Appeals from a judgment of the District Court, Central District of California, that Rosemount's patent No. 3,440,525 ('525 patent) is valid and infringed by Beckman (No. 83-947), and from a judgment of that court that Beckman's post-trial infringement violated its injunction, but denying damages to Rosemount for that infringement (Nos. 83-1238, 83-1251). We affirm.

In 1965, when Charles Cardeiro made the inventions claimed in the '525 patent, Beckman was the leading manufacturer of pH meters, with 40% of the process pH market. Beckman had a continuous development program designed to maintain and preferably increase its market share. Universal Interlock, Inc. (Uniloc), Rosemount's predecessor in interest, began marketing the pH meter of the '525 patent shortly after Cardeiro invented it. By 1975, Uniloc's annual sales of the patented inventions had grown from zero to many millions of dollars, and Uniloc employees had increased from its 5 founders to 170. Uniloc sold $25,000,000 worth of the patented inventions.

Beginning in 1966, within a year of Uniloc's first sales, Beckman's engineers and salesmen were sending a stream of reports to Beckman about the superiority of Uniloc's pH meters over Beckman's and those of other manufacturers, about customer preference for the patented pH meters, and about sales lost to Uniloc. The reports and Beckman's internal memos praised the patented pH meters as offering "unique and advantageous features", as preferred by users because of no "downtime" and "complete lack of maintenance", and as "low in price".1

Having lost sales of the Model "J" it was selling when Cardeiro made his invention, Beckman unsuccessfully attempted to design and market improved pH meters. After studying designs it found in the prior art, Beckman developed a pH meter embodying Cardeiro's concept in 1968. It modified that design upon issuance of the '525 patent in 1969. By 1973, Beckman found its market share had dropped to 27%, due principally to the advent of the patented pH meter, which had captured 25% of the market and was selling at 10% less than Beckman's pH meter. At that point, after evaluating the pH meters of six manufacturers, including its own and Uniloc's, Beckman elected to make and sell the pH meters found to have infringed claims of the '525 patent.

Rosemount sued Beckman on May 4, 1978. Beckman denied infringement and asserted that the '525 patent was invalid under 35 U.S.C. Secs. 102(a), 102(b), 102(e), 102(f), 102(g), 103, and 112. Beckman also asserted invalidity for incorrect naming of an inventor and unenforceability for fraud on the Patent and Trademark Office (PTO) and for unclean hands because Rosemount prosecuted the suit when the art cited by Beckman was not cited in the PTO, because Rosemount's licensing practices were wrongful, and because an element of the invention was not invented by Cardeiro. Beckman counterclaimed for a declaratory judgment that the patent is invalid and not infringed.

An eight-day trial was conducted (October 20-28, 1981). Judge Waters filed a succinct but comprehensive Memorandum of Decision on February 2, 1983, (reported at 569 F.Supp. 934), entered 137 findings of fact and 10 conclusions of law, and signed a judgment on February 24, 1983. In those documents, the court held the '525 patent valid, found that Beckman's Models 940/A, 941/A, 940B, 940BF, 960A, 980, and 8700 pH meters (Models 940/A-941/A) infringed asserted claims 1-3, 8, and 12, found Beckman's infringement to have been willful, awarded Rosemount treble damages and its attorney fees, and enjoined infringement. An accounting was stayed pending appeal.

On May 13, 1983, Rosemount charged Beckman with violation of the injunction by making and selling another infringing pH meter (Model 960B). The court found infringement and held Beckman in contempt, but declined to award Rosemount damages for that infringement because Rosemount knew of Beckman's sales of the Model 960B throughout the trial and did not include it with the accused Models 940/A-941/A.

The invention found to have been appropriated is set forth in claims 1-3, 8, and 12:

A system for determining the pH of a variable solution comprising:

a pH sensitive electrode, including a high impedance material responsive to the hydrogen ion activity of the variable solution and an internal electrical conductor for generating a voltage at said internal conductor indicative of the pH of the variable solution; and

a DC amplifier circuit for developing a signal as a function of the voltage developed on said internal conductor, said circuit including a preamplifier stage having a field effect transistor mounted with said electrode and in close proximity to said high impedance material and with its gate terminal connected to said internal conductor to provide a low impedance signal output from its output terminals for transmission to a remote circuit responsive to said low impedance signal output for indicating the pH of the variable solution.

The system of claim 1 wherein:said field effect transistor is disposed in a heat exchange relationship with said high impedance material that subjects both said field effect transistor and said high impedance material to substantially the same temperature variations.

a probe structure engaging said pH sensitive electrode, said field effect transistor being disposed within said probe structure adjacent the pH sensitive electrode, and one end of said internal conductor being connected directly to the gate terminal of said field effect transistor.

said field effect transistor is operated with its drain-to-source bias current at the level where the gain characteristic is independent of temperature variation over an ambient temperature range.

In a system for measuring the pH of a circulating solution, an arrangement comprising:

a pH sensitive electrode including a material responsive to the hydrogen ion activity of the circulating solution for developing a voltage indicative thereof, a hollow insulating member mounting said material for immersion in said circulating solution, an internal conductor, and a salt bridge solution contained within the hollow insulating member for establishing electrolytic contact between said material and said internal conductor; and

an amplifier circuit responsive to the voltage on said internal conductor, said amplifier circuit having a field effect transistor attached to and in close proximity with the insulating member and having its gate electrode connected to said internal conductor and its source and drain terminals connected to provide a low impedance signal indicative of the pH of the circulating solution for transmission to a remote location.

Whether the district court erred in: (1) holding the '525 patent valid;2 (2) finding claims 1-3, 8, and 12 infringed; (3) finding willful infringement and awarding treble damages and attorney fees; (4) holding Beckman in contempt; (5) excluding testimony; (6) declining to award damages for infringement by the Model 960B pH meter.

Beckman misconceives the role of an appellate court and the nature of the appellate process. The court does not sit to consider evidence de novo, and the appellate process is not a mere academic exercise. Yet Beckman asks us to say the invention "is"3 obvious after considering anew and only the claims and the prior art. Its briefs ignore all facts in the record that objectively establish the circumstances and human events surrounding the birth of the invention and its effect on the industry, as well as many of the fact findings militating against its positions on the issues. The district court's Memorandum Decision is nowhere mentioned in the 78 pages of Beckman's briefs.

Appeals in patent cases should not be mere games played with pieces of paper called references and the patent in suit. Lawsuits arise out of the affairs of people, real people facing real problems. So here, the technical problem out of which grew the present litigation was real. It was solved by inventor Cardeiro.

When Cardeiro's company began selling his problem-solving product, Beckman was confronted with a serious competitive business problem. Beckman first tried to solve that business problem with its own new designs, and when those failed, by appropriating Cardeiro's invention.

None of this real-world history is explicated in Beckman's briefs. An appellant cannot, within the page limitations for briefs, be expected to grapple with each of 137 fact findings. The foregoing history, however, is made clear in these ignored-by-Beckman findings of the district court:4

(1) Before Cardeiro's invention, repackaged laboratory pH meters were used in liquid streams of the process industry.

(2) The pH meters of (1) had a serious problem of drifting out of calibration. The leading pH meter in the field, Beckman's Model "J", had to be recalibrated as often as every eight hours.

(3) Beckman and others made numerous unsuccessful attempts to fill the longfelt need for a solution to the drifting problem.

(4) Scaling and corrosion had been a problem in watercooling towers since those towers began operating. No one had ever tried to control the cooling water remotely and automatically because it was known that available pH meters were not reliable for more than a few hours. Constant checking and recalibration of pH meters by a staff of technicians were required and accepted burdens.

(5) Three employees of Pacific Telephone Company tried to solve the problem in (4) with a system incorporating a Honeywell pH meter. When that failed, the three approached Cardeiro, who joined them in the recently formed Uniloc.

(6) Proceeding contrary to the prior art teachings and ignoring expressions by others that would discourage a search for new inventions, Cardeiro invented the pH meter of the '525 patent.

(7) The pH meter of the '525 patent solved the drifting problem. It could be used for a year without recalibration or "standardizing".

(8) The pH meter of the '525 patent met with immediate and rapidly growing commercial success. It replaced earlier pH meters and was used where others would not work.

(9) Virtually the entire industry converted to the pH meter of the '525 patent. The only major manufacturer of process pH meters not licensed under the '525 patent is Beckman.

(10) When Cardeiro made his invention in 1965, Beckman was making its Model "J". Though it had earlier made a space pH meter for NASA (discussed infra ) and others, it modified its Model "J" into the Model 900, which it advertised as the finest in the industry.

(11) Beckman at this point was doing all it could to design a pH meter that would preserve its lead in the marketplace. It failed in that effort.

(12) When Beckman's employees continued to make reports of sales lost to Uniloc, Beckman purchased a pH meter of the '525 patent and examined and tested it in 1967. Beckman's engineering department said the pH meter of the '525 patent "created a real crisis" at Beckman.

(13) Beckman established a project in 1968 to develop a new pH meter, because, as stated in an internal memorandum, "[T]he Beckman units are vulnerable to the Uniloc system...." The project planned to incorporate Cardeiro's concept of locating the input amplifier at the glass electrode.

(14) In 1969, after the '525 patent issued, Beckman modified its project pH meter, resulting in the Models 940 and 941 not accused of infringement.

(15) Models 940 and 941 continued to lose sales to the pH meters of the '525 patent.

(16) In 1973, because of the growing competitive advantage achieved by the pH meters of the '525 patent, Beckman decided to produce and sell the infringing pH meters.

Abandoning some of its many trial defenses, Beckman has elected to challenge the judgment here by presenting its view of portions of the evidence relating to issues (1)-(3) listed above, and citations of language in court opinions which Beckman relates to its presentation. Beckman's entire presentation has been fully considered and found without merit, as discussed infra.

(1) Validity

On appeal, Beckman has somewhat reduced its attack on the '525 patent, asserting invalidity under Secs. 102(a), 102(b), 102(g), 103, and 112.

(a) Secs. 102(a), 102(b), 102(g)

Beckman bases this three-sided defense on pH meters it produced and sold in 1963-64 to NASA and North American Aviation, Inc. (NAA) for use in space. Those pH meters were never used in space, and, as found by the district court, were never tested in zero gravity. Beckman insists that those pH meters anticipated the Cardeiro invention.

Though the parties have at length argued many aspects of these defenses, it is unnecessary to discuss the merits of those arguments here. The district court found that Beckman's "space pH meters" did not include all of the elements of the claimed invention.

Rosemount argues that anticipation under Sec. 102 requires identity, citing Kalman v. Kimberly-Clark Corp., 713 F.2d 760, 218 USPQ 781 (Fed.Cir.1983); SSIH Equipment, S.A. v. Int'l. Trade Comm., 718 F.2d 365, 218 USPQ 678 (Fed.Cir.1983); W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 220 USPQ 303 (Fed.Cir.1983); and Soundscriber Corp. v. United States, 360 F.2d 954, 175 Ct.Cl. 644, 149 USPQ 640 (1966).

Beckman argues that the differences are not patentable distinctions, citing Connecticut Valley Enterprises, Inc. v. United States, 348 F.2d 949, 172 Ct.Cl. 468 (1965); Tri-Wall Containers, Inc. v. United States, 408 F.2d 748, 187 Ct.Cl. 326, 161 USPQ 116, cert. denied, 396 U.S. 828, 90 S.Ct. 78, 24 L.Ed.2d 79 (1969); and In re Smith, 714 F.2d 1127, 218 USPQ 976 (Fed.Cir.1983). Though a non-identical device has been viewed as anticipating when the differences are not patentable distinctions, the result is the same under the more compartmentalized view that differences require application of Sec. 103. Beckman bases its argument, as it bases its argument devoted to Sec. 103 infra, on the assertion that the elements claimed in the '525 patent and not found in its space pH meters, as well as elements found in its space pH meters and not in the claims, are individually found somewhere in the prior art.

Beyond the effectively uncontradicted evidence that the space pH meters lacked the "DC amplifier circuit ... including a preamplifier stage having a field effect transistor ... with its gate terminal connected to said internal conductor" set forth in the claims, and beyond the uncontradicted evidence that Beckman employed a chopper in its space pH meters to convert the signal, lies the evidence most telling. It is uncontroverted that Beckman, with a laboratory staff of those skilled in the art continuously seeking an improved pH meter, did not find it obvious in 1965 to so modify its space pH meters as to produce the pH meters claimed in the '525 patent. It is only after learning of Cardeiro's invention, appropriating it, and getting sued for infringement, that Beckman attempts to show how its space pH meters might have been so changed as to remove the differences and to achieve the results achieved by Cardeiro. That approach illustrates the importance of the differences and is incapable of supporting Beckman's burden.

Thus Beckman's defenses under Secs. 102(a), (b), and (g) must fail, whether because identity is considered lacking between its space pH meters and the claimed inventions or because the differences between its space pH meters and the claimed inventions are not so inconsequential as to establish that the latter would have been obvious as a whole in 1965.

As above indicated, Beckman's defense under Sec. 103 consists of an effort to establish that each element of the inventions claimed in the '525 patent can be found in one of four places: Cameron patent disclosing a laboratory pH meter; two Zimmerli articles; literature of a parts manufacturer; and the Fein article. The district court found, and we agree, that this prior art array was not more pertinent that that considered by the PTO and was merely cumulative with the considered art.

Beckman's effort to establish that each element may be found somewhere in the prior art is unavailing. As this court has held, a combination may be patentable whether it be composed of elements all new, partly new, or all old. Connell v. Sears, Roebuck & Co., 722 F.2d 1542, 220 USPQ 193, 199 (Fed.Cir.1983); Environmental Designs, Ltd. v. Union Oil Co., 713 F.2d 693, 218 USPQ 865 (Fed.Cir.1983).

Fatal to Beckman's cause is its inability at trial and here to cite anything in the prior art items, individually or together, that can be seen as suggesting Cardeiro's solution to the drifting problem. Indeed, the facts of real world experience establish the contrary. The art now extolled by Beckman was available to all skilled in the art, including Beckman's own development staff, yet the solution to the drifting problem found by Cardeiro was not obvious to any of them. The same facts refute Beckman's assertion that the advent of field effect transistors rendered the inventions claimed in the '525 patent obvious.

The objective evidence, again composed of real world facts, is worthy of great weight in this case. As found by the district court, that evidence included tremendous commercial success, the filling of a long felt need, failure of others, including Beckman, to solve the problem, copying by Beckman, Beckman's praise of the invention, Cardeiro's departure from expert-accepted principles, and widespread recognition of the invention's significance in the industry. Beckman's argument that a nexus between commercial success and Cardeiro's invention is lacking and its conjecture that some of the commercial success here proven may have been due to elements in non-asserted claims are inadequate to overcome the objective evidence of record.

In sum, Beckman's attack on the validity of the patent under Sec. 103 wholly failed to meet the burden imposed on Beckman under 35 U.S.C. Sec. 282.

Beckman attacks the claims as indefinite, primarily because "close proximity" is not specifically or precisely defined. As stated in the district court's Memorandum Decision, "to accept Beckman's contention would turn the construction of a patent into a mere semantic quibble that serves no useful purpose." 569 F.Supp. at 940.

Indeed, the devotion of large portions of Beckman's briefs to this defense is difficult to appreciate in light of the record. The district court accepted testimony that one skilled in the art would understand all language in the claims when read in light of the specification, as the claims must be. Caterpillar Tractor Co. v. Berco S.P.A., 714 F.2d 1110, 219 USPQ 185 (Fed.Cir.1983). Beckman is confronted also with its own ease in applying "close proximity" to the prior art at trial and in its briefs here, the use of "close proximity" in the claims of one of its references, its own use of "close proximity" in describing its pH meters, its own witness' statement that he had no trouble understanding the claims of the '525 patent, the specific acceptance of "close proximity" by the examiner and by the industry, and the district court's finding that "close proximity" is as precise as the subject matter permits.

Beckman's argument that the district court erred in not upholding its Sec. 112 defense is without merit.

(2) Infringement

The district court found that the pH meters accused at trial were literal infringements of the asserted claims, and included with its Memorandum Decision, 569 F.Supp. at 940, n. 12, a direct comparison, reading the claims element for element on the accused pH meters.

Beckman cites testimony of its witnesses that the accused pH meters do not "use" the temperature compensation advantage of the patented pH meters, and that there is therefore no identity of function or operation. Rosemount, which bore the burden of proof at trial on infringement, cites contrary testimony of its witness, the testimony of a Beckman witness, the literature of the company manufacturing Beckman's input amplifier, the similarity of the accused pH meters to a figure and description in the '525 patent, and the mounting of Beckman's field effect transistor in the dome of the electrode station within one inch of the pH electrode, on the premise that temperature variations would equally affect elements arranged in such "close proximity."

As above stated, this court does not sit to reweigh the evidence or to decide which among opposing witnesses is to be believed. On appeal, the burden of establishing clear error in the findings on this fact issue rests on Beckman. In its briefs Beckman attacks only the finding that the close proximity of its field effect transistor to the glass membrane produces between them a heat exchange relationship (stated only in claim 2). Though Beckman says there is no evidence to support that finding, the presumption that elements an inch apart will be similarly affected by temperature, and the presence of other findings on the operation of the accused pH meters render futile the attack on this one finding.

Beckman ignores the district court's finding in its Memorandum Decision, 569 F.Supp. at 941, that the accused pH meter "does not perform its function in a substantially different way," and the finding in that Decision that its field effect transistor is in "close proximity to the high impedance material of the electrode." In attacking the district court's findings that the accused pH meters do have the temperature compensation feature Beckman only says it is not "using," and that Beckman does use that feature, Beckman improperly relies on material not in the trial record (proffered but excluded testimony and post-trial submissions--both discussed infra ).

In sum, Beckman has wholly failed to establish that the district court's finding of infringement was clearly erroneous. Fed.R.Civ.P. 52(a).

(3) Willful Infringement-Damages-Attorney Fees

Finding that Beckman knowingly, deliberately, willfully and wantonly infringed because of market pressure and without investigating the validity or scope of the patent, the district court awarded treble damages and attorney fees under 35 U.S.C. Secs. 284 and 285. That award is within the discretion of the district court and will not be overturned absent a clear showing of abuse of discretion. Hughes v. Novi American, Inc., 724 F.2d 122 (Fed.Cir.1984); Raytheon Co. v. Roper Corp., 724 F.2d 951 (Fed.Cir.1983); Underwater Devices Inc. v. Morrison-Knudsen Co., 717 F.2d 1380, 219 USPQ 569 (Fed.Cir.1983); White Consolidated Ind. v. Vega Servo-Control, Inc., 713 F.2d 788 (Fed.Cir.1983).

This court has held that a duty exists to obtain competent legal advice before initiating possibly infringing action. Underwater Devices Inc., supra. This court has also held that willfulness may include a determination that the infringer had no reasonable basis for believing it had a right to do the acts. Stickle v. Heublein, Inc., 716 F.2d 1550, 219 USPQ 377 (Fed.Cir.1983). There was no evidence at trial that Beckman had sought and obtained competent legal advice, and the district court's findings establish that it made the "no reasonable basis" determination required by Stickle.

In an effort to show abuse of discretion, Beckman cites in-house memos of the "not invented here" variety from its engineers and executives. In those memos, the authors say they see nothing patentable in Cardeiro's pH meters and that it employed old elements each of which was known to them. Rosemount points to other Beckman in-house memos lauding the Cardeiro invention, which in the form of Uniloc's product produced a crisis level threat to Beckman's leading market share. Nothing in the trial record establishes the presence of the "honest doubt" of validity and infringement referred to in Wilden Pump & Engineering Co. v. Pressed & Welded Products Co., 655 F.2d 984, 213 USPQ 282 (9th Cir.1981).

The evidence at trial refutes Beckman's assertion that the accused pH meters were occasioned by the availability of integrated circuits and had nothing to do with any intent to appropriate the invention of the '525 patent.

Lastly, Beckman improperly attacks the award as in conflict with "evidence" that was not presented at trial. That "evidence" consists of two ex parte affidavits of in-house counsel with attached opinions and a deposition, all of which were submitted in connection with Beckman's post-trial motion for a new trial.5 The district court thoroughly reviewed that material and found it inadequate, effectively making for the second time its determination that Beckman had no reasonable basis for believing it had a right to do its acts of infringement.

Nothing presented here is effective to demonstrate clear error in the finding of willfulness, or in the findings on which the district court based its award, or would justify this court in determining that an abuse of discretion occurred in the award of treble damages and attorney fees in the circumstances of this case.

(4) Contempt

Beckman's briefs do not challenge the district court's judgment of contempt based on its making and selling the Model 960B pH meter. Assuming reliance on its noninfringement assertions concerning Models 940A/941A, the foregoing affirmance of the district court's findings on those models would require affirmance of the contempt judgment, which rested on the finding that the Model 960B is an equivalent.

In support of its defense under Sec. 112, Beckman points out that its Model 960B has the field effect transistor within seven to ten inches of the glass electrode. Beckman says Rosemount's assertion of infringement by Models 940A/941A, where the distance is one inch, and by the Model 960B, where it is seven to ten inches, establishes indefiniteness in the term "close proximity." The district court, however, found Models 940A/941A and 960B equivalent and that all infringe.

No basis appears on which the district court's finding that Model 960B is an infringement, or its consequent judgment enjoining further infringement and holding Beckman in contempt may be overturned.6

(5) Exclusion of Testimony

The district court excluded expert testimony concerning ex parte tests prepared for this litigation by Beckman witness Sowa. The testimony, as described in Beckman's proffer, would describe tests of the accused pH meters and would, says Beckman, establish that it did not use temperature compensation.

During discovery, Rosemount served an interrogatory requesting identity of experts Beckman would call at trial, substance of their testimony, etc. Beckman responded that it had not determined who it would call. On Rosemount's motion, the court ordered Beckman to respond and to permit the deposition of its experts. Beckman then identified Dr. Anson, whose deposition was taken. At that deposition, Dr. Anson could not answer Rosemount's questions about the ex parte tests, the results of which had been supplied to Rosemount. When Rosemount asked for and could not obtain identification of any expert that might testify about the tests at trial it stated an intent to object to any such testimony that might be offered.

When Beckman offered the testimony on the tests, the district court, after hearing argument, sustained the objection and excluded the testimony. The testimony was offered at an advanced stage in the trial. Beckman's suggestion that a continuance be granted to permit Rosemount to take the witness' deposition was declined. The witness had already testified on the construction of the accused pH meters.

Beckman now argues that an abuse of discretion occurred in excluding the testimony, citing the Advisory Committee's Notes to Rule 403. The evidence was excluded, however, not under Rule 403 but under Rules 26(e) and 37(b) Fed.R.Civ.P.

Rule 26(e) requires expert identification before trial, as did the discovery order here. The Advisory Committee Notes to Rule 26(e) include:

"The duty will normally be enforced ... through sanctions imposed by the trial court, including exclusion of evidence, continuance, or other action, as the court may deem appropriate."

Sowa was not identified in response to the court's order, thus implicating Rule 37(b)(2)(B), which authorizes the court in such circumstances to prohibit Beckman "from introducing designated matters into evidence. Beckman says it considered Sowa's testimony on his tests as occurrence not expert testimony. It made the same argument to the district court. Though the court said it did not attribute bad faith to Beckman's counsel, it did not accept counsel's labeling of the testimony. Nor do we. One under a court order to identify its expert witnesses, and specifically requested by opposing counsel to identify who would testify on tests of the accused products, is ill advised to rely on his own characterization of the testimony as a basis for escaping the effect of Rules 26(e) and 37(b)(2)(B).

Rosemount cites cases in which the district court's exclusion of testimony was held on appeal not to have been an abuse of discretion. Beckman cites cases in which an exclusion was held an abuse and offers fact distinctions between the present case and those cited by Rosemount. It also says Rosemount was not prejudiced because it knew about the tests and had a rebuttal witness ready, and because, as above indicated, Beckman had suggested a continuance or recess of the trial to permit Rosemount to take Sowa's deposition.

The conduct of a trial, granting of continuances and the like, is not, however, solely or entirely a matter of balancing conveniences of the parties. The Federal Rules of Civil Procedure recognize another consideration--the need for the exercise of discretion by the trial court in carrying out its duty of managing the judicial process, the business of the court, and the administration of justice. The proper exercise of that discretion should not be impeded by second guessing at the appellate level, except in those rare instances when a clear abuse of discretion is firmly shown. The principle was well stated in United States v. Sumitomo Marine & Fire Ins. Co., 617 F.2d 1365, 1369 (9th Cir.1980) (discretion will not be disturbed unless appellate court has firm conviction trial court committed clear error of judgment after weighing relevant factors). The district court here weighed the relevant factors and selected from among the sanctions authorized by the Rules. We cannot, in light of all the circumstances, hold that the exclusion of Sowa's testimony on his tests to have been an abuse of discretion.

(6) Denial of Damages

Beckman informed Rosemount of the Model 960B shortly after the complaint was filed and long before trial. Rosemount's charge that the 960B infringed was made three years after it was first informed. The court, though it enjoined further infringement by the 960B, denied damages to Rosemount for the infringement that occurred during its long silence, finding it reasonable for Beckman to conclude that Rosemount was not going to charge the 960B as an infringement. We agree.

Rosemount mistakenly asserts that the court was establishing a rule that all post-complaint infringing products must be accused or the patent holder will be subject to laches. An equitable doctrine, laches and its applicability must be determined in light of the facts of each case. Leinoff v. Louis Milona & Sons, Inc., 726 F.2d 734 (Fed.Cir.1984); Advanced Hydraulics, Inc. v. Otis Elevator Co., 525 F.2d 477, 186 USPQ 1 (7th Cir.), cert. denied, 423 U.S. 869, 96 S.Ct. 132, 46 L.Ed.2d 99 (1975).

In the present case, nothing impeded Rosemount from filing a supplemental complaint or otherwise notifying Beckman of its intent to accuse the Model 960B. Rosemount correctly cites the general rule that only those acts committed before the complaint is filed may be considered at trial, but cites no reason why it did not file a supplemental complaint as provided for in Rule 15(d), Fed.R.Civ.P. As above indicated, the facts are controlling. Though it is true that Beckman had begun selling the 960B some two months before notifying Rosemount, that fact does not militate against the district court's recognition that Beckman was prejudiced in continuing the sale of that model for almost three years with Rosemount's acquiescence as evidenced by its silence. It is not necessary, as Beckman does, to attribute to Rosemount an ulterior trial strategy. Rosemount's long silence, unexplained before the district court and this court, can hardly rebut Beckman's showing of unreasonableness of the delay. See Leinoff, supra, 726 F.2d at 738. That delay raised equitable considerations respecting Rosemount's right to compensation for infringement by the Model 960B. The district court committed no error in denying Rosemount damages for that infringement.

The judgments appealed from are in all respects affirmed.

AFFIRMED.