536 F.2d 145
190 U.S.P.Q. 6
AIRTEX CORPORATION, Plaintiff-Appellant and Cross-Appellee,
v.
SHELLEY RADIANT CEILING COMPANY, Defendant-Appellee and
Cross-Appellant.
Nos. 75-1643, 75-1644.
United States Court of Appeals,
Seventh Circuit.
Argued Dec. 10, 1975.
Decided April 14, 1976.
As Amended on Denial of Rehearing and Rehearing En Banc May 27, 1976.
George N. Hibben, Chicago, Ill., for Airtex.
William A. Marshall, Chicago, Ill., for Shelley Radiant.
Before CASTLE, Senior Circuit Judge, TONE, Circuit Judge, and WARREN, District Judge.*
CASTLE, Senior Circuit Judge.
Plaintiff Airtex Corporation (hereinafter "Airtex"), owner of U.S. Patent No. 3,698,475 relating to radiant heating and cooling panels, appeals from the judgment of the district court holding the patent invalid. Plaintiff also appeals the district court's award of expenses under Fed.R.Civ.P. 37 to defendant Shelley Radiant Ceiling Company (hereinafter "Shelley") for plaintiff's failure to answer certain of defendant's interrogatories. Defendant cross-appeals the district court's refusal to award full attorney fees under 35 U.S.C. § 285 and the district court's holding that plaintiff was not guilty of unfair competition in sending notices of infringement to several of defendant's customers.
I.
The parties are competing manufacturers of radiant heating and cooling ceiling panels. Each ceiling panel consists of copper tubing bent into a serpentine shape and metallurgically bonded by soldering to the unexposed side of a thin flat aluminum sheet. Several panels are fitted together and installed as a ceiling in a room with the copper tubing interconnecting to form an integrated network. By circulating heated or cooled water through the copper tubing, the room in which the panels are installed can be either heated or cooled, respectively, by means of radiant energy.
Heating and cooling through radiant energy is accomplished by the transfer of energy from warmer objects to cooler objects. When used to cool a room, cooled water is circulated through the copper tubing of the panels. The water cools the tubing and in turn the aluminum panels to which the tubing is attached. The cooled aluminum panels in turn absorb the radiant energy, i. e. the warmth, from objects in the room. Conversely, when used to heat a room, heated water is circulated through the tubing which in turn warms the aluminum sheet. The warmth from the aluminum is then radiated into the room and transferred to less warm objects.
The concept of radiant heating and cooling predated the creation of the patented panel involved in this appeal. Long before copper tubing was metallurgically bonded to aluminum sheets, it was mechanically attached by way of metal clips or snaps. The capacity of these standard "snap-on" panels to cool a room was, however, quite limited. Because the mechanical bond between the tubing and the aluminum sheet was not sufficiently uniform and strong enough, energy was lost due to the uneven contact between the tubing and aluminum. This insufficient contact did not harm the panel's capacity to heat since the problem could be overcome by simply increasing the temperature of the water circulated through the tubing. However, when used to cool, the answer to the problem was not as simple; the temperature of the water could not be further lowered to compensate since once the temperature dipped below the dew point of the air around the panel, condensation would occur and the panels would drip. Thus, the "snap-on" panel's commercial marketability was hindered by its limited cooling capacity.
Airtex manufactured and marketed these standards "snap-on" panels pursuant to the Baran patent (U.S. Patent No. 2,818,235) which it had previously acquired. Airtex, recognizing the limited marketability of the standard "snap-on" panel, desired to market a better panel the cooling capacity of which would be as sufficient as the heating capacity of the standard panels.
Beginning in the mid-1950's, Airtex sought to develop a "high performance" panel, i. e. a panel that would overcome this cooling limitation.1 To achieve this goal, Airtex immediately realized that the tubing and sheet would have to be joined in a manner that would achieve the greatest possible uniform contact between them. In its search for a high performance panel, Airtex solicited the aid of, among others, Edward G. Beck, Jr., an engineer employed by the Stolle Corporation, and the Kawneer Corporation, manufacturer of standard "snap-on" panels for Airtex.
During 1961 and 1962, upon the request of Airtex, Beck developed and produced sample panels comprised of two aluminum sheets joined together with the water passages integrally formed in one sheet. Despite satisfactory performance tests, this project was abandoned by Airtex when corrosion problems occurred in the water passages of the panels. Beck continued, however, to experiment in developing a high performance panel that would overcome the corrosion problem.
Meanwhile, some time in 1960 or 1961, Airtex turned to Kawneer Corporation for assistance in the attainment of its quest for a high performance panel. At that time Kawneer was a manufacturer of standard "snap-on" panels for Airtex. Airtex requested Kawneer to develop a high performance tube-on-sheet radiant panel. Over the period encompassing the years 1961 to 1964, Kawneer developed a high performance panel in which copper tubing was adhesively secured to an aluminum sheet by an epoxy resin. These resin-bond panels performed satisfactorily in heating and cooling tests. Airtex found the panel to be feasible but nevertheless hesitated to commit itself and accept these panels as the answer to its search.
Contemporaneous with the developments at Kawneer, Beck pursued his work toward development of a high performance panel using a metallurgical bond to join the two components of the panels. To avoid the corrosion problem encountered by another in "all-aluminum" tube-on-sheet panels, Beck substituted copper tubing for the aluminum tubing. Encountering his own problems in soldering the two metals together, Beck experimented with various thicknesses of aluminum and copper tubing, types of flux, solder compositions, and alloy-plated aluminum sheets. Finally, some time between late 1964 and the middle of 1965, Beck reduced to practice a high performance tube-on-sheet radiant panel composed of copper tubing soldered to a sheet of aluminum.
Three separate patents in connection with this panel were issued to Beck: one for the process, one for the apparatus, and one for the product itself.2 Of these three, only the product patent is in suit and before us on appeal. The Beck product patent is entitled "Flat Sheet of Metal Having an Elongated Member Secured Thereto" and describes a radiant panel composed of the following preferred embodiments: a flat sheet of aluminum of a thickness of 0.040 of an inch and copper tubing having an 0.50 of an inch inside diameter and a wall thickness of 0.028 of an inch. The solder used in joining these two elements is described as consisting of a composition of 91 percent tin and 9 percent zinc using a flux under the trade-name ALCOA No. 62. The patent states that the assembled panel "will possess" an "excellent joint bond" from the mechanical standpoint and the heat transfer standpoint. That patent further states that the aluminum sheet will retain its high degree of flatness without distortion.
The Beck patent, therefore, attempts to cover not only a functional relationship of parts but also an esthetic aspect since a flat appearance is essential for commercial acceptance of the panels for use as a ceiling. As Airtex points out, not only must there be a strong solder bond to insure a long life for the panels, but the thin aluminum sheet must also be flat when the ceiling is installed and remain flat during the operation of the ceiling heating and cooling system.
Subsequent to the issuance of the Beck patent, Shelley developed, manufactured, and began marketing a radiant heating and cooling ceiling panel of its own. Shelley's panel also consisted of copper tubing metallurgically bonded to a flat thin aluminum sheet as was Beck's panel. Prior to this and contemporaneous with its marketing of the soldered tube-on-sheet panel, Shelley also marketed a standard "snap-on" panel.
Learning of Shelley's marketing of both panels, Airtex mailed notices of infringement to sixteen of Shelley's customers and thereafter filed suit for infringement of both the Beck and Baran patents.
Shelley defended the suit on the ground that the Beck patent was invalid due to obviousness under 35 U.S.C. § 103. Shelley further maintained that the Beck panel had been placed "on sale" more than one year before the patent was applied for in violation of 35 U.S.C. § 102(b). Finally, Shelley argued the Beck patent was invalid under 35 U.S.C. § 112 for lack of specificity of its claims. With regard to Airtex's charge that it infringed the Baran patent, Shelley denied infringement and contended that the Baran patent was nevertheless unenforcible because of Airtex's "unclean hands."
Shelley also pleaded three counterclaims. The first counterclaim charged Airtex with unfair competition. The second charged price discrimination under 15 U.S.C. § 13, and the third charged antitrust violations under 15 U.S.C. §§ 15 and 26. Airtex denied all three charges in its reply.
The third counterclaim was severed before trial and the second counterclaim was dismissed with prejudice during trial after Shelley sought its voluntary dismissal.
After a seven-day trial, the district court held the Beck patent invalid due to obviousness, lack of specificity, and prior sale of the panels. The court further held that Shelley had infringed the Baran patent but that any infringement was de minimis and did not warrant an award of attorney fees. With respect to Shelley's remaining counterclaim charging Airtex with unfair competition, the court held that Airtex had been justified in sending notices of infringement to Shelley's customers since there was no evidence that Airtex knew or believed the Beck and Baran patents to be invalid and since it believed in good faith that Shelley was infringing both patents.
The court also denied Shelley's request for an award of attorney fees, finding the case not to be "exceptional" under 35 U.S.C. § 285 since there had been no showing of any fraud perpetrated on the patent office by either Airtex or Beck. The court did however grant Shelley's post-trial motion that under Fed.R.Civ.P. 37 Airtex bear expenses incurred by Shelley in investigating matters which were not disclosed in certain of Airtex's answers to Shelley's interrogatories.
Airtex appeals the district court's holding the Beck '475 patent invalid and the court's award of expenses to Shelley. Shelley cross-appeals the court's orders concerning its first counterclaim and its denial of full attorney fees.3
II.
The district court held the Beck patent invalid on the ground that "(t)he combination covered by the (patent's) claims (is) obvious to persons skilled in the art." We agree with this ultimate conclusion and therefore affirm the court's holding of invalidity.
A.
As a preliminary matter, Airtex maintains that the district court failed to follow the dictate of 35 U.S.C. § 103 and consider the obviousness of the invention "at the time the invention was made." Rather, Airtex argues the court determined obviousness as of the time of trial as evidenced by such statements in its opinion as: "The Subject Matter of the Beck Patent Is Obvious" (emphasis added); and other similar uses of the present rather than the past tense. Airtex maintains that this is reversible error, relying on our decision in Walt Disney Productions v. Niles, 369 F.2d 230, 234 (7th Cir. 1966). In Walt Disney we found that the record disclosed that the district court had employed "hindsight" as of the time of trial in determining the obviousness of the patented invention before it. Here the record supports no such view. Rather, we believe that the record supports our conclusion that the cited examples of the court's alleged application of an erroneous standard was merely the loose use of verb tense. Under the record before us, no grounds exist for challenging the district court's application of the proper standard mandated by Section 103.
B.
Airtex concedes and the district court found that each individual element of the Beck panel was old.4 However, Airtex maintains that the combining of these old elements for the first time by Beck resulted in a "novel, unobvious and unthinkable structural combination to accomplish the new and patentable relationship defined in the claims as entireties."
Where the claimed nonobviousness of the invention rests on a new combination of old elements, the claimed invention must pass a "rather severe test" consonant with the difficulty and improbability of finding invention in an assembly of old elements. Gettleman Manufacturing, Inc. v. Lawn 'N' Sport Power Mower Sales & Service, Inc., 517 F.2d 1194, 1197 (7th Cir. 1975), citing Panduit Corporation v. Burndy Corporation, 517 F.2d 535, 539 (7th Cir. 1975); see also Skil Corporation v. Lucerne Products, Inc., 503 F.2d 745, 749 (7th Cir. 1974), cert. denied, 420 U.S. 974, 95 S. Ct. 1398, 43 L. Ed. 2d 654 (1975).
In considering the question of the obviousness of a combination of old elements, we shall employ the analysis set forth by Mr. Justice Stevens while still a member of this court in E-T Industries, Inc. v. Whittaker Corporation, 523 F.2d 636 (7th Cir. 1975). In that case, Mr. Justice Stevens presented a three-step process to be followed in this type of situation. First, we must consider whether each element of the invention is obvious. We need not delve into this inquiry because Airtex concedes that each element of the Beck panel was old per se. The second step in the analysis is to determine whether the combination is obvious in itself. And if it is, the third step is to decide whether "the rejection of the contrary teaching in the prior art requires a different conclusion." Id. at 641.
The combination of old elements comprising the Beck panel, viz. copper tubing soldered to a thin, flat sheet of aluminum, was unquestionably obvious when viewed in light of the relevant prior art. The superior performance obtained from the use of copper tubing with aluminum sheets in a radiant panel was clearly taught by the patented standard "snap-on" radiant panels and the Kawneer epoxy resin-bonded panels, both of which used copper tubing in conjunction with aluminum sheets. Moreover, the feasibility and benefits of joining copper to aluminum by means of a metallurgical bond was taught by at least three publications: Leopold, Design Factors in Panel and Air Cooling Systems, Heating, Piping and Air Conditioning, at 126 (May 1951); Miller, Joining Aluminum to Other Metals, the Welding Journal, at 731-33, 735 (August 1953); and Aluminum Development Association, Soldering Aluminum (A.D.A. Information Bulletin No. 23, 1957); as well as the Scott patent (U.S. Patent No. 2,311,579; issued Feb. 16, 1943) relating to solar heaters in which copper tubing is soldered to the face of a thin sheet of aluminum.5
(3) Airtex maintains, however, that this conclusion must be altered in light of the prior art's teaching against obtaining the essential characteristics of the Beck panel as stated in claim 1, to-wit: (1) "a strong fracture resistant and heat conductive bond between the copper tubing and a relatively thin commercially desirable sheet," and (2) the aluminum sheet's retention of flatness and freedom from localized "wrinkles," "ripples," "undulations," and other "distortion" (a) after the application of high soldering temperatures necessary to bond the copper tubing to the sheet, and (b) during continued operation and use of the panel as an integrated component of a heating and cooling system. These characteristics, Airtex submits, makes what appears to be an obvious combination of old elements in reality nonobvious. We disagree. When examined closely and carefully, the prior art does not teach against obtaining these essential characteristics of the Beck panel, but in fact teaches the opposite.6
The first "unexpected" result obtained by Beck, Airtex maintains, was the "strong fracture resistant and heat conductive bond" achieved by soldering copper tubing to a thin, flat aluminum sheet. Airtex points out that the strength of the bond in a radiant panel composed of copper tubing and an aluminum sheet is crucial because aluminum possesses a greater thermal coefficient of expansion than copper. Hence, as aluminum is cooled by the cooled water circulating through the copper tubing, it will contract to a greater degree than the copper tubing. Conversely, during the heating cycle, the aluminum sheet will expand to a greater degree than the copper tubing.7 This phenomenon causes any bond between the tubing and sheet to undergo continuous and constant strain. Therefore, the bond between the two elements must be of such strength and durability to withstand such punishment; otherwise the bond will fracture and deteriorate, causing the panel to lose its superior ability to radiate and thus its "high performance" characteristic.
Despite the existence of this phenomenon, the "strong fracture resistant and heat conductive" bond found in the Beck patent cannot be labeled an unexpected result of soldering copper to aluminum. Nor can it be asserted that the prior art taught against obtaining such a bond by soldering two metals together which possess different thermal coefficients of expansion. To the contrary, the prior art, most notably the Miller article, supra, teaches the superior strength of a metallurgical bond over that obtained mechanically or by means of an adhesive resin. Furthermore, the ability of a metallurgical bond to withstand the rigors of differing thermal coefficients of expansion was disclosed by two items of the prior art: the Scott patent, supra, and the Collins patent (U.S. Patent No. 2,934,917).
The Scott patent relates to a hot water solar heater which is composed in part by copper tubing soldered to a metal sheet. The Scott patent thus teaches that a metallurgical bond comparable to that found in the Beck panel may withstand the stress and strain imposed upon it by the different thermal coefficients of expansion possessed by copper and another metal when heated water is circulated through the tubing.
The ability of the Beck solder bond to withstand similar stress and strain during the cooling cycle of the panel is disclosed by the Collins patent. That patent relates to a refrigerator evaporator composed in part by copper tubing metallurgically bonded to an aluminum sheet. Liquid cooled to a temperature below that of the water used in the cooling cycle of the Beck panel is passed through the tubing to refrigerate or freeze the evaporator's contents. Thus the Collins patent clearly teaches the ability of a soldered bond between copper tubing and an aluminum sheet to withstand the stress caused by the aluminum contracting to a degree greater than that of the copper tubing.
Together, both of these patents disclose the feasibility of the Beck panel's metallurgical bond surviving the continuous rigor caused by the different thermal coefficients of expansion.8 Hence, the prior art does not teach against obtaining the strong, fracture resistant bond found in the Beck panel.
The differing thermal coefficients of expansion of copper and aluminum also serve as the crux of Airtex's assertion that the Beck panel obtained an unexpected result by remaining flat and free from localized distortion during use and operation. Airtex maintains that because copper and aluminum expand and contract at different rates, the prior art discouraged soldering the two together in a radiant heating and cooling panel because the flat, thin sheet of aluminum would wrinkle when heated or cooled water was circulated through the tubing. It was essential that the panels not distort during operation of the heating and cooling system since the esthetic appearance of flatness of the ceiling is important in the panel's commercial marketability.
Airtex implicitly argues that nothing in the prior art suggested how this problem could be overcome. Airtex correctly points out that "flatness" of the sheets is of no importance in either the Scott or Collins panel. Nevertheless, we believe that one other relevant item of the prior art clearly discloses that this distortion problem could indeed be overcome.
The Kawneer panel employed an adhesive epoxy resin bond between the copper tubing and the aluminum sheet. While perhaps not to the same degree, the resin-bond panel was nonetheless subject to the same basic problem as the Beck panel. While the use of resin in the Kawneer panel may have resulted in a more flexible bond, the basic problem of distortion during operation must nevertheless still be present because copper tubing is bonded to a thin aluminum sheet just as in the Beck panel. It is obvious that the Kawneer panel's overcoming of the problem signaled the way for Beck. By experimenting with different fluxes and compositions of soft solder, Beck was also able to overcome the problem as the Kawneer panel had previously. While Beck may have been the first to overcome the problem with bonding the two metals together with a solder,9 the Kawneer panel disclosed to persons skilled in the art that the problem of distortion during operation resulting from bonding copper tubing to a thin flat sheet of aluminum was not insurmountable, but rather quite solvable. The substitution of solder in place of resin by Beck only served to aggravate the problem previously encountered in the Kawneer panels. This aggravation, however, was not of such a degree that persons skilled in the art and knowledgeable in the Kawneer panel would think experimentation with various soft solders and fluxes would be "purposeless" in repeating the success achieved by the Kawneer panel.10
A very similar situation was before this court in E-T Industries v. Whittaker Corporation, supra. In that case the patentee had made an automobile wheel adaptable to different makes of automobiles by elongating the lug apertures of the wheel sufficiently to accommodate lug bolts arranged in circles of different radii. See generally id. at 637. Different adapters were then inserted to fit snugly in the enlarged lug apertures in order for the wheel to fit on the lug bolts of different makes of cars. The patentee argued that the prior art taught against enlarging the lug apertures since the result would be to weaken the wheel. Mr. Justice Stevens, then a member of this court, however, did not view the prior art in a similar manner:
In this case, the contrary factor was merely the rather obvious notion that enlarging holes in a structural member will weaken it to some extent. But the record establishes that it was equally clear that some enlargement might be tolerated and that modifications in design or material strength might overcome the effect of the loss of some material. As the testimony of plaintiff's own expert indicates, the problem was not so serious as to deter any investigation at all of enlarged lug apertures, but rather it was a problem . . . which would normally require careful testing and perhaps further experimentation and modification. (Id. at 642 (emphasis added).)
Similarly, in this case the distortion problem encountered during the panel's operation, it seems to us, was not so serious as to deter any investigation at all; the Kawneer panel demonstrated that distortion during operation was not an insurmountable problem. Rather, it was merely a problem which would "require careful testing and perhaps further experimentation and modification" as to the type of solder and thickness of aluminum to be employed. "This prior art warning to be careful . . . is not, in our judgment, equivalent to the uniform prior art teaching in Adams (United States v. Adams, 383 U.S. 39, 86 S. Ct. 708, 15 L. Ed. 2d 572 (1966)) that any investigation of the use of magnesium electrodes in a water activated battery would be purposeless." Id. at 642. Any teaching of the prior art here was not that it was "purposeless" to attempt to overcome the problem of distortion during operation, but rather only that careful testing and experimentation would be necessary to overcome it when solder was employed to join the two metals.
Identical reasoning applies to Airtex's next assertion that the Beck panels achieved an unexpected result in retaining their "flatness" and freedom from distortion despite the high temperatures the aluminum sheets endured in the soldering process. Airtex contends that the prior art discouraged soldering copper tubing to thin flat sheets of aluminum because the temperature reached in soldering (c. 510o F) would cause the aluminum panel to distort. Any such teaching of the prior art, however, related not to the fact of soldering itself, but rather to the heat generated by soldering. The prior art, however, clearly taught that this heat problem was not insurmountable. The Blessing patent (U.S. Patent No. 2,423,870), relating to adhesive resin bonded panels, involved thermosetting temperatures for the resin of between 392o and 572o F. In the file history of Beck's related process patent, Beck equated Blessing's thermosetting temperature with Beck's soldering temperature and Airtex, in its brief before this court, states that Blessing and Beck would therefore have encountered the "same distortion problems resulting from heat in securing tubing to a metal sheet."11 Therefore the Blessing patent taught that the aluminum sheet could in fact remain flat despite the high temperatures experienced during the soldering process.12
Hence, as demonstrated, the Beck patent does not survive the analysis set forth in our E-T Industries decision. It has not successfully withstood the "rather severe test" which an invention consisting of a combination of old elements must satisfy. Gettleman Manufacturing Inc. v. Lawn 'N' Sport Power Mower Sales and Service, Inc., supra. We accordingly hold the Beck '475 patent invalid as a matter of law due to obviousness.13
III.
Under 35 U.S.C. § 285, a district court may award attorney fees if the court finds the case to be "exceptional."14 After trial, Shelley requested such an award and the district court refused on the ground that the case was not "exceptional." Shelley cross-appeals this holding of the district court, contending that it is entitled to fees under section 285 for three reasons.
First, Shelley maintains that Airtex failed to advise the Patent Office of certain pre-1965 activities of Airtex, Beck, and the Stolle Corporation. Shelley contends that Stolle's transfer to Airtex of 12 panels in November of 1963 together with Airtex's subsequent sale of similar panels to a third party constituted prior commercial activity under 35 U.S.C. § 102(b) of which Airtex was under a duty to advise the Patent Office. The district court found, however, that the shipment of the 12 panels to Airtex did not constitute a "sale" since the panels were experimental and shipped at Airtex's request for purposes of inspection. The district court further found that knowledge of the subsequent sale of similar panels by Airtex could not be imputed to Stolle or Beck prior to one year before the patent was applied for.
Second, Shelley contends that Airtex brought an infringement action against it while knowing the Beck patent was invalid. The district court, however, found that there was no evidence to support such a contention.
And third, Shelley maintains that Airtex failed to answer certain interrogatories in good faith, thus necessitating Shelley to conduct its own investigation to ferret out information which Airtex should have disclosed in its answers. The interrogatories in question sought information and documents concerning the dates of conception, first disclosure, first written description and reduction to practice of the Beck panel. In its answers, Airtex professed not to have first hand knowledge and referred Shelley to Beck. The district court refused to find this a ground for awarding fees under section 285, but did grant Shelley's post-trial motion under Fed.R.Civ.P. 37 for an award of expenses incurred by it in investigating the matters not disclosed in the answers.
Our review of the record results in the same conclusion as the district court's viz. this did not constitute an "exceptional case" under section 285 to warrant an award of attorney fees. We cannot say that the district court was clearly erroneous in finding either that Airtex believed in good faith that the Beck patent was valid, or that there was an absence of fraud on the Patent Office. While the " 'exceptional case' justification for the allowance of fees is not . . . limited to the fraud category," L. F. Strassheim Company v. Gold Medal Folding Furniture, 477 F.2d 818, 824 n.9 (7th Cir. 1973), we believe section 285 requires an unambiguous showing of extraordinary misconduct. See H. K. Porter Company v. Black & Decker Manufacturing Company, 518 F.2d 1117, 1178-79 (7th Cir. 1975); Technograph Printed Circuits, Ltd. v. Methode Electronics, Inc., 484 F.2d 905, 909 (7th Cir. 1973). No such showing has been made here.
Airtex appeals the district court's award of expenses in connection with Airtex's failure to satisfactorily answer certain of Shelley's interrogatories. After reviewing the interrogatories and answers in question, we agree with the district court that while such conduct on the part of Airtex does not warrant an award of fees under section 285, plaintiff should bear Shelley's expenses in investigating the matters the interrogatories concerned. Airtex's answers do not comport with the duty of cooperation and disclosure imposed by the discovery provisions of the federal rules. Airtex's answers were evasive and incomplete and appear to have been framed to impede discovery rather than to facilitate it. Under these circumstances, the district court was well within its discretion in awarding expenses to Shelley pursuant to Fed.R.Civ.P. 37.
Although the district court did not cite the specific subsection of rule 37, its award was necessarily based on subsection (d). Airtex contends that at no time did Shelley move for expenses under rule 37(d). Our reading of the record indicates, however, that Shelley indeed brought rule 37 to the attention of the court in connection with its post-trial request for attorney fees under section 285. Relying on language in Butler v. Pettigrew, 409 F.2d 1205, 1207 (7th Cir. 1969), Airtex argues that Shelley waived any right under rule 37 by failing to seek sanctions before trial. Under the circumstances of this case, however, it would be unjust to find a waiver on the basis of Shelley's failure to request an order to compel answers under rule 37(a). The evasive and incomplete character of Airtex's answers was not immediately apparent to Shelley and did not become so until Shelley conducted further discovery. It was useless to ask the court to compel Airtex to provide information which Shelley had subsequently uncovered through its own efforts. Thus the only practical occasion to seek sanctions against Airtex was after trial in the form of requesting expenses. Since such request was made after trial and before judgment, no grounds for finding a waiver exit. Cf. Popeil Brothers, Inc. v. Schick Electric, Inc., 516 F.2d 772, 778 (7th Cir. 1975). Airtex also argues that rule 37(d) relates only to the situation where a party totally fails to file any answers to interrogatories, citing 8 Wright & Miller, Federal Practice & Procedure § 2291, at 810. It is true that rule 37(d) does not specifically cover giving answers that are evasive and incomplete, as distinguished from failing to answer at all. Compare Fed.R.Civ.P. 37(a)(3). We believe, however, that when, as here, the fact that answers to interrogatories are evasive or incomplete cannot be determined until further proceedings have been conducted to obtain the information later determined to have been withheld, the evasive or incomplete answers are tantamount to no answer at all, and rule 37(d) is applicable. To hold otherwise would be to deny the interrogating party any remedy for a deliberate violation of rule 33.
IV.
As a final matter, Shelley appeals the district court's adverse holding on its first counterclaim charging Airtex with unfair competition. Shelley's counterclaim was in response to Airtex's mailing of notices of infringement to 16 Shelley customers. The district court found nothing in the letters actionable and held that "(a) good faith belief that a patent is being infringed, particularly when promptly followed up by filing a lawsuit in good faith, is a defense to a charge of unfair competition," citing Virtue v. Creamery Package Mfg. Co., 227 U.S. 8, 37-38, 33 S. Ct. 202, 208, 57 L. Ed. 393 (1913), and Applied Biochemists, Inc. v. A and V Inc., 353 F. Supp. 949 (E.D.Wis.1973). We have reviewed the letter and agree that it contained no disparaging remarks. Nothing in the record sufficiently indicates bad faith on the part of Airtex in either its belief that the Beck patent was valid or its motive in sending the notices. We believe a further indication of its good faith is the fact that the notices were selectively sent only to those architects, contractors, and others who were in fact considering the purchase of Shelley panels rather than an indiscriminate wholesale mailing of notices to anyone who might be interested in radiant heating and cooling panels.
Accordingly, the judgment of the district court is affirmed.
Affirmed.
The Honorable Robert W. Warren, District Judge of the United States District Court for the Eastern District of Wisconsin, is sitting by designation
High performance panels have been defined as those possessing a surface temperature substantially equal to the temperature of the fluid circulating medium. See Howarth, Aluminum Ceiling Panels for Heating and Cooling, Heating, Piping and Air Conditioning, at 129 (Sept. 1951)
All three patents were based upon the same description and drawings and each is entitled to the same filing date of the abandoned parent application (SN 523,619), January 28, 1966. The process for producing the Beck panel is covered by U.S. Patent No. 3,514,834 (issued June 2, 1970), and is entitled "Method for Securing an Elongated Metal Member to a Flat Metal Sheet." The apparatus used in producing the panels is covered by U.S. Patent No. 3,703,758 (issued Nov. 28, 1972), and is entitled "Apparatus for Securing an Elongated Metal Member to a Flat Metal Sheet." The product patent '475 (issued Oct. 17, 1972) and apparatus patent applications were divisional applications of the process patent application
Neither Airtex nor Shelley appeals the district court's holding concerning Shelley's de minimis infringement of the Baran patent. Hence, we need not consider the Baran patent, other than its relevance to Shelley's cross-appeal
In response to Shelley's interrogatory No. 25.00, Airtex stated:
. . . A radiant heating and cooling panel per se comprising a flat sheet of aluminum less than 0.100 inch thick was not novel. Copper tubing per se having an inside diameter of the order of one-half inch and wall thickness of the order of 0.028 was not novel. Solder per se was not novel. A solder per se composed of 91 percent tin and 9 percent zinc was not novel. A radiant panel per se having substantial flatness and absence of wrinkles, ripples and undulations was not novel.
Also, an article published in 1957 describes in detail a building in Albuquerque, New Mexico which is heated by solar energy collectors. These collectors are constructed from flat aluminum sheets with sinuous copper tubing soldered thereto. Water is circulated through the tubing and collects energy from the sun. Bridgers, Paxton & Haines, Performance of a Solar Heated Office Building, Heating, Piping and Air Conditioning, at 165-71 (Nov. 1957)
In our discussion to follow concerning the alleged "unexpected" results obtained by Beck, we refer to certain items of the prior art set forth in the record before us but not set out and discussed by the district court in its opinion. Airtex maintains that the district court disobeyed the mandate of Graham v. John Deere & Co., 383 U.S. 1, 17, 86 S. Ct. 684, 693, 15 L. Ed. 2d 545, 556 (1966), by failing to fully describe the scope and content of the prior art. While we are constrained to agree with that observation, we do not agree that remand is necessary. We possess authority in patent-validity cases involving obviousness claims to review the record below. See Julie Research Laboratories, Inc. v. Guildline Instruments, Inc., 501 F.2d 1131, 1135-36 (2d Cir. 1974) and cases cited therein. While we would have preferred that the district court had thoroughly followed the dictates of Graham, its failure to do so does not prevent this court from relying on items of the prior art found in the record on review to counter the arguments of Airtex. Compare University of Illinois Foundation v. Blonder-Tongue Laboratories, Inc., 422 F.2d 769, 778 (7th Cir. 1970), cert. denied, 409 U.S. 1061, 93 S. Ct. 559, 34 L. Ed. 2d 513 (1972); see also Cloud v. Standard Packaging Corporation, 376 F.2d 384, 391 (7th Cir. 1967)
The temperature range of the water circulating through the tubing is designed to be from 50-55o F. in the cooling cycle to 200-250o F. in the heating cycle
We find it of no significance that the metallurgical bond of the Beck panel must undergo the stress and strain of both the heating and cooling cycles while the Collins and Scott constructions only undergo one cycle
We also note that Airtex failed to show at any time that Beck in fact was the first to overcome the distortion problem in relation to a tube-soldered-to-aluminum sheet construction. While the freedom from distortion was not an essential characteristic of the Scott or Collins constructions, there was no showing that either of these did in fact distort during operation
Airtex points to the statement of R. E. Boyar, Chief Development Engineer on radiant ceilings for Burgess-Manning, Inc. and later Inland Steel Products. In evaluating the Beck panel in 1965, he expressed disbelief that the panel could be operated without distorting because of the difference in thermal coefficients of copper and aluminum. Boyar stated in a memorandum: "In my opinion, it is perfectly safe to point out that no known manufacturing process can equalize the different thermal coefficients of expansion of copper and aluminum." Boyar's statement cannot be taken literally since, of course, the Beck panel did not "equalize" the different thermal coefficients of copper and aluminum. Regardless, there is no indication that he was aware of the Kawneer panel at the time this statement was made. Boyar's "skepticism" therefore does not warrant the treatment given similar statements of skepticism in United States v. Adams, 383 U.S. 39, 44, 86 S. Ct. 708, 710, 15 L. Ed. 2d 572, 576 (1966), and Skil Corp. v. Cutler-Hammer Inc., 412 F.2d 821, 824 (7th Cir. 1969)
Plaintiff's Reply Brief at 28
Airtex submits that corrosion was another "discouraging factor" the prior art taught but that Beck overcame. Airtex states that while soft solders might have been the answer to the distortion problem during soldering, soft solders were known to be less corrosion resistant that hard solders. We are of the opinion, however, that such a problem was not so "discouraging" as to teach against testing and experimentation. Airtex has failed to demonstrate that any corrosion problem was severe enough to actually discourage persons skilled in the art from attempting to overcome it
Airtex also points to certain "secondary factors" as bearing on the question of obviousness, to-wit: the longfelt but unsolved need for such a high performance panel, the commercial success of the Beck panel, and imitation of the Beck panel by Shelley. The district court's failure to consider these factors in its opinion is of no consequence. While such factors may be of relevance as indicia of obviousness, Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S. Ct. 684, 693, 15 L. Ed. 2d 545, 556 (1966), the district court need only turn to them in a close case. Potter Instrument Company, Inc. v. ODEC Computer Systems, Inc., 499 F.2d 209, 211 (1st Cir. 1974). This case is not a close one.
Since we hold that the Beck patent in suit invalid due to obviousness, we need not consider the district court's further grounds for holding the patent invalid
That section provides: "The court in exceptional cases may award reasonable attorney fees to the prevailing party."