Champion Spark Plug Company, Plaintiff-appellee, v. the Gyromat Corporation, Defendant-appellant, 603 F.2d 361 (2d Cir. 1979)

This appeal is from the decision and judgment (unreported) of the District Court for the District of Connecticut declaring invalid and unenforceable claims 5 and 6 of U. S. Patent No. 3,219,276 ("Plural Nozzles Having Intersecting Spray and Control Therefor"), issued November 23, 1965,¹  to Edward O. Norris, whose rights to the patent were acquired by appellant from the Norris estate following his death in 1968.²  We reverse.

The Invention

The invention relates to fluid (E. g. paint) spraying devices. The essential features, insofar as the involved claims are concerned, are illustrated in Figure 1 of the patent. Devices 29 are spray gun nozzles which issue a fine spray of highly charged particles (E. g. paint) toward a workpiece 44 to be sprayed. In order to spray the surface of a large workpiece, electrostatic spray guns 27 are moved up and down by means of a reciprocator slide 15. As they move up and down (in relation to a stationary source of spray fluid), the pressure of the spray from the guns varies with the height of the guns, there being a loss of pressure when the reciprocator slide moves the guns upward and a gain in pressure when it moves the guns downward. This occurs because of the change in height of a vertical column of the spray fluid. For example, if the stationary source of spray fluid is at floor level, the fluid must travel upward to reach the guns; and it must move upward a greater distance when the guns are at the top of their stroke than when the guns are at the bottom of their stroke. This greater upward travel of the fluid results in a greater weight of fluid pressing downward in opposition to the pressure from the stationary source of the fluid.³  Without more, a loss in pressure would cause less fluid (E. g. paint) to be applied to the workpiece when the guns are in their highest position and more to be applied to the workpiece when the guns are in their lowest position, with variations in between.

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

The problem of pressure variation with the changing vertical movement is solved by adjustable, fluid pressure regulators 37 ("control valves"), which are mounted on the reciprocator slide 15 and move up and down with the spray guns. There is one regulator for each gun. The fluid pressure to be delivered to the guns is determined by adjustment of the pressure regulators, also known as "primary regulators." Because these pressure regulators and the guns move up and down together, the fluid pressure Between the regulators and the guns does not vary during vertical reciprocation. Although the fluid pressure from the stationary source to the Input side of the regulators 37 varies during vertical reciprocation, the regulators compensate for such variation, providing a steady pressure at the Output side for delivery of the fluid to the spray guns.

A standard fluid pressure regulator is adjusted by a manually-controlled knob, thumb screw, or similar device. However, when the apparatus is in operation, such adjustment "on-the-fly" (I.e., without stopping the reciprocating spraying apparatus) is difficult.⁴  To meet this problem, the Norris '276 patent replaces the manual control of the primary regulator with a stationary, remote, secondary Air pressure regulator (202a). This regulates the pressure of air that flows to the primary regulator through flexible plastic tube 192. By adjusting the air pressure, the variable output of the vertically moving primary regulator is controlled, and the control is precise because the weight of the air in tube 192 remains practically constant while the tube moves up and down with the reciprocator slide.

The Claims

The involved claims 5 and 6 refer to claims 3 and 4, which were disclaimed by appellant Gyromat in earlier litigation involving the '276 patent (Gyromat Corp. v. H. G. Fischer & Co., 167 U.S.P.Q. 326 (N.D. Ill. 1970)) and are reproduced in the footnote below.⁵  Claims 5 and 6 read as follows:

5. Apparatus as set forth in claim 4 in which said adjusting means comprises a pressure actuated diaphragm and means including a remote pressure control valve connected to control the pressure applied to said diaphragm for thereby varying the pressure supplied by said first control valve to said nozzle.

6. Apparatus as set forth in claim 5 in which said pressure control valve is connected to said first control valve by a flexible coupling having insulating characteristics and an electrostatic potential is supplied to said nozzle for charging the spray material ejected therefrom.

Prior Art

All but two of the essential features of the invention described above are to be found in claims 3 and 4. These features have been disclaimed by appellant, and the parties agree that they may be considered prior art; also appellant does not dispute the district court's findings that the flexible coupling in claim 6 is old in the art. As to a stationary, remote, secondary (air) pressure regulator controlling pressure in a primary regulator that "reciprocates through a vertical plane," the district court found that this was not disclosed by the prior art. With respect to this feature, the district court commented that "the combination of a self-relieving pressure regulator connected to a Stationary fluid regulator" is "well known." (Emphasis added.)⁶ Proceedings Below

The district court based its determination that claims 5 and 6 were invalid on its holding that the subject matter of the claims was obvious for purposes of 35 U.S.C. § 103.⁷  As we have seen, the critical subject matter of both claims is the stationary, remote, secondary pressure regulator controlling pressure in a primary regulator that "reciprocates through a vertical plane."

In declining to follow Judge Hoffman's holding of validity of claims 5 and 6 in Gyromat Corp. v. H. G. Fischer & Co., supra, the district court noted that two of the remotely-controlled regulators cited in Fischer contained no bleed holes and that "the third was clearly not suitable for a paint spraying device since the bleeding means was placed in a portion of the system that corresponded to the paint line of a spray paint device"; also that Judge Hoffman had rested his conclusion of nonobviousness "in part" on the bleed device disclosed in the Norris '276 patent in stating:

The provision of a manually controlled air pressure regulator, with a continuously bled regulated output, for effecting the precise control of paint pressure in a reciprocating spray apparatus was not "obvious" within the meaning of 35 U.S.C. § 103.

167 U.S.P.Q. at 341. The court below went on to point out that Champion had cited prior art references which the parties agreed disclosed bleed holes in pressure regulating systems. It then said:

Thus the issue is whether the combination of two components, a self-relieving pressure regulator and a reciprocating fluid regulator, was obvious within the meaning of § 103 of the Patent Act of 1952. Each component was well-known. Also well-known was the combination of a self-relieving pressure regulator connected to a stationary fluid regulator. Was it "obvious" to decide to connect a self-relieving pressure regular to a (vertically) reciprocating fluid regulator?

Having analyzed the scope and content of the prior art and the difference between the prior art and the subject matter of claims 5 and 6, the district court considered the level of skill of a person of ordinary skill in the art to which said subject matter pertained as of 1962 (the time when the Norris invention was made). Graham v. John Deere Co., 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545, 148 U.S.P.Q. 459 (1966). It found that the pertinent art was "the field of hydraulics and fluid controls," and that the testimony of Champion's expert, Russell Henke, "is persuasive that in 1962 the idea of connecting a pressure regulator to a (vertically) reciprocating fluid regulator required no more than the ordinary skill of a mechanic reasonably knowledgeable in the field of hydraulics and fluid controls."⁸ 

Stating that the benefit of hindsight is to be discounted,⁹  the district court concluded as follows:

While the combination of components achieved a useful result, it was the same result to be expected whenever a fluid regulator is remotely controlled by a pressure regulator. The fact that the fluid regulator is (vertically) reciprocating presented no special problem either in deciding to use a remotely controlled pressure regulator or selecting means to accomplish its connection. . . .

. . . That the fluid regulator in the patent reciprocates while the secondary pressure regulator remains stationary is not, on the evidence presented, a sufficiently different change in the context in which the combination of components operates to be considered a new function or a non-obvious improvement.

In responding to Gyromat's argument that Champion failed to come up with the idea of attaching a remote pressure regulator to a vertically reciprocating fluid regulator until after its personnel had seen such an arrangement constructed by the Fischer Company,¹⁰  the district court said:

There is no evidence that plaintiff's (Champion's) staff was using a traveling fluid regulator but was unable to come up with a device for remotely controlling it.

Rather, the district court emphasized that Champion had rejected traveling fluid regulators for reasons unrelated to the problem of fluid adjustment.

Finally, the court, recognizing that the district court in Fischer "was understandably impressed with the comparison between the benefits of the '276 patent and the lack of success of Fischer personnel," stated that it had reached the opposite conclusion on the basis of "augmented evidence presented here . . . of pertinent prior art disclosing continuously bled pressure regulators and persuasive testimony concerning the obviousness of remotely controlling a fluid regulator (whether stationary or reciprocating) with a pressure regulator. . . ."

OPINION

Presumption of Validity

Before the district court, Champion stated that "the prior art evidence to be produced . . . was not considered by the Patent Office and the presumption (under 35 U.S.C. § 282) is thus undercut or substantially weakened." However, the district court did not mention the statutory presumption of validity of the Norris '276 patent.¹¹  Instead, it disposed of the holding of validity of claims 5 and 6 in Gyromat Corp. v. H. G. Fischer & Co., supra, on which Gyromat has heavily relied. In refusing to follow Fischer, the district court said it was persuaded by Champion's "augmented evidence" of prior art bleed-type pressure regulators not in evidence in Fischer. However, the relevance of such evidence is not apparent, because the bleed feature of the secondary regulator in the Norris '276 patent is not even claimed. To the extent that Judge Hoffman in Fischer referred to the bleed feature of the secondary regulator, this, too, would lack relevance. Nevertheless, other evidence in Fischer was relevant and clearly supports Judge Hoffman's conclusion of nonobviousness. Thus, after noting that H. G. Fischer's first commercial use of such a primary regulator was in 1959, but that the remote, secondary (air) pressure regulator was not reduced to practice (by Norris) until late December of 1961, Judge Hoffman said (167 U.S.P.Q. at 341):

During the intervening period of almost three years, it was necessary to shut down the paint spray equipment in order to make an adjustment in the paint spray pressure for most installations. Moreover, Mr. Walberg (assignor of rights in certain patent applications to H. G. Fischer & Company to whom the patents were issued) admitted that there were occasions when it was necessary to adjust the paint pressure regulator of the defendant's equipment by jabbing or poking at the handle of the reciprocating regulator with an object such as a broomstick. This experience of the defendant is wholly inconsistent with its argument that the remote control of a paint pressure regulator would have been obvious to one skilled in the art.

Although we agree with the district court that the holding in Fischer is not controlling in this case, See Jamesbury Corp. v. Litton Industries, Inc., 586 F.2d 917, 920-21, n. 9 (2d Cir. 1978),¹²  we conclude that the statutory presumption of validity of claims 5 and 6 of the Norris '276 patent still stands. Thus, the dispositive question is whether this presumption of validity has been rebutted by Champion. See Solder Removal Co. v. United States International Trade Commission, 582 F.2d 628, 632, 199 U.S.P.Q. 129, 133 (CCPA 1978).

Champion's Expert Testimony Insufficient

As earlier pointed out, the district court, in determining the level of skill (in 1962) of a person of ordinary skill in the pertinent art of hydraulics and fluid controls, found the testimony of Champion's expert, Russell Henke, to be persuasive. Mr. Henke, an engineer experienced in hydraulics and fluid control systems, testified that in 1960 or 1961 there were commercially available both manual controlled fluid pressure regulators and remote operated air pressure regulators, and the concept of remote control by air pressure of a fluid pressure regulator was well known. It was his belief that those possessing skill in the art such as he possessed would have had the knowledge and know-how to connect pilot operated regulators for liquid service with a pilot regulator operating on air.¹³  He also testified

It's my opinion that anybody who would have that knowledge of the field similar to what I might have, would be led to using a remote pilot controlled fluid regulator to solve the problem.

. . . .

It's my feeling that anyone skilled in the prior art we have been discussing would be led to the same approach of applying a remote pilot control on a fluid regulator to solve this problem.

Although, as a general rule, the weight of an expert's testimony is to be determined by the trial court, such a determination may, nevertheless, be shown to have been erroneous. Universal Athletic Sales Co. v. American Gym, Recreational & Athletic Equipment Corp., 546 F.2d 530, 537, 192 U.S.P.Q. 193 (3d Cir. 1976), Cert. denied, 430 U.S. 984, 97 S. Ct. 1681, 52 L. Ed. 2d 378 (1977). We note other factors which detract from the weight of Henke's testimony. He had no experience with paint spraying devices, with which the "on-the-fly" adjustment problem was associated and to which problem the Norris '276 patent was addressed. Moreover, his testimony was often vague. The witness did not define "the problem"; nor did he indicate what (other than the hindsight obtained from the Norris invention) would have led "a person having ordinary skill in the art" of hydraulics and fluid control systems to the "on-the-fly" solution achieved by Norris with respect to a vertically reciprocating fluid pressure regulator. On cross-examination, Mr. Henke was asked whether he had occasion to work with equipment in which there was a vertical movement between a controlled regulator and a controlling regulator. His response was as follows:

A. I suspect there must be instances where the motion was vertical as opposed to horizontal or rotary.

Q. What type of thing are you thinking about when you say that?

A. Well, mostly machine motion I don't have specific examples in mind, but I have worked in general field of machines that are fluid operated or have fluid systems on them and there are motions in these machines, and they could be anything. It could be automation systems, it could be machine tools, they could be different kinds of equipment, construction equipment, agricultural equipment, industrial equipment, in general.

Q. This would involve the use of a vertically reciprocating regulator that was

A. In some instances, it could, yes.

Q. In some instances? Can you give us a specific example?

A. Not off the top of my head, I can't, no.

Thus, although he had testified that the concept of remote control by air pressure of a fluid pressure regulator was well known in 1960 or 1961, he could not, on November 16, 1976 (date of testimony), think of a single example of his having worked with a Vertically reciprocating fluid pressure regulator. As we have seen, however, the "on-the-fly" problem Norris solved arose only from prior art vertically reciprocating fluid pressure regulators.¹⁴  Thus, we conclude that the testimony of Mr. Henke was clearly insufficient to support the district court's finding that in 1962 the idea or concept of connecting a pressure regulator to a (vertically) reciprocating fluid regulator required no more than the ordinary skill of a mechanic reasonably knowledgeable in the field of hydraulics and fluid controls.

District Court's Finding Clearly Erroneous

Bearing crucially on a determination of the level of skill of a person of ordinary skill in the art (hydraulics and fluid controls) is the district court's finding, quoted earlier, that there is no evidence that Champion's staff was using a traveling fluid regulator but was unable to come up with a device for remotely controlling it. This was in response to Gyromat's argument, pursued on appeal, that persons of ordinary skill in the art (namely, the research staff of DeVilbiss Company, a division of Champion) were not able to solve the problem of "on-the-fly" adjustment of vertically reciprocating fluid pressure regulators. The following evidence supports Gyromat's argument and contradicts the district court's finding:

1. Deposition testimony of William D. Gauthier, Director of Research and Engineering of DeVilbiss. He had been employed by DeVilbiss since January of 1963, first as Assistant Director of Research, and had been involved in, among other things, work on paint pumps, reciprocators, and gun mounts, with work on a short-stroke reciprocator starting in early 1963.

Q. In the course of developing a control system for the reciprocating equipment that was being developed by your group, were various arrangements considered and evaluated for effecting control of paint pressure at the gun at the spray gun from a remote location, or from any location, for that matter?

A. My recollection is rather hazy in that area. I believe I recall cases where fluid regulators were mounted on the reciprocator, and some mechanical linkage was evolved to provide the adjustment.

. . . .

Q. Do you remember what it looks like or what it looked like, rather?

A. My recollection was just some insulating rods . . . with universal joints at either end connected from the reciprocator to a remote location.

Q. Was equipment ever built and delivered commercially using such an arrangement for control of paint pressure?

A. No.

2. Deposition testimony of Robert J. Della Flora, a Senior Project Engineer in the Research Group of DeVilbiss. He had been employed by DeVilbiss since April of 1960, first as a Project Engineer, and had been involved in, among other things, paint pumps and some work on control systems for reciprocators and associated equipment and some work on guns.

Q. Did your work on the reciprocator commence at around the time that Mr. Gauthier came with the company?

A. Yes.

. . . .

Q. What does this report, or what does this document marked as Defendant's Exhibit 20 represent, Mr. Della Flora?¹⁵ 

A. A report on possible methods of controlling paint flow. . . .

Q. To whom did you hand the report in, do you recall?

. . . .

A. I believe it went to Mr. Gauthier.

. . . .

Q. Did this work that is reflected in document Defendant's Exhibit 20, relate to the very first prototype reciprocator that you worked on?

. . . .

A. It was a separate item. It didn't refer to the reciprocator.

Q. Oh, this was not work with respect to a reciprocator?

A. No.

Q. I take it then that it was limited to known reciprocating equipment?A. It was paint feed control, period. Reciprocated or stationary, no difference.

Q. In other words, it was without regard to whether it was reciprocated or not?

A. Right.

. . . .

Q. Did the investigations that are reflected in Defendant's Exhibit 20 result in any changes in the design of the AGC (paint system using an air atomizing concept) gun, or the paint control for the AGC gun?

A. I think it all developed concurrently.

. . . .

Q. Thank you. Was this paint feed control incorporated in the AGC equipment?

A. No.

. . . .

Q. You discussed . . . the ultimate paint control system that was finally adopted in the AGC system, and I believe you also testified that you did not have anything directly to do with the development of that ultimate system. Do you have any personal knowledge as to where the concept or the genesis of that remote control fluid regulator came from?

A. All of this was a group effort. The fluid regulator portion was a regulator that we had in our line. The remote air control of this regulator was developed to satisfy a request by the Research Director, Mr. Juvinall to make this unit remotely adjustable without having to turn off the electrical power or stop the reciprocator.

. . . .

Q. Did Mr. Juvinall's request then immediately result in the ultimate system being developed, or did this take some period of time?

A. Today it seems as though it was a step right onto it, but I can't recall how long it did take.

Q. Were any other systems attempted or tried and discarded following this request by Mr. Juvinall and prior to the development of the ultimate system? In other words, were there any false steps along the way, or did the group just go out and put this thing together?

A. I know there were some false starts made, but I can't recall how many or how long they took.

From the foregoing testimony of Mr. Gauthier and Mr. Della Flora, it is clear that the research group at DeVilbiss, commencing early in 1963, was working on development of vertically reciprocating equipment with fluid regulators mounted on the reciprocator and on development of a paint feed control system for that equipment; that development of the paint feed control system, covered by Della Flora's report of 10/23/63, was carried out without recognition of the "on-the-fly" control problem, much less a solution to it, since it was without regard to any difference between reciprocating and stationary equipment; that the results of Della Flora's investigations were not incorporated into the AGC equipment or commercially developed; that it was not until Juvinall requested that the fluid regulators be adjustable "on-the-fly" that development of remote air control of those regulators was undertaken; and that, even then, "there were some false starts."

Providing still greater support to Gyromat's argument regarding the level of skill of a person of ordinary skill in the pertinent art is a report dated 8/31/64¹⁶  by Erherd Kock, Director of Research and Engineering for DeVilbiss, who, like Gauthier, had been employed by DeVilbiss since January of 1963, first as a senior research engineer, and had been continuously involved in development of paint spray equipment. The report states that the objective of Project No. 39 ("Paint Feed Control") was to develop a suitable means for controlling, with precision and reliability, the paint flow rate for the new DeVilbiss electrostatic system, which consisted of an array of vertically reciprocating spray guns; that it was desirable for the paint flow to be readily adjustable while the system was in operation, I. e., while the guns were at high voltage and reciprocating; further:

In the early consideration (¹⁷  of this problem, fluid regulators had been rejected because of their bulk, cost and incompatibility with quick color change. Further, the use of fluid regulators did not seem to offer an easy solution to the requirement of remote on-the-fly adjustment, because in order to minimize fluid flow fluctuation as the height of the vertically reciprocating gun changed, it would have been necessary to reciprocate the fluid regulators with the guns.

A second approach, consisting of controlling the flow by varying the fluid needle opening was investigated at greater length. It had the obvious advantage of simplicity and low cost, but was finally rejected because it was difficult to control individually on-the-fly . . . .

The fluid control method that was finally selected consisted of a modification of the small "gun-mounted fluid regulator" # 44167-008 (which had just become commercially available), the diaphragm pressure of which was adjusted remotely by means of an air pressure regulator. . . . The fluid regulator itself was small enough that it could be mounted on the reciprocator adjacent each gun without creating a weight problem, and it was sufficiently cheap, that one regulator could be supplied for each gun without unduly increasing the cost of the system. The only relatively expensive component was the air regulator (a Norgren regulator was found to give the best performance), one of which had to be used in conjunction with each fluid regulator. However, the resulting ease of remote paint feed control for both conductive and nonconductive paints while the system was in operation, facilitated by the fact that the operator could now select and maintain a given paint flow by merely adjusting the air pressure regulator . . . seemed to be well worth the additional expense.

Thus, a contemporaneous report shows that the DeVilbiss research group found the requirement for remote "on-the-fly" adjustment to not be easily satisfied when using fluid regulators because they had to be reciprocated with the guns; also, that varying the fluid needle valve opening of the gun (recommended as "practical" by Della Flora, note 16 Supra) was not satisfactory because of the difficulty of "on-the-fly" control. The solution that was eventually selected was made after Gauthier and Kock had (in 1963) viewed an embodiment of the Norris invention constructed by the Fischer Company.¹⁸ 

This report, along with the above-quoted testimony of Gauthier and Della Flora, compels the conclusion that the district court's finding, that there is no evidence that Champion's staff was unable to come up with a device for remotely controlling a traveling regulator, is clearly erroneous.¹⁹  Indeed, the evidence is compellingly to the contrary. Although early in 1963 the DeVilbiss research group commenced work on a paint feed control system for vertically reciprocating equipment, it was not until a year and a half later that Kock reported a satisfactory solution to the "on-the-fly" control problem, and that solution, which is clearly an embodiment of the Norris invention, was not approved until some nine months later. Accordingly, Champion's argument that the level of skill of a person of ordinary skill in the pertinent art in late December 1961 was such that the Norris invention would have been obvious must fail.

Combination of Old Elements Not Obvious Per Se

Citing Sakraida v. Ag Pro, Inc., 425 U.S. 273, 96 S. Ct. 1532, 47 L. Ed. 2d 784 (1976), Champion argues that claims 5 and 6 of the Norris '276 patent simply define "an arrangement of old elements," each performing "the same function it had been known to perform," and that such combinations are not patentable. In the factual setting of the Sakraida case, we have no difficulty with the holding that the invention there involved was not patentable. However, we do not agree with what amounts to an oblique suggestion that the dicta in the Supreme Court's opinion overruled the statutory test of nonobviousness established by 35 U.S.C. § 103 along with the analytical guidelines for that test established by the Court in Graham v. John Deere Co., supra, which the opinion in Sakraida cites with approval. See Republic Industries, Inc. v. Schlage Lock Co., 592 F.2d 963, 970, 200 U.S.P.Q. 769, 777 (7th Cir. 1979). Most, if not all, inventions involve a combination of old or known elements. Shaw v. E.B. & A.C. Whiting Co., 417 F.2d 1097, 1102, 163 U.S.P.Q. 580, 584 (2d Cir. 1969), Cert. denied, 397 U.S. 1076, 90 S. Ct. 1518, 25 L. Ed. 2d 811 (1970); Reiner v. I. Leon Co., 285 F.2d 501, 503, 128 U.S.P.Q. 25, 27 (2d Cir. 1960), Cert. denied, 366 U.S. 929, 81 S. Ct. 1649, 6 L. Ed. 2d 388 (1961); B.G. Corp. v. Walter Kidde & Co., 79 F.2d 20, 22, 26 U.S.P.Q. 288, 289-90 (2d Cir. 1935). If the inventions are new, useful, and nonobvious, they are patentable. If the level of skill of a person of ordinary skill in the pertinent art is such that the differences between the subject matter sought to be patented and the prior art would not have been obvious to that person, the test for nonobviousness is met.

In view of all the foregoing, we hold that the statutory presumption of validity of the Norris '276 patent has not been rebutted by Champion.

The judgment of the district court is reversed and the case is remanded for further proceedings consistent with this opinion.