Plaintiffs Colortronic Reinhard & Company, K.G., and Colortronics, Inc. (hereinafter referred to collectively as "Colortronics") appeal from the district court's judgment, 496 F. Supp. 259, declaring Colortronics' patent, No. 3,814,296 (more particularly its Claims 1 and 5) invalid for obviousness under 35 U.S.C. § 103; its holding that defendant Plastic Controls, Inc. ("PCI") did not infringe on this patent; and the district court's award of attorneys' fees to PCI under 35 U.S.C. § 285. Defendant Plastic Controls, Inc. appeals from the district court's award of only nominal damages to PCI on PCI's counterclaim alleging unfair competition. Colortronics, Inc. is the United States representative of Colortronic Reinhard & Company of West Germany. The German parent company owns 50 percent of the American subsidiary. One Max Reinhard is the chief executive officer of the German company while Leo Meyer is president of Colortronics, Inc. Because the history of this case bears upon the issues we resolve, we begin with a review of the relationship between these parties.

PCI's principal officer and owner, Bruce A. McLean, was the exclusive U.S. sales and marketing agent for Colortronics from March 1974 until he was terminated in October 1975. In this capacity, he sold Colortronics' color blending machines-devices used in mixing coloring ingredients into kneadable plastics in precise, metered quantities. Colortronics had two U.S. patents-No. 3,985,345 ("Jakob") and No. 3,814,296 ("Tschritter")-covering these blending units. The Jakob patent covered the overall automatic mixing and feeding process of the blender. The Tschritter patent related to a very specific metering device used in the feeding apparatus.

McLean's termination in 1975 was sparked by an accusation that McLean was withholding receipts, although there may have been additional reasons for the break. Thereafter, the parties steered a collision course with each other. McLean immediately sent his customers a form letter assuring them that he would soon be marketing color blenders using "the same metering principles" as the Colortronics units. Colortronics, meanwhile, sued McLean to obtain some $7,000 to $8,000 of customers' payments he had allegedly withheld. McLean counterclaimed for allegedly withheld commissions.

In the winter of 1977, this lawsuit was finally tried to a jury resulting in a net award to McLean of approximately $30,000. Colortronics appealed. In August 1977, McLean met with Reinhard and other Colortronics employees and was offered, according to McLean, some $6,000 to settle the claim.1 At this meeting he was also told that the machines he was then marketing through PCI (he had made some four machines and was marketing them for about $700 each) were in violation of Colortronics' Tschritter and Jakob patents. As part of the proposed settlement, McLean was to admit the validity of these Colortronics patents. McLean refused this package, and Reinhard threatened to sue for infringement.

In November 1977, nothing further having developed on either side, Colortronics sued McLean and PCI for infringement of the two patents. Another settlement discussion then took place with Reinhard's offer now up to $12,000-$14,000. Once again, McLean was to admit the validity of the Colortronics patents as part of the deal, and again McLean refused. By this time or soon thereafter PCI effectively stopped doing business (it never produced more than the original four units and McLean had stopped drawing a salary as early as March 1975).

In late winter or early spring of 1978, the lawsuit over McLean's salary dispute was finally resolved with McLean collecting some $34,000 including interest. The patent dispute continued, however, though Colortronics once again offered to drop the suit if McLean would admit the validity of the patents. McLean once again refused and his deposition was taken in July 1978. At his deposition, McLean informed counsel for Colortronics that Peter Jakob, a former Reinhard employee who was the author of the blending patent being sued upon and who had helped design Colortronics' commercial metering units based on the Tschritter patent, had personal knowledge regarding the invalidity of the Jakob patent. While McLean actually had some materials in his possession at this time which could have added weight to his testimony, he said only that Jakob had proof and would bring it with him when he came from Germany to testify. Colortronics did not pursue this matter until the day before Jakob was deposed on March 12, 1980. At that deposition, Jakob revealed that the blender patented as No. 3,985,345 had been written up in literature and displayed in Germany as early as 1970 and 1971, and he produced such literature, some of which was written by Reinhard. He further stated that for this reason a German patent application had been rejected, the law in Germany having been at that time that no invention was patentable if it had been displayed more than six months before the patent application. He admitted that he had signed an oath on his 1973 U.S. patent application that his invention had not been on sale, in use or described in a printed publication anywhere in the world more than one year before the 1973 filing date.2 He agreed that this was a false statement when made and that, while he had signed the document, it was Reinhard, his employer, who had actually pursued the U.S. patent application.

The day before the deposition, on advice of counsel, Reinhard disclaimed the Jakob patent and dedicated the invention to the public. Colortronics dropped the Jakob patent aspect of its suit on the first day of trial by not opposing a motion for summary judgment on that patent. The trial then went on strictly on the basis of the Tschritter patent. After two days of live testimony and the introduction of the depositions of Peter Jakob and Max Reinhard, the court rendered the decision we now review.

The district court found that the Tschritter patent was invalid for obviousness under 35 U.S.C. § 103.3 On appeal, Colortronics asserts that this finding was clearly erroneous. Before proceeding to examine this claim, we pause to give a brief description of the Tschritter device itself. A drawing of the mechanism is appended to this opinion.

The Tschritter patent was issued for an apparatus to dispense in precise, metered quantities "kneadable" materials used in blending coloring agents into plastics. In Tschritter, the metering principle is embodied in two rotating geared discs placed at the bottom of a large hopper. The larger disc acts as a pan at the bottom of the hopper to accept the material to be metered. This disc has gear "teeth" placed at regular intervals around its perimeter. When material is introduced into the hopper, the material works its way between these teeth.

A smaller geared wheel is fixed within the larger disc. As the two wheels rotate against one another, the teeth of the smaller wheel mesh with the spaces between the teeth of the larger disc proximate to an exit slot in the hopper's base. The kneadable material between the teeth on the perimeter of the larger disc is thus forced out the opening and broken up by a grating just outside the exit slot.

Since varying amounts of the metered material might tend to be pushed upwards and back into the hopper space by the action of the meshing gears, Tschritter placed a sickle-shaped plate over the interlocking gear system to effect the compression necessary to force all of the material between the two affected perimeter teeth out of the exit slot. This compression of the material by the use of a lid or "coverplate" over the chamber formed by the meshing gears enabled Tschritter to define more exactly the amount of material which would be forced out of the hopper with each rotation of the gears. In granting the patent, the patent examiner made much of this compressive feature, stating that "(a)pplicant's device forms a closed metering chamber. The contents of the chamber are, for example, horizontally ejected through a comminuting grate shortly after the chamber is formed."

In reviewing a ruling of obviousness under section 103, we are guided by the Supreme Court's decision in Graham v. John Deere Co., 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966). In Graham, the Court indicated that obviousness depends upon a comparison between the patent at issue and the sum and content of the "prior art." If the advances claimed to be innovative in the patent would have been apparent to a person having ordinary skill in the particular art, then the patent is invalid. Id., at 17, 86 S.Ct., at 694.

At the outset, Colortronics is entitled by statute to a presumption that its patent is valid.4 Where patents involve merely the combination of known mechanical elements in a new configuration, however, the presumption of validity is often "not a heavy one." Futorian Manufacturing Corp. v. Dual Manufacturing & Engineering, 528 F.2d 941, 943 (1st Cir. 1976). This is so because, unlike experimental combinations in genetics or chemistry, the uniting of elements "old in mechanics" less frequently results in "some new quality or function." A & P Tea Co. v. Supermarket Corp., 340 U.S. 147, 152, 71 S. Ct. 127, 130, 95 L. Ed. 162 (1950). See also Sakraida v. Ag Pro, Inc., 425 U.S. 273, 96 S. Ct. 1532, 47 L. Ed. 2d 784 (1976) (device for flushing manure from cow barns found obvious in light of old and well-known technical components). The Tschritter patent, involving as it did merely an alleged new combination of gear wheels, a covering lid and a grating, was a classic example of a mechanical combination patent. As such, the patentee might reasonably anticipate that the presumption, standing alone, would be a slender reed against any significant evidence of obviousness.

In spite of this, Colortronics came to the trial without an expert who could explain how and why the claimed invention was a notable advance in the art. Early in the trial the district court unsurprisingly indicated its interest in hearing expert testimony, but counsel for Colortronics indicated that it chose to rest its case solely on the statutory presumption, and informed the court that it might address any questions regarding the patent to Mr. Leo Meyer-president of Colortronics, Inc. At the end of Mr. Meyer's testimony, the district court availed itself of this opportunity and inquired if Mr. Meyer was familiar with the "Walker" patent-a specific item of prior art which PCI had suggested was indistinguishable in principle from Tschritter. Mr. Meyer stated that he was not familiar with the Walker patent. The following colloquy then occurred:

THE COURT: .... Do you know how a gear pump works?

THE WITNESS: Yes.

THE COURT: Would you tell us?

THE WITNESS: It works by two gears rotating toward each other, and the oil is brought in and compressed and pressed out through a hole.

THE COURT: In what way does the 296 patent (Tschritter) differ from a gear pump?

THE WITNESS: The gears are placed in upwards, and the ejection disk is placed on the inside of the gear-by the gear pump, runs against each other.

THE COURT: That is a matter of visual inspection. What is the difference in the principle?

THE WITNESS: The principle is basically similar.

THE COURT: Thank you. That is all.

The district court later wrote in its opinion that the Tschritter patent, by Mr. Meyer's own admission, merely put to a new use the venerable principle of a common gear pump-a device described by Mr. Meyer himself. It went on to hold the patent invalid for obviousness and lack of novelty. Colortronics now asserts that this was error. We disagree.

A common gear pump was not "unknown prior art" as alleged by Colortronics. Mr. Meyer-Colortronics' own president-described the mechanics of this type of pump and stated that it was similar in principle to the Tschritter patent. There was no objection to the observation and no subsequent qualification or withdrawal of it by Mr. Meyer. Meyer's testimony thus stands as evidence, and the gear pump principle he described could properly be assumed by the district court to have been a part of "the knowledge that would have been available to any person having ordinary skill in the art." Steelcase, Inc. v. Delwood Furniture Co., 578 F.2d 74 (5th Cir. 1978), cert. denied, 440 U.S. 960, 99 S. Ct. 1503, 59 L. Ed. 2d 774 (1979). Since Tschritter, by Meyer's admission, represented no material advance in principle on the ordinary gear pump Meyer described, the district court could take this into account in ruling that the patent was invalid.

Moreover, the admission regarding Tschritter's relation to a common gear pump was by no means the only evidence of obviousness before the court. PCI put in substantial evidence from which it was reasonable to conclude that Tschritter involved no more than an obvious application of principles embodied in other specific prior art-in particular, U.S. Patent No. 1,664,802, granted April 3, 1928 ("Walker"). The district court referred to the Walker patent in its opinion and was plainly well aware of its relation to Tschritter.5

The Walker patent was issued for an apparatus to dispense free-flowing powders such as fertilizers. In Walker, the metering principle was embodied in a rotating, geared disc placed at the bottom of a large hopper. A smaller, rotating geared wheel was fixed by a raised support arm within the larger disc so that the gears of the two wheels meshed proximate to an exit slot in the hopper base. As the two geared wheels turned, the powder was directed at the opening by the meshing gears. A grating near the opening was used to regulate the quantity metered out in each rotation.

Even at first glance, the Tschritter apparatus bears a marked resemblance to the Walker device. Tschritter's mechanism, like Walker's, consists of an interlocking set of gears placed at the base of a large hopper. The expulsion of material is achieved, as in Walker, by the rotation of these gears near a strategically placed exit slot in the hopper base. Unlike Walker, however, the gears in the Tschritter device are of equal height, the material metered is divided by a grate outside the perimeter of the larger disc, and the inner gear wheel in Tschritter is supported by a stationary "coverplate" which serves to compress the material being metered. This coverplate extends in a curving fashion over several of the spaces formed between the gear teeth around the edge of the outer disc, including the space aligned with the exit slot. According to Tschritter patent Claim 1, the device is thus characterized by "cogs (or outer gears) extending to the coverplate, their height being equal to the height of the teeth of the ejection disc (inner wheel) whereby during the metering a closed metering chamber is formed." Claim 5, also at issue here, relates to the device as described in Claim 1 with a "comminuting grate" placed outside the exit slot to break up the metered material as it comes out.

PCI produced evidence which was unrebutted by Colortronics that the Tschritter claims were unremarkable, indeed unnecessary, advances on the Walker mechanism, given the task of metering kneadable plastic. Peter Jakob, who worked with Tschritter himself, who redesigned the Tschritter machine to make it more practical and marketable, and who, when he left Reinhard's employ, was the top technical man in the company, testified that there were only two differences between the Walker and Tschritter patents, both of which "have nothing to do with the basic function of the equipment." These differences were the grating over the exit slot and "closed chamber" formed by the coverplate. Jakob testified it was "not an important point, to build this chamber." In addition, in a letter to Bruce McLean, which was introduced by Colortronics, Jakob stated that, in making the commercial adaptations of the Tschritter patent, "we always had exactly what Walker already described in 1928." This view was echoed by Bruce McLean, who stated that both the coverplate used to form the metering chamber and the outside grate made "no difference" and were of "no value" to the overall design of the metering station.6 Thus, the total improvement over the Walker apparatus embodied in Tschritter would seem to have been a lowering of the arm supporting the inner gear wheel so that it would help to contain the natural tendency of kneadable material to push upward as the gears meshed.7

Finally, we note Colortronics' failure to produce an expert, and the lack, therefore, of any testimonial evidence to support the validity of the patent and rebut PCI's evidence of invalidity.8 While the failure to put on an expert is not itself independent evidence of invalidity, a court is obviously entitled to take into account the weakness of one side's evidence in contrast to the other's.9

The district court could properly conclude, then, that PCI overcame Colortronics' presumption of validity, having shown that the Tschritter patent was but an obvious application of the prior art. Moreover, despite its somewhat elliptical nature, we think the district court's opinion adequately signals a proper application of the Graham inquiries and the relevant issues. No more is required. Mollura v. Miller, 609 F.2d 381 (9th Cir. 1979), cert. denied, 446 U.S. 918, 100 S. Ct. 1852, 64 L. Ed. 2d 272 (1980). We therefore affirm the district court's ruling of invalidity for obviousness. Having found the patent to be invalid, we need not discuss the district court's independent finding that the PCI equipment did not infringe on the Tschritter patent.

The district court found that this case was "exceptional" under 35 U.S.C. § 28510 and awarded attorneys' fees to the prevailing party, PCI. Such an award, while unusual, is within the sound discretion of the trial judge in patent cases, Contour Saws, Inc. v. L.S. Starrett Co., 444 F.2d 1331 (1st Cir. 1971), provided that the district court's discretion was not abused. Attorneys' fees are normally awarded under section 285 only where the district court finds strong evidence of unfairness and bad faith on the part of the losing party. Plymouth Rubber Co. v. Minnesota Mining & Manufacturing Co., 203 F. Supp. 595 (D.Mass.1962), aff'd, 321 F.2d 151 (1st Cir. 1963), cert. denied, 375 U.S. 969, 84 S. Ct. 489, 11 L. Ed. 2d 417 (1964).

With respect to the Jakob patent ('345), the court found that it was obtained by committing knowing fraud upon the patent office. We quote the findings of the district court on this matter:

In his 1974 application for the '345 patent Jakob, an employee of Colortronic, subscribed to the usual oath to the effect that the "invention (had not been) described in any printed publication in any country ... or on sale in the United States more than one year prior to this application." In point of fact, Jakob himself had written two leaflets, one a very detailed description of the device, which had been extensively distributed at the Hanover Fair in 1970. He at least participated in publishing several articles in German technical journals the year following. One had been written by Reinhard. In addition, after being on sale in Germany in 1970, the '345 device was put on sale at the National Plastics Exposition in Chicago in 1971. Both Jakob and Reinhard were present.

With this background Colortronic prosecuted the application that Jakob had signed. Colortronic was anything but a neophyte in such matters. I am abundantly persuaded that Jakob's oath was false both to his and Reinhard's, and hence to Colortronic's, knowledge at the time it was made, and was falsified for the purpose of obtaining the patent. Colortronic disclaimed on the eve of trial on advice of counsel and admitted the invalidity of the patent not because it had just learned the true facts, but because it had learned that Jakob was going to testify and counsel appreciated the predicament.

We find that the above facts, even were they not sufficient to support a finding of actual fraud, would be sufficient to demonstrate that Colortronics showed a reckless disregard for the integrity of the patent application process which would be enough to make this case "exceptional." Monolith Portland Midwest Co. v. Kaiser Aluminum & Chemical Corp., 407 F.2d 288 (9th Cir. 1969). To sue and continue to press suit on such a patent is manifestly an abuse of the protection provided by the patent laws and is a clear demonstration of bad faith.11

With respect to the Tschritter patent, the issue is somewhat more troublesome. The district court awarded attorneys' fees for this aspect of the case after it found that Colortronics "must have known" that the PCI device did not infringe on the Tschritter patent. It apparently based this conclusion on the fact that the PCI apparatus-like the Colortronics commercial units-did not have the full sickle-shaped coverplate depicted in the patent and therefore displayed only a "partially closed" metering chamber, as opposed to the "closed" metering chamber of Tschritter. This latter departure from the specifications of the patent could arguably have estopped Colortronics from claiming infringement in a normal case by operation of the doctrine of "file wrapper estoppel." Graham v. John Deere Co., 383 U.S. at 33, 86 S.Ct. at 701.

While the history of this litigation gives some indirect grounds for inferring that the Tschritter aspect of it, like the Jakob, was pursued in bad faith,12 we reluctantly conclude that the district court erred in finding that Colortronics must have known that PCI was not infringing before the trial in this case. Absent some disabling factor not here present, such as fraud in the procurement of the patent, Colortronics was entitled to presume that the Tschritter patent was valid and to seek its enforcement against those who it believed in good faith were infringing it. While it is true that both the Colortronics commercial unit designed by Jakob and PCI's alleged "Chinese copy" varied somewhat in design from Tschritter, they varied only by virtue of their narrower traverse bar and the fact that this bar was raised from the gears below it by a few millimeters. The function and result of the bar, however, was at least arguably identical with the coverplate of Tschritter. Both the bar and the coverplate had a "compressing" effect. McLean admitted that the placement of the bar over the gears improved the mechanism's metering capabilities, and Meyer reported that the bar was raised on Colortronics' commercial units only to prevent friction between the coverplate and the gears-while presumably preserving the compression concept embodied in Tschritter. We note that the legal test for equivalence in patent cases requires only that the offending unit perform substantially the same work in substantially the same manner to achieve substantially the same result. Borg-Warner Corp. v. Paragon Gear Works, Inc., 355 F.2d 400, 404 (1st Cir. 1965), cert. dismissed, 384 U.S. 935, 86 S. Ct. 1461, 16 L. Ed. 2d 536 (1966). We think that Colortronics, based upon the similarity between PCI's units and Colortronics' commercial units, could have believed in good faith that infringement had occurred because, functionally, the partially closed chamber of their own commercial units performed substantially the same task as the closed chamber of Tschritter. It was at least a valid legal question as to whether the technical departures from the exact specifications of Tschritter's Claim 1 would have estopped Colortronics from successfully asserting this equivalence. Without more evidence to suggest that Colortronics actually knew that its own units or the PCI device were not covered by Tschritter, we do not think the requisite showing of bad faith has been made.13 We therefore remand on the issue of attorneys' fees alone so that the district court may properly deduct from its award the amount it allocated to the Tschritter patent. General Instrument Corp. v. Hughes Aircraft Co., 399 F.2d 373 (1st Cir. 1968).14

As for PCI's claim that the district court erred in awarding PCI merely nominal damages on its common law claim of unfair competition, we agree with the district court that PCI's evidence on damages was simply too speculative to support any greater award.15 The court did not, as PCI asserts, find that PCI was not a business at all. Nor can we find any cases suggesting that PCI is entitled to compensatory damages as a matter of right because of the finding of fraud on the Patent Office, however relevant that factor is on the issue of attorneys' fees.

The district court found, with record support, that McLean had no manufacturing expertise and that his proposed stock purchasers for PCI were creditors who could have provided only a portion of the capital needed to get PCI's manufacturing operation off the ground. Issues such as this, which hinge in part upon the trial court's judgment regarding credibility as well as its review of the written record, are normally left to the sound discretion of the district judge. Fed.R.Civ.P. 52(a). We see no reason to disturb the court's findings here. Rivera Morales v. Benitez de Rexach, 541 F.2d 882, 886 (1st Cir. 1976), overruled on other grounds, Aufiero v. Clarke, 639 F.2d 49 (1st Cir.), cert. denied, --- U.S. ----, 101 S. Ct. 3052, 69 L. Ed. 2d 421 (1981).

Affirmed in part; remanded in part.

Drawings from the Tschritter Patent,

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE