at 555-58.
While plaintiff does not appear to argue that the Kaufmann patent and the Kiekhaefer and Gale Products drawings do not represent relevant prior art, plaintiff does claim that these references are not sufficient to invalidate Claim 4 of the gear structure patent. According to plaintiff, a reference can only render an invention obvious if it contains an enabling disclosure. See Ortho Pharmaceutical Corp. v. American Hospital Supply Corp., supra. Mooney maintains that Claim 4 of his patent remains valid because the references cited by the district court do not contain sufficient information to enable a person of ordinary skill in the art to make a one piece gear unit and thus, the unit is not "obvious" from the prior art.
Mooney makes a similar argument in suggesting that the court applied the wrong test to determine obviousness. See discussion infra at 736-737. Based primarily on expert testimony, the district court concluded that it would have been obvious to one skilled in the art "to make" a one piece gear. The thrust of Mooney's argument, however, is that a one piece gear could not have been obvious in the absence of a method for producing such a gear. In fact, the district court noted that ". . . it was the method of producing the gear and not the gear which was in fact unobvious." Mooney, 489 F. Supp. at 561.
We believe that the references relied upon by the district court contained sufficient enabling disclosures for the purpose of serving an invalidating function in this case. "It is well settled that a publication must contain sufficient disclosure to enable a person of ordinary skill in the art to make and use the invention described in the patent before the patent will be invalidated by the prior art publication." Ortho, 534 F.2d at 93 (emphasis supplied). But it is also generally known, as found by the district court, that fabrication of a one piece gear unit would have to proceed by one of the various well-known metal-working methods, e.g. forging, casting or machining. Mooney, 489 F. Supp. at 560. Further, the structure patent here in issue may be invalidated for obviousness in light of the prior art even though the means of manufacture disclosed in the prior art publication is not the same as that claimed by the patent holder. Application of Payne, 606 F.2d 303, 315 (C.C.P.A. 1979). The district court specifically found that four prior art patents and drawings revealed a one piece gear unit similar to plaintiff's design. Mooney, 489 F. Supp. at 556-58, 561. These examples of prior art, in relation to the level of skill then existing in the mechanical gear design field, were considered in detail by the district court when it held that the use of a one piece gear unit construction was obvious to those possessing ordinary skill in the art. The prior art patents and drawings relied on by the district court evince sufficient enabling disclosures (in light of the common knowledge of methods of metal fabrication) for that court to have properly relied on them for purposes of invalidation. Cf. Application of Payne, 606 F.2d 303 (C.C.P.A. 1979) (prior patent on analogous chemical compound contained sufficient enabling disclosure when new compound is made by atom substitution which is obvious to those skilled in art); Application of Collins, 462 F.2d 538 (C.C.P.A. 1972) (existing metal working techniques, when considered with prior art patent, disclosed method to make new heat exchanger). See also In re Donohue, 632 F.2d 123, 125-26 (C.C.P.A. 1980) and In re Sasse, 629 F.2d 675, 681-82 (C.C.P.A. 1980) (enabling disclosures sufficient for 35 U.S.C. § 102).
This conclusion about the adequacy of the enabling disclosures is buttressed by an analysis of the special factors relied on in cases where enabling disclosures in prior art publications have been held insufficient to render the publications adequate to invalidate subsequent inventions for obviousness. In Ortho, for example, the prior art consisted solely of a magazine article describing an unsuccessful attempt to prepare a similar chemical compound. 534 F.2d at 93; cf. Application of Brown, 329 F.2d 1006 (C.C.P.A. 1964) (prior art disclosing unsuccessful attempts to manufacture insufficient to constitute enabling disclosure). The prior art and publications relied on in the instant case, however, reveal successful efforts at utilizing one piece gear unit designs. In Application of Hoeksema, 399 F.2d 269, 271-72 n. 6, 274-75 (C.C.P.A. 1968), unchallenged affidavits from experts declared that the chemical compound described in the application could not be made prior to the time of patent application, and thus, the prior art failed to contain sufficient enabling disclosures. Plaintiff in this case, on the other hand, failed to rebut the testimony of the defendant's experts that it was obvious to make a one piece gear unit, given the state of the existing skill and art.
As noted in Payne, it is not necessary that the prior art disclose precisely the same means of manufacture of the one piece gear unit as that employed for the current invention because the prior art is to be considered together with the ordinary skill possessed by practitioners in the field (including their general knowledge of methods of manufacture — in this case, metal-working methods). The enabling disclosure requirement has been only uncertainly and relatively recently imported into the analysis of obviousness claims, not as a matter of statutory construction but rather by way of analogy to the anticipation defense of 35 U.S.C. § 102. See Application of Hoeksema, 399 F.2d 269 (C.C.P.A. 1968). We therefore decline to give this requirement the overly rigorous interpretation urged by plaintiff.
Here it was the Patent Office which required Mooney to bifurcate his patent application for the one piece gear unit into two application — one addressing structure and the other method. The structure patent was directed to the problem of structure and solved it (albeit in an obvious fashion). The method patent in turn purported to, and did, solve the problem of method. It would be absurd under these circumstances to require that the prior art and publications, against which the structure patent was to be tested, should somehow solve the problem of method in detail — under the rubric of an "enabling disclosure."
To require that prior art or publications, to be invalidating as to a structure patent, contain an "enabling disclosure" giving the details of manufacture would quite improperly render many undeserving structure patents unobvious as a matter of law. It is enough that the prior art contain sufficient disclosure to enable a person of ordinary skill in the art to solve the problem of structure (and thus to be able to make and use the subject of the patent).
It is quite immaterial (even if it be a fact) that the [prior art] specification is insufficient to teach a manufacturer how to make the patented [unit]. It is enough if it sufficiently describes the [unit] itself. . . . The plaintiff claims a manufacture, not a mode of making it; and the important inquiry, therefore is, whether the prior publication described the article. To defeat a party suing for an infringement, it is sufficient to plead and prove that the thing patented to him had been patented or described in some printed publication prior to his supposed invention or discovery thereof.
See also General Electric Co. v. De Forest Radio Co., 17 F.2d 90 (D.Del. 1927), modified, 28 F.2d 641 (3rd Cir.), cert. denied, 278 U.S. 656, 49 S.Ct. 180, 73 L.Ed. 565 (1928).
(2) The Differences Between the Prior Art and the Claim at Issue
This step involves a factual comparison of the gear units described in the Kaufmann, Kiekhaefer and Gale Products references with the gear described in Claim 4. We have carefully evaluated each of these references and conclude, as did the district court, that the differences from the claimed invention are minimal. For example, the Kaufmann patent, like the Mooney unit, discloses an annular ring of beveled gear teeth with a recessed ring of clutch teeth. The major difference between the two gears involves the direction of the engaging faces of the ratchet teeth. In the Kaufmann patent, the faces of the teeth incline in an outward direction while in the Mooney patent, the faces on the gear teeth incline inwardly.
The Kiekhaefer drawing discloses another gear unit very similar to the unit described in Claim 4. Here again, both units have an annular ring of beveled gears with a recessed ring of ratchet clutch teeth which are concentric with and interior to the ring of beveled gears. The only significant difference between the two gear structures involves their construction: the Kiekhaefer gear is manufactured in two pieces and brazed together while the Mooney gear is manufactured in a single unit. Arguably this distinction may not differentiate the Kiekhaefer gear to the degree Mooney contends. Claim 4 does not clearly require that Mooney's gear be manufactured from a single piece of metal. It merely describes "a one piece integral gear." The unit disclosed in the Kiekhaefer drawing, while admittedly manufactured in two pieces, was sold and used as a one piece integral unit. Thus, the Kiekhaefer drawing plays a significant role in determining whether the gear structure was obvious in light of the prior art.
Finally, the Gale Products drawing also closely resembles the gear described in Mooney's patent. The Gale Products gear, used in the first outboard motors, was machined in one piece. The most important difference between the Gale Products gear structure and the Mooney invention involves the design of the clutch teeth. The teeth in the Gale Products drawing are flat at the top with no slope to their backs and no incline, while the teeth in the Mooney patent have both sloping backs and engaging faces which are slightly inclined. Although the Gale Products drawing shows only two teeth and the Mooney diagram has six, this distinction is not relevant because Claim 4 is not limited to the six teeth shown in the diagram.
(3) The Level of Ordinary Skill in the Pertinent Art
The level of ordinary skill in the art is generally defined through the testimony of individuals who are familiar with the art and the patent at issue. See Malsbary Manufacturing Co. v. ALD, Inc., 447 F.2d 809, 811 (7th Cir. 1971). In the instant case, the district court reviewed the statements and qualifications of the experts who testified for both parties at the trial and concluded that
the level of skill in the art of gear design was very high in 1962. Those of ordinary skill in the art would at least have [had] a degree in Mechanical Engineering. In addition, it is clear that . . . those skilled in the art of gear design would also be skilled in the art of milling or metal cutting. Furthermore, one skilled in the art would also be expected to have a good deal of practical experience . . . and would probably hold a number of patents which originated from their work.
Mooney, 489 F. Supp. at 560. After examining the record and the thoughtful analysis of the district court, we believe that the district court made a proper assessment of the level of skill in the art.
Once the court has established the scope and content of the prior art, the differences between the prior art and the claim at issue and the level of skill in the art, it must determine whether the invention was obvious to one skilled in the art at the time the patent was issued. Here, the district court noted that the prior art disclosed one and two piece gears which were virtually identical to the unit described in Claim 4 of Mooney's patent. Although Mooney's one piece gear unit was stronger and more durable than the gears referenced in the prior art, the court concluded that this condition merely reflected the value and importance of one piece construction. Mooney, 489 F. Supp. at 560-62. See Enterprise Railway Equipment Co. v. Pullman Standard Car Mfg. Co., 95 F.2d 17 (7th Cir. 1938). The record is also replete with the testimony of experts provided by both parties that it would be obvious, when faced with a weakness in the brazings of a two piece gear unit, to make the transition to a stronger one piece unit. As a result, the district court concluded that Claim 4 was invalid under 35 U.S.C. § 103, because it would have been obvious to one skilled in the art "to make a one piece bevel gear with ratchet clutch teeth." Mooney, 489 F. Supp. at 561.
Plaintiff argues, however, that the district court's conclusion is erroneous because the court applied the wrong test. According to Mooney the proper test is not whether it is obvious "to try to make" or "to make" the gear, but instead "whether or not the one piece recessed ratchet clutch and bevel tooth gear is obvious to one skilled in the art." App.Br. at 19. This argument is similar to the contention that the prior art must contain an enabling disclosure, and we have already addressed this objection. See note 9 supra. The conclusion that it would be obvious "to make" a one piece gear and the conclusion that the one piece structure of the gear (or the "one piece . . . gear") was obvious to one skilled in the art are one and the same. Mooney's point amounts to the proposition, which we reject, that obviousness must extend to method in detail as well as structure before the structure patent can be declared invalid under 35 U.S.C. § 103. This argument, which we have essentially addressed in connection with the invalidating effect of the prior art, is without merit. The district court thoroughly evaluated the evidence in the instant case and properly found that Claim 4 of the gear structure patent was obvious and therefore invalid.
Finally, plaintiff argues that the district court erred in concluding that the process used by Brunswick to manufacture its one piece ratchet clutch and bevel gear did not infringe on plaintiff's method patent. In order to prove infringement, the plaintiff must demonstrate that defendant engaged in the "unauthorized performance of substantially the same process steps in substantially the same way to accommodate substantially the same result." International Glass Co. v. United States, 408 F.2d 395, 400 (Ct.Cl. 1969). See Tilghman v. Proctor, 102 U.S. 707, 26 L.Ed. 279 (1881). It is not enough that the defendant produce the same result. Infringement occurs only when both the method and the result are substantially the same. See Linde Air Products Co. v. Graver Tank Manufacturing Co., 167 F.2d 531 (7th Cir. 1948), modified, 336 U.S. 271, 69 S.Ct. 535, 93 L.Ed. 672 (1949), adhered to on rehearing, 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097 (1950).
The district court concluded that defendant's method differed from the procedure outlined in Claim 1 in two substantial respects. First, the Brunswick method used forgings instead of metal stock and therefore did not involve "machining" the work piece into outline shape as described in Claim 1. Second, the Brunswick method did not involve the movement of the cutting tool "in one direction" and then "in another direction" relative to the work piece as specified in Mooney's patent. As a result, the district court found that Claim 1 was not infringed. Mooney contends that both of these findings are clearly erroneous.
The initial step in Claim 1 explicitly states that a work piece is formed by "machining metal stock to outline shape." Brunswick's method, however, does not include this step. The forging used by defendant is already in outline shape before any machining. Although some machining is required to finish off the rough edges of the raw forging, the district court determined that this machining was merely a finishing process which did not amount to "machining metal stock to outline shape" as described in Claim 1.
After examining samples of the plaintiff's metal stock and the defendant's raw forging, we conclude that the district court's finding in this regard is not clearly erroneous. Brunswick's raw forging has definite shape and form prior to any machining and therefore substantially differs from Mooney's metal stock, which requires considerable machining before reaching the same form as Brunswick's initial forged work piece. In addition, in the Brunswick method the delineation of the ratchet teeth in the forging enables the sloping surface or ramp of each ratchet tooth to be cut with the single radial cut of an end mill thereby reducing the amount of material that has to be removed during the milling operation. Thus, the forging permits Brunswick to achieve the same result as Mooney but in a substantially different, and in some respects superior, manner. Since Mooney's method patent is specifically limited to metal stock which is machined into outline shape, we agree with the district court that Brunswick's use of a forging precludes infringement on Claim 1 of Mooney's patent.
In his brief, Mooney suggests that if the preparation of the forging and the machining of the forging to outline shape is not the same as "machining metal stock to outline shape" then it is at least the equivalent within the scope of the patent laws. Since this argument was not raised by Mooney at trial and there is no supporting evidence in the record, it is not properly before us on this appeal.
B. Method of Cutting the Ratchet Teeth
Claim 1 of the method patent also states that the work piece is moved relative to the end cutting tool in one direction and subsequently the work piece is moved relative to the side cutting tool in another direction. The district court determined that Brunswick's method does not infringe Claim 1 because both cuts in the Brunswick work piece are made in the same radial direction. We have thoroughly examined both procedures and conclude that the district court's finding is not clearly erroneous.
In the milling art, there are three basic directions: (1) axial (movement along an axis); (2) circumferential (rotation around an axis); and (3) radial (movement out from the center of an object). Each of these directions only has meaning insofar as it relates to a work piece.
In the instant case, we must examine the terms of Mooney's method patent to determine which of these directions are involved in Mooney's process. In describing the procedure for cutting the sloping ramps of the ratchet teeth, the patent states:
The . . . work piece is held in a suitable jig at an angle with respect to the cutting tool . . . corresponding to the desired slope of the face surface of the ratchet teeth. The end cutting face of the tool . . . is brought into contact with the work piece and cuts the ratchet tooth surface by a combination of relative rotational movement of the cutter about its own axis and rotational movement of the work piece about its own axis through an arc dependent upon the desired number of ratchet teeth.
U.S. Patent No. 3,248,782, Col. 2, line 38-48. This description indicates that the sloping ratchet teeth are cut by rotating the cutting tool around its axis while rotating the work piece around its axis. This would be defined as circumferential movement. With respect to cutting the engaging faces of the clutch teeth, Mooney's patent states that the "cutter . . . and the work piece move relative to one another along a path lying generally along a radius of the work piece." This description specifies movement in a radial direction. Thus, Mooney's method patent calls for one circumferential cut to form the sloping ramps of the ratchet teeth, and one radial cut to form the engaging faces. In defendant's method, however, the tool used to cut the sloping surfaces of the ramp and the tool used to cut the engaging faces of the clutch both move in a radial direction. Since Mooney's patent is restricted to a cut in one direction ( i. e., circumferential) and a cut in another direction ( i. e., radial), Brunswick's process which performs both cuts in the same direction ( i. e., radial) does not infringe Mooney's claim.
Although Mooney admits that both tools in the Brunswick method move in a radial direction, he nevertheless maintains that the defendant's method infringes his claim because one cut is made in a vertical radial direction and the other cut is made in a substantially horizontal radial direction. We find no merit in this argument.
Mooney's definition of "direction" contradicts the established meaning of the word in the milling art as found by the district court. "Direction" must lie in relation to some established frame of reference in order to have meaning. In milling, that frame of reference is the work piece, and the three directions which have accepted meaning in reference to the work piece are axial, circumferential and radial. It would be possible to further define the precise direction of a tool within these three general directions in relation to a geometric grid, e. g. radial-horizontal or radial-vertical. Mooney's patent claim does not specify any such refinement, however, and we think the district court properly construed the term "direction" as used in the milling art.
Claim 1 is expressly limited to a method which cuts the sloping ramps of the ratchet teeth in one direction and the engaging faces of the clutch in another direction. The definition of direction is supplied by the prevailing practice and argot of the milling art. Under that practice Brunswick's method makes both cuts in the same direction and therefore does not infringe Claim 1 of Mooney's method patent.
Accordingly, the judgment of the district court is, in all respects, affirmed.