Northlake Marketing & Supply, Inc. v. Glaverbel, S.A.

72 F. Supp. 2d 893, 1999 U.S. Dist. LEXIS 18123, 1999 WL 1063058

• Court: District Court, N.D. Illinois
• Decided: June 25, 1999
• Docket: 92 C 2732
• Status: Published

Opinion

FINDINGS OF FACT AND CONCLUSIONS OF LAW

SHADUR, Senior District Judge.

Over the protracted length of this litigation 1 this Court has found every claim advanced and every defense asserted by Northlake Marketing & Supply, Inc. (“Northlake”) and its principals James Hamilton (“Hamilton”) and Samuel May (“May”) to be devoid of merit. They have taken an unsuccessful appeal to the Court of Appeals for the Federal Circuit from a host of this Court’s orders (the memorandum opinion and order and permanent injunction order each dated November 13, 1997; the memorandum opinion and order and supplement, both dated March 11, 1997; the memorandum opinion and order dated December 16, 1996; the memorandum opinion and order dated August 29, 1995; and the memorandum opinion and order dated August 4, 1994), with that Court having rejected all of those claims in a per curiam one-sentence judgment order entered November 13, 1998 (unpublished, but available as 1998 WL 796051).

With Glaverbel S.A. (“Glaverbel”) and Fosbel, Inc. (“Fosbel”) having prevailed on the merits of their patent infringement counterclaim, the only issue remaining before the litigation at long last terminates is the quantification of the damages recoverable for that infringement. This Court has conducted the trial on that subject, only to find the issuance of its ultimate ruling delayed substantially by the lapse of many' months before the litigants tendered their respective proposed findings of fact and conclusions of law.

Those submissions have now been made, and what follows are this Court’s Findings of Fact (“Findings”) and Conclusions of Law (“Conclusions”) in accordance with Fed.R.Civ.P. (“Rule”) 52(a). To the extent (if any) that the Findings as stated may be deemed conclusions of law, they shall also be considered Conclusions. In the same way, to the extent (if any) that matters later expressed as Conclusions may be deemed findings of fact, they shall also be considered Findings. In both of those respects, see Miller v. Fenton, 474 U.S. 104, 113-14, 106 S.Ct. 445, 88 L.Ed.2d 405 (1985).

Findings of Fact

Parties and Background

1. Northlake Marketing was formed as a partnership in 1984 and incorporated as Northlake Marketing & Supply, Inc. in 1985. In 1995 Hamilton, one of the founders and a current owner of Northlake, formed Northlake Industries, Inc. as a separate company, and he is the sole owner of that company. In 1988 May, also one of the founders and a current owner of Northlake, formed N & E Refractories as a separate company, and he is the sole owner of that company. May left North-lake at the end of 1994 but did not relinquish his ownership position or his position as an officer-director. For convenience these Findings and Conclusions will also use “Northlake” as a singular collective noun to refer to the corporation and the two individual counterdefendants.

2. Glaverbel S.A. (“Glaverbel”) is a Belgian corporation that was the successor by merger of two firms, each of which dated back to the 1930s. Glaverbel is the owner of the United States patents involved in this action. Fosbel, Inc. (“Fos-bel”) is a joint venture based in Cleveland, Ohio, 49% of which joint venture is owned by Glaverbel and the other 51% of which is owned by the other joint venturer. Fosbel is the exclusive licensee of the Glaverbel-Fosbel patents in the United States.

3. This Court has previously determined that the two Glaverbel United States patents involved in this action— -Nos. 4,792,468 (“ ’468 Patent”) and 4,920,-084 (“’084 Patent”) — are “not invalid,” were not procured by inequitable conduct and were infringed by Northlake. This Court has accordingly issued a permanent injunction that terminates on expiration of the two Glaverbel patents. What remains is a determination of the monetary relief to which Glaverbel and Fosbel are entitled. They seek (1) an award of “lost profits” for some of the Northlake activity and (2) a “reasonable royalty” for the remainder of the Northlake activity, together with (3) prejudgment interest, (4) enhanced damages under 35 U.S.C. § 284 2 and (5) a determination that this is an “exceptional case” such as to call for an award of attorneys’ fees under Section 285.

4. Glaverbel-Fosbel’s ’468 and ’084 Patents are two of their four United States Patents involved in their litigation with Northlake and relating to the repair of industrial furnaces, a process referred to as “ceramic welding.” In non-legal terms that process involves forming a coherent refractory mass on a “target” (e.g., a furnace wall) by projecting a mixture of particles of an oxidizable substance and particles of a refractory substance against the target. When the oxidizable substance burns exothermically (i.e., gives off heat) at the wall, it welds the refractory particles to the wall, thus repairing the furnace wall. Finding 5 describes the four patents in non-legal terms.

5. Glaverbel’s now-expired United States Patent No. 3,684,560 (“ ’560 Patent”) refers to ceramic welding where the average size of the oxidizable particles is less than 50 microns (a micron is one one-millionth of a meter). Glaverbel’s United States Patent No. 4,489,022 (“ ’022 Patent”) refers to an improvement in ceramic welding where both silicon particles and aluminum particles are present, as the oxi-dizable materials, in certain amounts and proportions. Glaverbel’s ’468 and ’084 Patents refer to an improvement in the granulometry of the refractory particles. Granulometry refers to the quantity of particles at various different sizes, and also may be referred to as the particle size distribution. According to the two patents in suit, the “size range spread factor” (a coined term that can be used as a representation of the granulometry) must be at least 1.2.

6. One use of the process of the ’468 and ’084 Patents is in repairing the silica brick walls of a coke oven. In a coke oven the process of converting coal into coke may take from 12 to 20 hours, during which time the oven operates at a range of 1200"to 2000F. During the trial this Court observed a videotape (DX 104) 3 and heard an explanation of the ceramic welding process.

Lost Profits Analysis

7. Four factors (known as the Panduit factors, see Conclusion 5) are considered (if and when present) in a lost profits analysis. Those factors are (a) a demand for the patented product or method, (b) the absence of acceptable noninfringing substitutes, (c) the manufacturing and marketing capability of the patent owner (or its licensee) to exploit the demand and (d) the amount of profit that the patent owner would have made on the infringing sales.

1. Demand for the Patented Product

8. Northlake explained that the annual rate of growth in the ceramic welding industry from 1986 to 1996 was probably 35% or 40%, while in some years the market was actually doubling (DX 140 at 223-24 to 224-9). 4 Northlake further explained that the environmental trend has been to shut down coke plants because they give off large quantities of environmentally damaging emissions, so that many plants have closed down and more are closing down. That trend has created a drastic need to repair the remaining coke plants. Because the alternative of rebricking is so expensive, Northlake described the resulting demand for ceramic welding' as a “gold mine” (DX 140 at 246-9 to 246-24). Because coke plants are no longer being built in the United States, that has contributed to the creation of the “gold mine” (DX 140 at 527-3 to 527-22).

9. Northlake’s discovery answers in this case indicated that it had documents concerning its own sales projections, market share and the size of the ceramic welding market, but those documents were not produced in response to the Glaverbel-Fosbel discovery requests (DX 149).

10. Based on Northlake’s own testimony, its failure to produce its documents referred to in Finding 9, its infringing sales and the Fosbel sales, this Court finds that there was a substantial demand for coke oven repair using the patented ceramic welding technology.

2. Absence of Acceptable Noninfring-ing Substitutes

11. Alternative techniques for coke oven repair include gunning, rebricking, lava flame and harsh press silica dusting. Findings 12 to 18 describe them and compare them with ceramic welding.

12. Gunning is the application of a water-based cement bonded refractory material through a gunning machine, which mixes the water and the cement based powder, spreading it onto the hot silica refractory. As Glaverbel-Fosbel witness John Bacon (“Bacon”) testified (Tr. 293) and this Court finds:

Q. What is the expected life of a gunning repair?

A. Very short. It can be as short as a week or two. Sometimes it last [sic] as long as six months.

Q. What is the purpose of a short term repair such as gunning?

A. It’s a quick fix. We can get it up and back into service quickly. Then we can get around to doing a proper ceramic welding repair a little later.

13. Rebricking is a major repair to a coke oven wall, involving the replacement of bricks that make up the oven wall. In an even more major rebricking job, an entire oven wall can be replaced. To accomplish rebricking the oven has to be taken out of service, the burners must be turned off and the area being repaired has to be cooled down so that the workers can lay bricks physically. Then after the repair is made and the oven reheated, the silica brick goes through various silica transformations that can crack the new brick. As Northlake explained, cooling down the ovens sometimes takes two to three weeks, and reheating th,e ovens can take three months (DX 141 at 99).

14. Lava flame is a flame spraying process.

15. Harsh press silica dusting is a method by which silica powder is sprayed into the oven to fill cracks and holes and seal the oven (DX 141 at 141). It is not a long-term fix at all.

16. By contrast, ceramic welding can extend the life of the oven lining for years. Fosbel advertises that the life of its oven repair is in excess of two years. In fact the repairs typically last two to four years, with some oven repairs having been in place for 10 years. As an additional advantage, in ceramic welding the repairs are performed from outside the oven, so that all ovens can remain in service except the oven under repair, and all ovens (including the oven under repair) remain at operating temperatures. As Northlake agreed, not having to cool down the ovens is one of the things that make ceramic welding unique (DX 141 at 97). Importantly, among the various furnace repair techniques only ceramic welding and rebricking have a bona fide life expectancy of two to four years. Finally, the Glaverbel-Fosbel ceramic welding process is certified under ISO 9000, which is an international quality certification.

17. Fosbel never lost a ceramic welding bid to any of the other furnace repair techniques. No Fosbel customer ever indicated that any of the other repair techniques would be an acceptable substitute for ceramic welding. Northlake agreed that it did not compete with companies that rebrick ovens (DX 141 at 98), and it never lost any work to companies that do silica dusting (DX 141 at 158).

18. Based on the differences in the methods of making the repairs, the times that the furnaces must be taken out of service and the expected longevity of the repairs, this Court concludes that none of the non-ceramic welding repair methods has benefits comparable to those of the patented ceramic welding technique. Hence none of them is an acceptable non-infringing substitute.

19. Glaverbel-Fosbel’s ceramic welding competitors for coke oven repair in the United States have included Northlake (a/ k/a Exo-Ram), United Refractories, Fuse Tech and Certek. Certek competed during the period 1989 through 1991, having approximately 13% of the total market, and was thereafter acquired by Fosbel. Collectively the competitors (including Northlake but excluding Certek) accounted for less than 10% of the ceramic welding business. While Northlake’s share was about 2% of the ceramic welding market, Fosbel’s share was about 80% before the Certek acquisition and about 92% thereafter.

20. No direct evidence was presented during the trial as to the details of the ceramic welding powder used by the competitors (other than Northlake) listed in Finding 19. Conclusions 7 to 9 set out the reasons why on the evidence of record those competitors also did not provide “acceptable” alternatives. As the predicate for those Conclusions, this Court finds that the history of ceramic welding started with the invention of the ’560 Patent, progressing to the invention of the ’022 Patent and then to the inventions of the ’468 and ’084 Patents. Northlake sought to invalidate the two later patents (those now in suit) on the theory that the invention was “on sale” for more than one year before the filing dates of their applications, in alleged contravention of Section 102(b). In response to Northlake’s summary judgment motion, Glaverbel-Fosbel submitted substantial evidence from the inventors explaining the history of their research to discover the improvement that became the basis of the ’468 and ’084 Patents. This Court’s August 29, 1995 memorandum opinion and order provides a summary and timetable of the Glaverbel experimentation leading up to the invention of the ’468 and ’084 Patents (1995 WL 534290, at *4-*6). Because all of Northlake’s invalidity challenges failed, on the present record the formulation of the ’468 and ’084 Patents is a novel and non-obvious improvement that provided material benefits over the prior ceramic welding powders.

21. There is circumstantial evidence of the absence of any acceptable alternative that the ceramic welding techniques of the competitors (other than Northlake) were not acceptable noninfringing alternatives because they did not have the advantages that formed the basis of the invention of the ’468 and ’084 Patents. If- on the other hand they had the novel and non-obvious attributes of the invention of the ’468 and ’084 Patents, then those competitors (like Northlake) would have infringed. That circumstantial evidence of the absence of any acceptable alternative is corroborated by Northlake’s failure to change to a noninfringing formulation during the nine years of its dispute with Glaverbel-Fosbel. Such a change to a size range spread factor of less than 1.2 would have been relatively straightforward (even some of the samples tested by Zvosec had a size range spread factor of less than 1.2), so that the absence of change supports the inference (and hence the finding) that there was a meaningful benefit or advantage to using powder with the size range spread factor of at least 1.2.

22. In addition, Northlake’s antitrust claim (statements in which may be used against Northlake under Fed.R.Evid. 801(d)(2)(A) even though disputed by Glav-erbel-Fosbel) includes the allegation of Glaverbel-Fosbel’s ceramic welding monopoly power — an assertion that means the absence of acceptable ceramic welding substitutes.

23. Based on Findings 11 to 22, this Court finds that there were no acceptable noninfringing substitutes of any kind — either non-ceramic methods or noninfringing ceramic welding- — -for the Glaverbel-Fos-bel patented ceramic welding.

3. Manufacturing and Marketing Capability To Exploit the Demand

24. According to Northlake, there are between 30 and 33 coke plants in the United States (DX 141 at 134). Northlake’s evidence at the time of the March 1992 trial in the Indiana case between Glaver-bel-Fosbel and Northlake showed that Glaverbel-Fosbel was servicing at least 23 or 24 accounts, United Refractories was servicing one or two, Lava Flame was not servicing any accounts and Northlake was servicing one account (DX 149 at 102 and 105).

25. Glaverbel-Fosbel knew the customers for ceramic welding repair at least as early as 1988, the issue date of the first of the two patents now in suit (DX 117 at 6 to 7 and Tab 5). Because you can’t hide a coke oven, everybody having any relationship to the industry knew where all of the coke ovens were, and Glaverbel-Fosbel had understandably identified those customers. As for Northlake, even though it had only one customer as of 1992, during the 1989-96 period it had provided ceramic welding materials and services to 10 customers (DX 117 at 7).

26. As Finding 19 reflects, less than 10% of the ceramic welding furnace repair business went to Glaverbel-Fosbel competitors (other than Certek, which was acquired- by Fosbel in 1991). Glaverbel-Fos-bel had a procedure in place to handle an increase in market share of 10 to 20%, including hiring and training additional crews, transferring people among crews, scheduling the rotation and maintenance of equipment and providing for the necessary lead time to purchase additional equipment. In addition, Glaverbel-Fosbel’s expert took into consideration the capital cost of additional equipment.

27. This Court finds that Glaverbel-Fosbel did have the manufacturing and marketing capacity both to make the ceramic welding sales that were made by Northlake and to provide the ceramic welding services that were provided by Northlake.

4. Profits the Patent Owner Would Have Made

28. Glaverbel-Fosbel is claiming lost profits on Northlake’s sales to four customers. Fosbel had called on those customers before Northlake entered the ceramic welding market, and in some cases it had actually provided them with ceramic welding services before Northlake’s entry into that market (DX 117 at 10).

29. Glaverbel-Fosbel’s expert Dr. Lewis Koppel explained his lost profits calculations. Among the variables in those calculations was the size of the crew on a particular job. For example, Northlake frequently used only a single person ceramic welding crew (where the customer provided the additional crew members), while Glaverbel-Fosbel recommended a three person crew. Dr. Koppel’s initial calculations proceeded on the assumption that Glaverbel-Fosbel would have been able to provide a three person crew at the Glaverbel-Fosbel price even in those instances where Northlake provided a single person crew at a lower price, or even where Northlake provided a three person crew at a lower per-erewmember price. To assist in evaluating the financial impact of alternatives to that assumption, Dr. Koppel prepared a supplemental report (DX 148): a grid summarizing the lost profits (and reasonable royalty) calculations under different assumptions. Those different assumptions include things such as Glaverbel-Fosbel price vis-a-vis North-lake price, a three-person crew vis-a-vis a one-person crew and whether or not there were acceptable noninfringing alternatives. In addition, the fourth column of DX 148 is based upon Glaverbel-Fosbel’s ratable or relative market share.

30. These Findings are based on the assumptions (less favorable to Glaverbel-Fosbel than those in Dr. Koppel’s initial calculations) that Glaverbel-Fosbel would have made the sales at the Northlake price and with the Northlake crew-member size. It was agreed during the trial that Glaver-bel-Fosbel’s proposed findings and conclusions would be based on the right-hand column 4 of DX 148, with Glaverbel-Fos-bel reserving the right to file a memorandum supporting the use of any other sets of assumptions. They have done so, and these Findings have already accepted one such set by confirming that there were no acceptable noninfringing alternatives to Glaverbel-Fosbel’s patented ceramic welding. Although Glaverbel-Fosbel have also made a strong legal case for the added acceptance of Dr. Koppel’s initial assumptions referred to in Finding 29, these Findings reflect a more conservative (that is, lower) damages award predicated on the revised calculations in DX 148. In the interest of providing a complete record in the event that Glaverbel-Fosbel were to take an appeal or cross-appeal, footnotes to appropriate Findings will be included to set out what the damages figures would be if Dr. Koppel’s initial assumptions had been fully accepted instead.

31. Northlake has not challenged any of Glaverbel-Fosbel’s specific calculations as such (thus Northlake did not argue that its actual sales records reflected a number different from that used by Glaverbel-Fos-bel’s expert). Instead Northlake disputes the underlying factual assumptions. It is therefore unnecessary to make specific findings on the unchallenged items such as the total amount of ceramic welding material that Northlake actually sold or the total number of crew days for which Northlake actually charged its customers.

32. This Court fully credits the report and testimony of Glaverbel-Fosbel’s expert Dr. Koppel, whose analysis included a number of conservative (and valid) components — for example, that Fosbel would have to purchase additional machines in order to make the sales for which lost profits are sought, even though its witness Bacon testified that additional machines were in fact available. In addition, Dr. Koppel made a relative market share reduction by reducing the Northlake sales numbers ratably to take into account the sales during the three year period (1989-91) that Certek was in the market before it was acquired by Glaverbel-Fosbel. Dr. Koppel also determined the incremental costs (those that would have increased if Glaverbel-Fosbel had made the additional sales) associated with the lost profits analysis. Examples are the cost of additional ceramic welding powder and the cost of additional direct labor involved in providing the ceramic welding repair service, as contrasted with general overhead costs (items that do not increase because of increased sales, such as rent and the cost of new product development). Finally, Dr. Koppel has properly accounted for some costs that are partially incremental.

33. As for the time period for which Glaverbel-Fosbel are entitled to recover damages, see Conclusion 2. Based on the foregoing Findings, this Court finds that Glaverbel-Fosbel’s lost profits amount to $694,231. 5

Reasonable Royalty Analysis

34. Glaverbel-Fosbel also seek a reasonable royalty on Northlake’s sales to its remaining customers. Such a reasonable royalty is determined based upon a hypothetical negotiation between the patent owner and the infringer, at the time the infringement began, with both parties to the negotiation assuming that the patent is valid and would be infringed but for the license. Courts consider a variety of factors (known as the Georgia-Pacific factors, see Conclusion 15) as part of the reasonable royalty analysis.

35. There is an existing license between Glaverbel and Fosbel that initially called for a 10% royalty, a figure later reduced to 9%. In a 1995-96 independent audit that license was found to represent an “arms’ length transaction” (DX 117 at 15). This Court rejects both 9% and 10% as a reasonable royalty as of the time that Northlake’s infringement began on December 20, 1988, the issue date of the ’468 Patent — not only because of the time differential involved between the negotiation date of the actual license and the theoretical negotiation date of the hypothetical Northlake license, but also (and importantly) because in addition to receiving a royalty, Glaverbel received a share of profits by reason of its approximately ownership position in Fosbel (a financial benefit that Glaverbel would not of course have derived from Northlake sales under the hypothetical Northlake license).

36. According to Northlake’s testimony in its unsuccessful antitrust claim against Glaverbel-Fosbel in the Indiana case, Northlake had a projected gross profit of 40%, a figure that may have dropped to the 30%-plus range in the mid-1990s (DX 140 at 221 22). Northlake’s net profit would have been at least in the mid-teen percentage range if Northlake had not sustained its legal expenses in its dispute with Glaverbel-Fosbel (id. at 222-23).

37. If Northlake had taken a license at the beginning of 1989, it would not have sustained the legal expenses from its dispute with Glaverbel-Fosbel and also would not have had the distractions resulting from such litigation. Instead it would have had the opportunity to develop its business further, which would have increased its revenue without increasing its fixed overhead costs. Moreover, at least as early as 1987 each of Northlake’s three principals at the time (Hamilton, May and Frank Zlamal (“Zlamal”)) was earning in excess of $100,000 (DX 140 at 537-23 through 538-11). This Court concludes that if Northlake had taken a license its net profit would have exceeded the mid-teen percentages without any need to make any change in its manner of operation.

38. This Court credits the Koppel report and explanation and finds that a reasonable royalty under all the circumstances was Dr. Koppel’s original 14% figure, as adjusted to 13.1% to correct a fractional error (see Finding 39). Other than purportedly challenging Dr. Koppel’s report on cross-examination (a challenge that this Court finds to have been unpersuasive), Northlake presented no evidence as to the numerical amount of a reasonable royalty.

39. According to the alternative calculation in DX 148, at a 14% royalty rate and using the actual Northlake revenue figures the reasonable royalty payable to Glaver-bel-Fosbel for the Northlake use of the patented inventions would have been $105,552. Because Dr. Koppel’s testimony revealed a fractional error in that figure, this Court finds the reasonable royalty rate would have been 13.1% rather than 14% (Tr. 485), for a reasonable royalty calculation of $98,767. 6

Prejudgment Interest

40. Although Northlake disputes the appropriateness of any prejudgment interest award, the parties have agreed that if prejudgment interest is in fact awarded, the rate should be prime plus 1%. Prejudgment interest should ordinarily be awarded in patent cases under Section 284, absent some justification — such as where the patent owner is responsible for undue delay in prosecuting a lawsuit — for withholding such an award (General Motors Corp. v. Devex Corp., 461 U.S. 648, 654-57, 103 S.Ct. 2058, 76 L.Ed.2d 211 (1983)).

41. Northlake has offered no evidence upon which a denial of prejudgment interest could be predicated, arguing instead that it had a good faith belief of patent invalidity. And Northlake’s further position in its June 7, 1999 Response to the Glaverbel-Fosbel submission — that prejudgment interest should not be awarded because “Glaverbel delayed filing this infringement action for about 8 years”' — is totally unpersuasive. It was after all Northlake’s ultimately rejected contention that the enforcement of the patents at issue violated the antitrust laws that effectively forced any further efforts along those lines (hence forcing the delay) until that groundless antitrust claim was dispatched. Northlake cannot complain of such a self-inflicted wound. Because the purpose of awarding prejudgment interest is to compensate the patent owner, not to punish the infringer, factors such as asserted good faith on the part of the in-fringer and such as the untenable argument regarding delay are irrelevant to a prejudgment interest analysis.

42. Here the record reflects no reason not to award prejudgment interest. Prejudgment interest was calculated by Dr. Koppel at the prime rate, then recalculated at prime plus 1% compounded quarterly. Because the total damage award under the preceding Findings is the sum of lost profits plus a reasonable royalty ($694,231 plus $98,767 or a total of $792,998), prejudgment interest on that amount .through August 1, 1998 was $489,866. According to the reported figures in the Wall Street Journal, of which this Court takes judicial notice, the prime rate since 1998 has been 7.75%. Additional prejudgment interest at 8.75% is awarded on the base of $792,998, compounded quarterly from August 1, 1998 to the date of entry of judgment in accordance with these Findings and Conclusions.

Allocation Among Northlake Parties

43. Northlake suggests that May should not share liability for the period beginning January 1,1995, when he ceased to work through Northlake marketing. As Finding 1 reflects, Northlake Industries, Inc. was formed by Hamilton in 1995. But all billing was still made by Northlake Marketing, and all checks were deposited into the Northlake Marketing bank account and were reflected in the. Northlake Marketing tax returns. There is no question that Northlake Marketing and Hamilton are jointly and severally liable on the entire judgment, while May is jointly and severally liable on all aspects of liability through December 1994. Hence the only issue as to possible allocation of liability is whether May is also liable for the infringement that occurred after January 1, 1995.

44. During the relevant time period through and including the grant of, the permanent injunction in this case in 1997, both May and Hamilton were officers and directors, and each was a 44% shareholder, of Northlake Marketing. Northlake Marketing’s practice was that every owner was-an officer and board member — there were no officers or board members who were not owners. Northlake Marketing’s 1989 federal income tax return showed that its liabilities exceeded its assets, and the lia bilities continued to exceed the assets thereafter.

45. Northlake failed to establish the legal effect of May’s claimed resignation on his status with the corporation, for there was no amendment to Northlake’s by-laws eliminating the requirement that all shareholders were officers and directors. Nor did Northlake’s corporate record book (DX 109) even reflect May’s claimed resignation or any acceptance of such claimed resignation by Northlake.

46. If Northlake had prevailed on its antitrust claims, either in this Court or on appeal from the adverse decision in the Indiana case, May as a 44% owner stood to reap a substantial gain. Furthermore, May’s activities on behalf of Northlake at least through 1994, including his initial work from 1984 through 1988, were part and parcel of putting Northlake into the infringing position in which it found itself.

47. All of Northlake, May and Hamilton have been found liable for infringement. None of them has provided this Court with a sufficient factual and legal basis for allocation of responsibility. Accordingly all three of those parties are jointly and severally liable for the entire award in this case. 7

Enhanced Damages Pursuant to Section 281

48. Glaverbel-Fosbel seek an award of enhanced damages. For that purpose this Court considers the issues of (a) willful infringement and (b) bad faith conduct during litigation.

1. Willful Infringement

49. Northlake seeks to defend against enhanced damages on the theory that there was no willful infringement because it had subjective good faith beliefs (a) that its ceramic welding powder came first, so that the patents were invalid, and (b) that the invention had been on sale by Glaver-bel-Fosbel itself more than one year prior to the filing date of the patent applications, thereby invalidating the patents on an alternate basis. Because the acceptance of a subjective belief, however unfounded or unreasonable, as the basis for defeating a claim for willful infringement could entirely eviscerate the concept of willfulness, an objective standard must be used instead. And that finding is not changed by North-lake’s June 7, 1999 Response that seeks to recast its defense in a manner that makes no difference on the issue of its willfulness.

50. On the subject of Northlake’s willfulness vel non, the oral opinion of its counsel John Brezina (“Brezina”) was provided in a discovery answer and was read into the record. But there is no evidence (a) that Brezina ever made an independent investigation into the facts or (b) that Bre-zina ever attempted to verify the facts or (c) importantly in terms of whether North-lake could reasonably rely on such an opinion of counsel, that Northlake ever provided Brezina with any documents (as opposed to oral information) about its alleged prior powder.

51. Northlake’s previously rejected contention that its initial refractory powder had a size range spread factor of more than 1.2, thus invalidating the patent claims, must be examined for the purpose of considering enhanced damages. In that regard Northlake received its technical information from Zlamal and former Fosbel employee Mickey Whisman (“Whisman”), but from no other source (DX 141 at 7-8, 128). During the 1986-87 time frame Northlake received a sample of Fosbel ceramic welding powder and had that analyzed (DX 140 at 107-09). 8 During the period from 1984 through 1988 Northlake performed literally hundreds of experiments varying the components and percentages of ingredients used in its ceramic welding process. But with all of that trial- and-error activity, Northlake had no knowledge whatever of the size range spread factor of its powders at the time of its use (DX 141 at 13-14). There is no evidence that anyone acting on Northlake’s behalf attempted at any time before trial to obtain verification from its supplier Harbison-Walker as to the size range spread factor of the two different products used by Northlake, the originally employed Flintgrain 1604 or the smaller Calcined Quartz (the other product to which it shifted by early 1988). To the contrary, in response to Requests for Admissions Northlake agreed that it had no documents that specifically identified the size range spread factor of any refractory particles sold by Harbison-Walker (DX 123 at 1, 2). Northlake admittedly had “no concern about particle size” (DX 140 at 411).

52. As Finding 51 reflects, the proofs adduced by Northlake during the liability (validity) phase of the trial also did not indicate any effort to obtain size range spread factor information from Harbison-Walker. Indeed, when the Harbison-Walker witness Nale appeared at trial and produced a document, and when Glaver-bel-Fosbel then objected to its use because it had not been previously produced in discovery, Northlake argued that it had just received the document. Nor is there any evidence that Northlake sought to depose Harbison-Walker or to subpoena documents or to take any steps, other than to rely on oral testimony and purported recollection, to support its contention as to the details of its prior ceramic welding material.

53. This Court finds, based upon the foregoing, that there could not be a competent opinion of counsel, and certainly not an opinion on which Northlake (knowing its counsel’s total lack of knowledge of the critical relevant facts) could reasonably rely, that the prior Northlake powder had a size range spread factor of at least 1.2. This Court further finds that, given the Northlake changes in its formulation percentages and the vagaries of the amounts of ingredients it used, coupled with its knowledge that its counsel was unaware of all such matters, there could not have been an objective good faith reliance on its part on any opinion of counsel.

54. Northlake’s second branch of its attempted advice of counsel defense is the alleged prior commercial (i.e., public) use of the powder by Glaverbel-Fosbel, which allegedly triggered the on-sale bar of Section 102(b) to invalidate the Glaverbel patents. Brezina’s asserted opinion in that regard was based on documents produced by Glaverbel-Fosbel during the Indiana case under a protective order (DX 119) that restricted its use to that lawsuit. That being the case, reliance on that document would appear to be improper as a matter of public policy. But even if that were not so, Brezina’s oral opinion was no better than cursory and conclusory, and was not worthy of reliance, for it clearly did not include any investigation into the issue of experimental use, which is a universally understood and critical exception to the on-sale defense to patent validity. Brezina, an experienced patent lawyer, had to be well aware that a competent opinion of counsel would at a minimum have addressed that exception to the on-sale defense and would have advised that further investigation was necessary.

55. Moreover, Northlake’s advice of counsel defense must be considered in the context of the Indiana case and the prior Belgian litigation among the parties. Glaverbel and Fosbel sued Northlake and its principals for infringement of the ’560 and ’022 Patents in August 1988 in the Indiana Court in Civil Action No. H88 383. Northlake counterclaimed, asserting patent invalidity, patent noninfringement, patent misuse, unfair competition and antitrust violations by Glaverbel and Fosbel and inequitable conduct by Glaverbel. Then in November 1989 Northlake sought reexamination of the ’560 and ’022 Patents pursuant to Section 302, a statutory provision that permits new prior art to be brought to the attention of the Examiner. On September 3, 1991 the United States Patent and Trademark Office (“PTO”) confirmed the patentability of three claims of the ’560 Patent (which originally had 22 claims), and on April 16, 1991 the PTO confirmed the patentability of all original 14 claims of the ’022 Patent plus 14 new claims.

56. Before trial on the ’560 and ’022 Patents, the Indiana Court granted North-lake’s motion for summary judgment of noninfringement. Trial in the Indiana Court then proceeded on Northlake’s counterclaims. At the conclusion of Northlake’s proofs in March 1992, the Indiana Court ruled in response to a Glav-erbel-Fosbel Rule 52(c) motion that Northlake had not established (1) invalidity of the ’560 and ’022 Patents or (2) inequitable conduct by Glaverbel or (3) antitrust violations or unfair competition by Glaverbel or Fosbel. On January 23, 1995 the United States Court of Appeals for the Federal Circuit (a) affirmed the decision of the Indiana Court as it related to validity, lack of inequitable conduct, lack of antitrust violation and lack of unfair competition, (b) vacated in part the summary judgment decision as it related to infringement and (c) remanded the matter to the Indiana Court. Thus Northlake was not able to make out a prima facie case on any of its allegations during the Indiana Court trial.

57. In January 1989 Glaverbel sued Northlake in Belgium on three Belgian patents, the first two of which corresponded to the ’560 and ’022 Patents and the third of which corresponded to both the ’468 and ’084 Patents. In February 1995 the Belgian court found in favor of Glaverbel and against Northlake on the validity and infringement of the three Belgian patents and made a partial award of 2 million Belgian francs against Northlake. Northlake has not paid any part of the Belgian judgment.

58. Any prudent business person, in deciding objectively whether to rely on any advice of counsel (importantly in determining if such advice were indeed competent), should take into account the consistent inability to make out a prima facie case on the various theories that had been advanced in the Indiana case. Moreover, a series of adverse summary judgment decisions in this Court, where Northlake (despite the benefit of affidavits submitted in opposition to summary judgment) was not able to make out a prima facie case on any of its theories, would have confirmed to the prudent business person that it was engaged in nothing more than blind reliance on its counsel’s unsupported speculation.

59. Northlake’s second patent invalidity theory in this Court (see Finding 48(b)) was that the published prior art, particularly the powder of Example VII of Glav-erbel-Fosbel’s own ’022 Patent, had a size range spread factor of at least 1.2 and thus invalidated the claims of the two patents now in suit. For one thing, when North-lake lost that issue in early 1995 in the Belgian litigation, a prudent business person would not have continued its infringement, or at a minimum would have sought a supplementary opinion o'f counsel as to the impact of the adverse Belgian decision. But independently of Northlake’s lack of success on that issue in Belgium, the proofs presented to this Court also did not amount to a prima facie case of invalidity. As noted in this Court’s November 13, 1997 opinion (986 F.Supp. 471, 476), North-lake did not explain which claims of the patents were actually alleged to be invalid on this basis. In fact Northlake’s expert Dr. Nash testified that his theory did not apply to every claim of the ’468 Patent or the ’084 Patent. Furthermore, the ’022 Patent had been considered by the PTO in evaluating, and in not finding invalid, the ’468 and ’084 Patents (DX 101 and DX 102 at 1 of each exhibit under “References Cited”).

60. Based on the total absence of proof on the details of Northlake’s own prior powders, an equally total absence of proof that the prior Glaverbel-Fosbel powder was anything but experimental, and the fundamental shortcomings of Northlake’s purported remaining proofs on patent invalidity, the Court finds that this has not been a close case and that Northlake cannot claim good faith reliance on any assert-edly competent advice of counsel.

2. Conduct During Litigation

61. It is undisputed that even after being sued in Belgium in 1989, and with knowledge at least since 1990 of the corresponding ’468 and ’084 Patents — indeed, even after an adverse judgment in the Belgian litigation in February 1995— Northlake made no effort simply to change the size range spread factor of its refractory powder to avoid further infringement, until Northlake’s activities were enjoined by this Court in late 1997.

62. Moreover, Northlake’s litigation conduct here, as noted in the docket entries (cited “Dkt. — ”) of which this Court takes judicial notice, reflects a continued pattern of arbitrary and intransigent conduct. For one thing, there was no basis to include additional defendant Foseco in the suit to begin with — in response to the Fo-seco motion to dismiss (Dkt.36, 37) North-lake proceeded with the deposition of Anthony Money of Foseco and thereafter, rather than submitting a dismissal with prejudice, required this Court to rule on the Foseco motion (Dkt.56). In addition, Northlake sued Glaverbel in this action for a declaration of invalidity of the ’022 Patent even while Northlake’s challenge to that patent was on appeal from the Indiana Court — thus engaging in the improper splitting of a claim. In response to a motion to dismiss as to the ’022 Patent (Dkt.26, 27), Northlake filed its opposition (Dkt.57) but presented no valid legal grounds for maintaining a challenge to the ’022 Patent in this action, triggering this Court’s opinion granting the motion and dismissing as to the ’022 Patent (Dkt.59).

63. Another unreasonably proffered aspect of the present action was Northlake’s allegation of Glaverbel-Fosbel antitrust violations and unfair competition. Having lost its antitrust claims at trial in Indiana in March 1992, Northlake had no basis for Northlake to reassert those claims in this Court in April 1992. 9 Thus Northlake had no proper basis to oppose as it did (Dkt.61) Fosbel’s motion in this Court to dismiss as to all events occurring before the end of the March 1992 Indiana trial (Dkt.42, 43). And because the conduct alleged to violate the antitrust laws was no different before that 1992 date than after that date, there was equally no basis for Northlake to oppose the motion to dismiss all of its antitrust claims. Then after this Court dismissed the remainder of the Northlake antitrust and unfair competition claim (Dkt.82), there was no basis for Northlake to argue that it did not understand the ruling and seek reconsideration, resulting in further briefing and a further opinion by this Court (Dkt.126).

64. As to Northlake’s inequitable conduct allegations (Count III), which this Court dismissed on summary judgment on March 11, 1997 (958 F.Supp. 373), they were based on an impermissibly stretched reading of Ninth Circuit law — a reading that was contrary to the express and controlling ruling of the United States Court of Appeals for the Federal Circuit. This Court similarly found Northlake’s nonin-fringement claim (Count I) to be based on an improper argument, for Northlake was simultaneously contending that its powder had come first and thus met the language of the patent claims. Finally, Northlake’s laches argument was equally improper, for Northlake conceded the absence of harm, one of the two requirements for a laches defense.

65. In addition to all of that substantive misconduct, Northlake repeatedly failed to cooperate in discovery in this action, forcing the filing of several discovery motions by Glaverbel-Fosbel. It appears that some documents were never produced. Further, there was no justification for Northlake’s position that a new entity, Northlake Industries, was involved after May had left Northlake Marketing, because as found earlier all business continued to be carried out in the name of the original Northlake Marketing. Those aspects of Northlake’s conduct ultimately compelled this Court to enter an order excluding any documents that Northlake had not produced during discovery. As stated earlier, despite that order North-lake attempted to use a nonproduced Har-bison-Walker document at trial.

66. Northlake used certain Fosbel documents, 10 produced under a protective order in the Indiana case, to support its summary judgment motion charging that the Glaverbel patents were invalid for allegedly violating the on-sale bar. Even though that protective order prohibited use of the documents other than in the Indiana litigation (DX 119), Glaverbel-Fos-bel did not raise the issue initially because the documents were part of a Northlake claim of fraud and Glaverbel-Fosbel did not want to be accused of attempting to cover up the purported fraud. With no on-sale bar having been found, Glaverbel-Fosbel may now appropriately object to Northlake’s tactics in that respect. Similar tactics were repeated during the damages phase of the trial, with Northlake attempting to rely on another exhibit (PX 66) that had been produced in the Indiana case under the same restrictions.

67. After this Court had denied North-lake’s summary judgment motion urging the invalidity of the ’468 and ’084 Patents for violation of the on-sale bar, Northlake took no discovery from the inventors and no further discovery from either Glaverbel or Fosbel, notwithstanding this Court’s ruling in the summary judgment opinion that there were defects or gaps in the Northlake proofs. Accordingly, after the close of discovery this Court granted the Glaverbel-Fosbel motion asserting that the patents were not invalid on the basis of the on-sale bar.

68. This Court does not have to address the applicability or inapplicability of Rule 11 or 28 U.S.C. § 1927 as such to determine that the Northlake litigation conduct was not in good faith for purposes of Section 284. It is significant in the context of the overall litigation that North-lake took almost no discovery. It appears that only Anthony Money of Foseco was deposed on the question of the propriety of Foseco being a party, while only Charles Zvosec of Fosbel was deposed on the question of the 20 samples — no other testimony of Glaverbel-Fosbel personnel, nor any Glaverbel-Fosbel responses to interrogatories, nor any Glaverbel-Fosbel responses to requests for admissions were used at either trial in this Court. Northlake consistently could not make out a prima facie case either on summary judgment issues or at trial: It offered an expert witness who apparently had not studied all the patent claims, it had no witness to explain its patent invalidity theories as to all patent claims, and it had no witness to explain the specific patent claims for which its anticipation (Section 102) or obviousness (Section 103) defenses purportedly applied.

69. Based on the totality of Findings 49 through 68 and comparable subsidiary findings in the earlier Findings, this Court finds that the Northlake infringement has been willful, that this litigation was conducted in bad faith and that Northlake’s extrajudicial conduct in not changing its ceramic welding powder after repeated adverse decisions amounts to bad faith conduct as that term has been interpreted under Section 284.

Attorneys’ Fees Under Section 285

70. Glaverbel-Fosbel also seek an award of attorneys’ fees under Section 285. As Conclusions 30 and 31 reflect, the case-law teaches that an attorneys’ fees award requires a finding of an “exceptional case” and that such a finding can properly be based on either willful infringement or bad faith conduct. Where both of those factors are present (as is true here), the basis for an “exceptional case” finding becomes even stronger.

71. In light of the prior Findings of willful infringement and bad faith conduct, further aggravated by the fact that North-lake consistently could not make a prima facie case either at the summary judgment stages or at trial, this Court exercises its discretion and finds this to be a truly exceptional case justifying an award of attorneys’ fees.

Conclusions of Law

Jurisdiction

1. This Court’s prior conclusions as to jurisdiction continue to apply during the proceedings leading to these Findings and Conclusions.

Damages — General

2. Under Section 284 a patent owner is entitled to damages “adequate to compensate for infringement but in no event less than a reasonable royalty, for the use made of the invention by the infringer, together with interest and costs as fixed by the Court.” Among the different elements of damages for which the patent owner may recover such compensation are (1) lost profits on sales that the patent owner lost as a result of the infringement (State Indus., Inc. v. Mor-Flo Indus., Inc., 883 F.2d 1573, 1577 (Fed.Cir.1989)) and (2) a reasonable royalty on infringing sales even if the patent owner would not have obtained those sales (see Bio-Rad Labs., Inc. v. Nicolet Instrument Corp., 739 F.2d 604, 616-17 (Fed.Cir.1984)). As for the time period for which Glaverbel-Fosbel are entitled to recover damages, this Court has already held that their counterclaim (because it is a compulsory counterclaim under Rule 13(a)) dates back to the filing of their Answer by virtue of Rule 15(c)(2)(958 F.Supp. at 376 (1997); accord, 6 Charles A. Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure: Civil 2d § 1430, at 228 (2d ed.1990)). Despite this Court’s invitation in its 1997 opinion, Northlake has offered nothing to support the notion that the Section 286 reference to “the filing of the... counterclaim” should be read in terms of the file stamp date rather than the effective date mandated by Rule 15(c)(2), while Glaverbel-Fosbel have cited one early case whose holding conforms to the more normal (and more sensible) reading that gives full effect to Rule 15(c)(2)— indeed, if Northlake’s position were accepted, a situation could arise in which a later-tendered compulsory counterclaim could be filed but could confer no relief at all on the injured patentee. All of that being so, the six-year limitations period specified in Section 286 does encompass the entire period of the Northlake infringement addressed in these Findings and Conclusions. 11

3. Damages need to be proved only by a preponderance of the evidence, and the patent owner’s burden of proof is not absolute but rather one of reasonable probability. This Court is free to use its discretion in choosing a method for calculating damages as long as the measure of damages is just and reasonable (Korn Corp. v. Wilco Marsh Buggies & Draglines, Inc., 761 F.2d 649, 653-54 (Fed.Cir.1985).) Because the computation of damages is not always amenable to precise determination, it is acceptable if the evidence shows the extent of damages as a matter of just and reasonable inferences, even though the result is only approximate (Paper Converting Mach. Co. v. Magna-Graphics Corp., 745 F.2d 11, 22 (Fed.Cir.1984)). Any doubt as to the correctness of a damages award is to be resolved against the infringer (State Indus., 883 F.2d at 1577). If the patent owner is able to establish a reasonable probability that it would have made only some of the infringer’s sales but for the infringement, the damages award may be in the amount of lost profits to the extent so established plus a reasonable royalty for the remainder of the sales (id.).

Damages — Lost Profits

4. It is unnecessary to prove lost profits with absolute certainty. It suffices to prove a reasonable probability that Glaver-bel-Fosbel would have made the sales of the ceramic welding powder and would have provided the ceramic welding services (one of the two patents is for the powder and the other for the method of repair). There is no need to negate all possibilities that a purchaser might have bought a different product or might have forgone the purchase altogether (State Indus., id.).

5. Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152 (6th Cir.1978), cited with approval in various Federal Circuit cases (see, e.g., Rite-Rite Corp. v. Kelley Co., 56 F.3d 1538, 1545 (Fed.Cir.1995) (en banc)), identifies four factors to be considered (if and when present) in a lost profits analysis. Those four factors are (a) a demand for the patented product or method, (b) the absence of acceptable noninfringing substitutes, (c) the manufacturing and marketing capability of the patent owner to exploit the demand and (d) the amount of profit the patent owner would have made on the infringing sales. Under the Panduit-taught analysis, evidence of those four factors permits a court to draw the reasonable inference that the lost profits claimed were in fact caused by infringing sales, thus establishing the patentee’s prima facie case of “but for” causation.

6. Here the demand for the patented invention was established, among other things, by the infringers’ sales and the infringers’ sales projections that were presented in the Indiana case and were part of DX 140 and DX 141. This Court concludes that there was a demand for the patented invention.

7. There are no acceptable non-infringing substitute methods for ceramic welding. None of the other techniques or methods has the considerable advantages of the patented invention. In addition, even ceramic welding as practiced by other competitors was not a substitute for the patented process and powder. In that respect, because the other competitors’ ceramic welding powders lack the attributes of the patented invention, they are not acceptable alternatives (TWM Mfg. Co. v. Dura Corp., 789 F.2d 895, 901 (Fed.Cir.1986)). Mere existence of such a competing alternative does not make it an acceptable substitute (id.). Hence a court should apply a two-supplier market approach when products other than those of the infringer lack the advantages of the patented invention (Kalman v. Berlyn Corp., 914 F.2d 1473, 1484 (Fed.Cir.1990); Uniroyal, Inc. v. Rudkin-Wiley Corp., 939 F.2d 1540, 1545-46 (Fed.Cir.1991)). Under that approach the damage calculation should be based on the premise that Glav-erbel-Fosbel would have made all of the Northlake sales, not just a ratable or relative market share portion of those sales.

8. If the ceramic welding powders of the competitors other than Northlake did not in fact infringe on the patents in suit, they did not provide the benefits or advantages of the patented invention. For damage calculation purposes the question is not whether those powders were competing products, but rather whether there were acceptable substitutes (see, e.g., Minco, Inc. v. Combustion Eng’g, Inc., 95 F.3d 1109, 1119 (Fed.Cir.1996) (per curiam)). Moreover, Panduit, 575 F.2d at 1160-62 teaches that the argument of “acceptable substitutes” must be viewed as of limited influence where the infringer knowingly made and sold the patented product for years while ignoring the claimed “substitute” (Stryker Corp. v. Intermedics Orthopedics, Inc., 96 F.3d 1409, 1418 (Fed.Cir.1996)). Here Northlake continued with its infringing powder for nine years (December 1988 through 1997), rather than making the purportedly simple change to a refractory powder having a size range spread factor of less than 1.2.

9. Finally, the issue of assertedly acceptable noninfringing ceramic welding alternatives may be considered in the context of the Northlake antitrust claims that Glaverbel-Fosbel have assertedly monopolized the ceramic welding repair market. But that would mean that Glaverbel-Fos-bel must have monopoly'power (which is not presumed in the patent-antitrust interface). To look at the converse of that possibility, Abbott Labs. v. Brennan, 952 F.2d 1346, 1355 (Fed.Cir.1991) has quoted Justice O’Connor’s concurrence in Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 37 n. 7, 104 S.Ct. 1551, 80 L.Ed.2d 2 (1984) (emphasis added): 12

A common misconception has been that a patent or copyright, a high market share, or a unique product that competitors are not able to offer suffices to demonstrate market power. While each of these three factors might help to give market power to a seller, it is also possible that a seller in these situations will have no market power: for example, a patent holder has no market poiuer in any relevant sense if there are close substitutes for the patented product.

It follows as a logical matter that if a patentee does possess market power (monopoly power), there cannot be close substitutes for the patented product. And Northlake has indeed stated (it has actually conceded in its pleading, albeit while looking in a different direction) that there is monopoly power, a concession that can be used against Northlake under Fed. R.Evid. 801(d)(2)(A).

10. Accordingly the lost profits analysis applies to all of the sales made by Northlake to its customers with whom Fosbel had any contact before the North-lake infringement. That analysis applies without any ratable reduction to take into account any noninfringing ceramic welding.

11. Based upon the undisputed testimony, Glaverbel-Fosbel had the marketing and manufacturing capability for a 10% to 20% increase in sales, and Glaverbel-Fosbel’s expert Dr. Koppel took the cost of new machinery into consideration in his calculations. Glaverbel-Fosbel had the full marketing capability to have made all of the sales that Northlake made, if North-lake had in fact not made them.

12. Panduit’s fourth factor calls for a calculation of the amount of the additional, or incremental, profit that Glaverbel-Fos-bel would have made if Northlake had not infringed. That calculation is made by (1) determining the amount of additional sales Glaverbel-Fosbel would have made and (2) subtracting from it the additional costs Glaverbel-Fosbel would have incurred in order to make the additional sales. That incremental profit approach is well established in patent damages cases (Paper Converting, 745 F.2d at 22).

13. This Court has approved Dr. Koppel’s expert report with the modifications reflected in the- Findings. It concludes that the lost profits award for Glaverbel-Fosbel is $694,231 (see Finding 33). 13

Damages — Reasonable Royalty

14. “Reasonable royalty” is the amount that a willing licensee would pay, and a willing licensor would accept, before the commencement of the infringement (Section 284; Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075, 1079 (Fed.Cir.1983)).

15. Georgia-Pacific Corp. v. United States Plywood Corp., 318 F.Supp. 1116 (S.D.N.Y.1970), cited in such cases as Rite-Hite, sets forth the factors generally to be considered (if and when present) in a reasonable royalty analysis. Dr. Koppel’s report correctly analyzes the first fourteen factors, and that analysis is accepted by this Court. And the fifteenth factor, about which Dr. Koppel testified, is the amount upon which a licensor (such as the paten-tee) and a licensee (such as the infringer) would have agreed if both had been reasonable in trying to reach a voluntary agreement — that is, the amount that a prudent licensee who desired, as a business proposition, to obtain a license to manufacture and sell a particular article embodying the patented invention would have been willing to pay as a royalty and yet be able to make a reasonable profit, and that would have been acceptable by a prudent patentee who was willing to grant a license (Georgia-Pacific, 318 F.Supp. at 1120).

16. Any reasonable royalty adequate to compensate Glaverbel-Fosbel for any infringing sales made by Northlake for which Glaverbel-Fosbel is not awarded lost profits should take into account the incremental profit margin for Glaverbel-Fosbel at the time infringement began (Rite-Hite, 56 F.3d at 1554). Here North-lake’s infringement began when the ’468 Patent issued on December 20, 1988. Hence the period of time closest to that date for which a statistically valid royalty rate can be computed is 1989, when the Glaverbel-Fosbel incremental profit margin was 16.2%. Next that incremental profit margin is adjusted for relative market share, and Glaverbel-Fosbel had 85.69% of the coke oven repair business in 1989. On that basis a reasonable royalty rate would be in the range of 16.2% x 85.69% or 13.9%, a figure reduced to 13.1% to adjust for a fractional error (see Finding 39).

17. Northlake’s net profit was in the mid-teen percentage range (without taking legal fees in its dispute with Glaverbel-Fosbel into account). But in addition to the substantial salaries paid to Northlake’s three principals, once Zlamal left North-lake in July 1989 he received continued compensation calculated at $2,200 per month plus six additional installments of $10,000 each (DX 140 at 227-4 to 227-15).

18. This Court concludes that taking into account Northlake’s gross profit in the 30% to 40% range and its net profit in the mid-teen percentage range, plus the substantial compensation paid to its principals, 13.1% is indeed a reasonable royalty rate. This Court concludes that the reasonable royalty payable, for the infringement is $98,767 (see Finding 39). 14

19. Awards of prejudgment interest are the rule, not the exception (Sensonics, Inc. v. Aerosonic Corp., 81 F.3d 1566, 1574 (Fed.Cir.1996), citing General Motors Corp. v. Devex Corp.). As Finding 41 states, the purpose of prejudgment interest is not to punish the infringer but to compensate the patent owner for its losses. Because an award of prejudgment interest is the rule, and because Northlake has proffered no evidence to justify an exception, this Court concludes that an award of prejudgment interest is proper.

20. Based on Findings 33 and 39, the total damages award (the lost profits portion plus the reasonable royalty portion) is $694,231 plus $98,767 or $792,998.

21. Northlake does not dispute that the prejudgment interest rate should be the prime rate plus 1%. Dr. Koppel calculated the damages (lost profits and reasonable royalty) by using that rate, compounded on a quarterly basis. With the adjustments reflected in Finding 42, the prejudgment interest calculated through August 1, 1998 was $489,866. This Court awards that amount of prejudgment interest through August 1, 1998. It also awards additional prejudgment interest at 8.75% compounded quarterly from August 1, 1998 to the date of entry of an order of judgment pursuant to these Findings and Conclusions.

Allocation of Liability

22. There is no justification for an allocation of liability among the North-lake parties. Despite the fact that May resigned from Northlake as of the beginning of 1995, it would be inequitable for him to abdicate his responsibilities as an officer-director, and thus to create a potential for avoiding further liability, after his years of contributing to the position in which Northlake found itself and his years of taking a substantial salary, yet to remain in a position in which he sought to reap benefits from any potential success of Northlake (in its antitrust claims). At a minimum, to avoid further liability for continued infringement May should have tendered his stock (or had the corporation cancel his stock) at the same time that he purportedly resigned his position as a Northlake officer and director.

23. As Northlake acknowledged, its liabilities exceeded its assets from 1989 to the present. Even though the technical knowhow of the company was considered a valuable trade secret by Northlake, May used some of that technical knowhow in his new business without any payment to Northlake. Northlake has not paid any part of the Belgian judgment, or even the taxable costs awarded by the Federal Circuit in Northlake’s unsuccessful appeal. It thus appears to this Court that Northlake has continued its infringing activities merely on the chance that it might recoup on its antitrust claims, but with neither the ability nor the intent to meet its legitimate financial obligations.

24. On balance, it would be inequitable to allow May to escape even partial liability for the damages award here, even though it is true that about 15% of the award, based on the proportion of North-lake sales in the years 1995 through 1997 to the total Northlake sales for 1998 through 1997, is attributable to those post-1994 years. For the reasons already stated, no allocation is appropriate.

Enhanced Damages

25. Under Section 284 a court may increase the damages up to three times the amount found or assessed by the factfinder. Any award of enhanced damages requires a showing of either willful infringement or bad faith (Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1460-61 (Fed.Cir.1998) (en banc)). As with Section 285 awards of attorneys’ fees in “exceptional cases,” where as here both of those factors are present the case for enhanced damages becomes even stronger.

26. Any infringer has the obligation to avoid infringement once it has knowledge of the patent. Several factors to be considered on the issue of willfulness are explained in Read Corp. v. Portec, Inc., 970 F.2d 816, 827 (Fed.Cir.1992), 15 and the single most important factor is whether or not there was reliance on a competent opinion of counsel. Other factors include (but are not limited to) whether evidence of infringement (or the magnitude of infringement) was concealed, evidence of copying, closeness of the case, duration of the infringement, remedial action by the infringer (if any) and the infringer’s motivation for harm.

27. There is an affirmative duty to obtain validity and infringement opinions (Great Northern Corp. v. Davis Core & Pad Co., 782 F.2d 159, 166-67 (Fed.Cir.1986)). Any such opinion must be sufficiently thorough to support the reasonableness of a client’s reliance on the opinion (Ortho Pharm. Corp. v. Smith, 959 F.2d 936, 944 (Fed.Cir.1992)), because the required showing is one of justifiable reliance on the opinion (see Datascope v. SMEC, Inc., 879 F.2d 820, 828 (Fed.Cir. 1989)). Hence a cursory or conclusory opinion does not suffice (Underwater Devices Inc. v. Morrison-Knudsen Co., 717 F.2d 1380, 1390 (Fed.Cir.1983); Kori Corp., 761 F.2d at 656), and oral opinions are not favored (Minnesota Mining & Mfg. Co. v. Johnson & Johnson Orthopaedics, Inc., 976 F.2d 1559, 1580 (Fed.Cir.1992)). When the situation here is judged against those legal requirements, this Court concludes that Northlake did not receive a competent opinion of counsel upon which a prudent business person could reasonably rely.

28. This Conclusion addresses the remaining factors for enhanced damages separately. There was a failure by North-lake to cooperate appropriately in discovery (cf. Russell Box Co. v. Grant Paper Box Co., 203 F.2d 177, 183 (1st Cir.1953)), its challenge to validity was not of substantial quality (cf. Delta-X Corp. v. Baker Hughes Prod. Tools, Inc., 984 F.2d 410, 413 (Fed.Cir.1993) (same requirement as to a challenge to infringement)), it did not discontinue its infringing activities or find a noninfringing alternative during the pen-dency of this suit, even after a series of adverse rulings (Read, 970 F.2d at 827), and it demonstrated a motivation to harm Glaverbel-Fosbel. For example, North-lake received technical information from former Fosbel employee Whisman (DX 141 at 8-1 to 8-6 and at 128-2 to 128-14) and had access to Fosbel pricing to the same customers where Northlake was bidding for work (id. at 11-17 to 11-19, 13-12 to 13-20; 14^18 to 14-23; 24^18 to 24r-22 and 26-18 to 26-25). Northlake’s corporate minute book (DX 109) contains no references, at any time from 1988 to the present, to the Glaverbel-Fosbel claims, to Northlake’s alleged defense, to the present litigation or to the claimed opinion of counsel.

29. Based on all of the foregoing, this Court concludes both that the infringement was willful and that Northlake acted in bad faith. It therefore awards additional enhanced damages of twice the amount of the actual damages award. Prejudgment interest will not be applied to the enhanced damages portion of the judgment.

Attorneys’ Fees and Costs

30. Any determination of whether a case is “exceptional” so as to be eligible for an award of attorneys’ fees under Section 285 is a two-step process. First the court must determine if the case is exceptional and then, if the answer is affirmative, it must decide whether an award of such fees is appropriate (Cybor Corp., 138 F.3d at 1460).

31. Exceptional circumstances (the first step) may be found if there was misconduct during litigation, vexatious or unjustified litigation or a frivolous suit (Bayer Aktiengesellschaft v. Duphar Int’l Research B.V., 738 F.2d 1237, 1242 (Fed.Cir.1984)). Willful infringement alone will suffice to support a finding of an “exceptional ease” for an award of attorneys’ fees, though it does not mandate such relief (Cybor Corp., 138 F.3d at 1461).

32. As already found and concluded, here there was both willful infringement and bad faith conduct. There was no basis for Northlake to commence the present suit encompassing all of the issues it sought to advance, and there was no basis for it to continue infringement (or continue with the suit) in the face of successive adverse summary judgment or exclusionary rulings. This Court concludes that Glaverbel-Fosbel have demonstrated by clear and convincing evidence that this is indeed an exceptional case within the meaning of the statute.

3. Having made that “exceptional case” determination, this Court exercises its discretion in favor of awarding attorney fees to Glaverbel-Fosbel. In that respect counsel for the parties are ordered to proceed in accordance with this District Court’s General Rules (“GR”) 46 and 47. Under Budinich v. Becton Dickinson & Co., 486 U.S. 196, 108 S.Ct. 1717, 100 L.Ed.2d 178 (1988) the pendency of proceedings looking to such an award does not affect the finality of the substantive judgment to be entered at an early date.

34. Finally, this Court awards taxable costs in favor of Glaverbel-Fosbel and against Northlake.

% % H* ‡ ❖

It is hereby ordered that judgment shall be entered on June 25, 1999 in favor of Glaverbel S.A. and Fosbel, Inc. and against Northlake Marketing & Supply, Inc., James Hamilton and Samuel May, jointly and severally, in the following amount:

1. $792,998, representing the sum of lost profits and a reasonable royalty; plus

2. prejudgment interest on that sum to and including that date of judgment; 16 plus

3. enhanced damages in the sum of $1,585,996 (2 x $792,998), without prejudgment interest.

In accordance with the Findings and Conclusions, Glaverbel, S.A. and Fosbel, Inc. shall also be awarded their reasonable attorneys’ fees (to be determined in accordance with GR 46 and 47) and are hereby awarded their taxable costs.

JUDGMENT ORDER

This Court’s lengthy June 10,1999 Findings of Fact and Conclusions of Law ended by anticipating the entry of a judgment order on June 25, 1999, with counsel for the parties being directed to have previously submitted a statement as to the agreed amount of that judgment (if agreement were possible). Counsel have indeed since succeeded in reaching an “Agree[ment] as to arithmetic only,” with “All other objections reserved” (a copy of their calculations is attached to this judgment order). In accordance with that agreement, it is' ordered that judgment be entered in favor of Glaverbel S.A. and Fosbel, Inc. and against Northlake Marketing & Supply, Inc., James Hamilton and Samuel May, jointly and severally, in the sum of $2,992,918.

1 . This is by a substantial margin the oldest pending case on this Court’s calendar — only one other multifaceted patent case comes within hailing distance.

2 . All further citations to Title 35 provisions will simply take the form "Section — ,” employing the numbering within Title 35.

3 . "DX” refers to the Glaverbel-Fosbel exhibits and "PX” refers to Northlake’s exhibits. Where numbers such as "223-24” or "224-9” appear in reference to a DX (see, e.g., the Finding 8 text at n. 4), they refer to the page and line citations in a documentary exhibit.

4 . DX 140 and DX 141 are excerpts from testimony by May and Hamilton, respectively, primarily from depositions and from trial testimony in the case between the parties ("Indiana case") in the United States District Court for the Northern District of Indiana (“Indiana Court").

5 . On the added assumption referred to in the last sentence of Finding 30, that figure would have been $863,219.

6 . On the added assumption referred to in the last sentence of Finding 30, that figure would have been $88,611 (reduced from the DX 148 figure of $94,699).

7 . According to the portion of Dr. Koppel’s report (DX 117) that allocates liability over the several years involved, only about 15% of the total liability was incurred after January 1, 1995 in any event.

8 . This evidence comes from DX 140 and 141, testimony during the period when Northlake was pursuing its antitrust claims against Glaverbel-Fosbel. To some extent Northlake sought to retreat from that position during the damages trial, suggesting that it did not receive technical information from Whisman and that the test of the Fosbel powder occurred later. But this Court discredits that attempted disavowal in favor of accepting the earlier testimony.

9 . If Northlake's appeal from the Indiana decision had been successful, the antitrust claims would have been remanded and complete relief would have been available in Indiana. If on the other hand the Northlake appeal proved unsuccessful, that would have completely disposed of the Northlake claim. In either event the matter plainly did not belong in this District Court.

10 . This was the "grid” or data of 20 "sam-pies.”

11 . Glaverbel-Fosbel’s Objections to North-lake's Proposed Conclusion 12 sets out additional grounds for reaching the same conclusion. This Conclusion's silence as to the express adoption of those additional grounds should not be mistaken as any adverse reflection on (let alone a rejection of) those grounds, which may indeed provide sound and independent bases for the result reached here.

12 . This Court's memorandum opinion and order dismissing Northlake's antitrust claims quoted from the same case (see 861 F.Supp. 653, 663 (N.D.Ill.1994)).

13 . See Finding 33 n. 5 for a potential alternative figure for that element of' the damages award.

14 . See Finding 39 n. 6 for a potential alternative figure for that element of the damages award.

15 . Read has been overtaken by Markman as to the handling of claim interpretation, but that of course has no impact on the principle stated in the text.

16 . In that respect counsel for Glaverbel-Fos-bel are ordered promptly to calculate, and to submit to Northlake’s counsel for prompt review, the amount of prejudgment interest from August 1, 1998 to the contemplated June 25 judgment date (and hence the total amount of prejudgment interest), and then counsel for the parties are ordered to submit a statement as to the agreed amount (or if not agreed, statements as to the parties’ respective calculations and the basis therefor) to this Court's chambers on or before June 21, 1999.