This appeal and cross-appeal are from the April 22, 1991 judgment and the April 23, 1991 amended judgment of the United States District Court for the Northern District of California, In re Hayes Microcomputer Products, Inc. Patent Litigation, 766 F. Supp. 818, 20 USPQ2d 1836 (N.D.Cal.1991). On January 25, 1991, a jury returned a verdict in favor of Hayes Microcomputer Products, Inc., the assignee of U.S. Patent 4,549,302 (the '302 patent), against Ven-Tel, Inc., Omnitel, Inc., and Everex Systems, Inc. The jury found that the patent was not invalid and was willfully infringed by each defendant, and it awarded Hayes damages. The district court subsequently denied the defendants' motion for judgment notwithstanding the verdict (JNOV). Ven-Tel appeals on the issues of validity, infringement, and willful infringement. Hayes cross-appeals from the district court's denial of enhancement of royalties payable during pendency of this appeal and denial of sanctions under Fed.R.Civ.P. 11.1 We affirm the judgment in all respects.BACKGROUND

The '302 patent, entitled "Modem with Improved Escape Sequence Mechanism to Prevent Escape in Response to Random Occurrence of Escape Character in Transmitted Data," was issued in the name of Dale Heatherington on October 22, 1985. The application was a division of a division of a parent application filed on June 19, 1981, from which it claims priority. The invention relates to a mechanism for controlling the mode of operation of a modem. A modem is used for modulating and demodulating signals, both analog and digital, over telephone lines.2 It has two modes: (1) a transparent mode, in which the modem performs the modulation-demodulation function3, and (2) a command mode, in which the modem responds to predetermined commands and performs operations by executing a set of instructions stored in Read-Only-Memory (ROM) or firmware. A predetermined command or escape command tells the modem when to switch between transparent and command modes.

The patent claims an improved mechanism for detecting an escape command by a modem. It contains five claims, claims 1, 2, 4, and 5 being in independent form. Claim 1 is representative of the claims in suit, and reads as follows:

1. In a modem including a data input port for connecting said modem to a utilization device, and a telephone port for connecting said modem to a telephone line, said modem being of the type having two distinct modes of operation:

(a) a transparent mode of operation for which said modem provides modulated signals to said telephone port in response to data signals provided to said data input port; and

(b) a command mode of operation for which said modem responds to said data signals provided to said data input port as instructions to said modem;

said modem including means defining a predetermined sequence of said data signals as an escape character; the improvement comprising:

timing means for detecting each occurrence of a passage of a predetermined period of time after provision of one of said data signals to said data input port; and

means, operative when said modem is in said transparent mode of operation, for detecting provision of said predetermined sequence of said data signals, and for causing said modem to switch to said command mode of operation, if and only if said predetermined sequence of data signals occurs contiguous in time with at least one said occurrence of said passage of said predetermined period of time during which none of said data signals are provided to said data input port.

(Emphasis added). Each claim includes "timing means" and "means, operative."

The specification discloses that a microprocessor used in the invention is preferably a Z-8 type as manufactured by Zilog Inc., but states that "other microprocessors or other collections of memory, registers, counter timers and an arithmetic logic unit providing equivalent functions could be used." ' 302 patent, col. 4, lines 20-25. The preferred processor includes a central processing unit (CPU) and a program ROM for storing instructions for the functions to be executed by the processor. Id. at col. 5, lines 14-17. The ROM is programmed to include instructions which are used to control the mode in which the modem operates. Id. at col. 5, lines 58-61. In the transparent mode, all data coming into the computer's data port are treated as data to be modulated and then transmitted over telephone lines. The preferred embodiment will switch from the transparent mode to the command mode when it receives a predetermined sequence of data bits, the escape command. The specification recites the escape sequence as one full second of no data, followed by the predetermined escape command, followed by another full second of no data. Id. at col. 6, lines 24-29. The periods of no data have been commercially referred to as "guardtime."

Ven-Tel manufactures and markets a modem for use with personal computers that detects the escape code signal "-+++-", wherein "-" represents one second of guardtime and "+++" is the industry standard escape signal. Ven-Tel first marketed its modems in 1983, designed to be compatible with commercially available software by using the standard escape code signal. In October 1986, Hayes sent approximately 125 manufacturers, including Ven-Tel, notices of potential infringement of the '302 patent and offered licensing arrangements. In May 1987, at the direction of the Modem Patent Defense Group (MPDG), a consortium of such manufacturers, Charles Call, a patent attorney, prepared a written validity opinion on the patent and opined that it was invalid. This letter became known as the "green light letter." Potential members of MPDG were asked to contribute $10,000 each to fund litigation against Hayes' patent. Ven-Tel paid its membership fee and, in return, received a copy of the green light letter.

On January 7, 1988, Ven-Tel filed suit under the Declaratory Judgment Act, 28 U.S.C. § 2201 (1988), for a declaration of invalidity and non-infringement. Hayes counterclaimed that Ven-Tel was infringing the '302 patent and violating the antitrust laws. On January 25, 1991, a jury returned a general verdict that the '302 patent was not invalid and was willfully infringed. The jury awarded as past damages a reasonable royalty of 1.75% of net sales of Ven-Tel's infringing products, totalling $1,010,116.50. The court entered judgment accordingly and reserved judgment on the issues of permanent injunction, prejudgment interest, enhancement of damages, and attorney fees. Ven-Tel filed motions under Fed.R.Civ.P. 50(b) for JNOV4 and for a stay of injunction pending appeal. Hayes moved for sanctions, enhanced damages, attorney fees, and an accounting of infringing sales.

The district court denied Ven-Tel's motion for JNOV on all grounds, granted Hayes' motion for an injunction, and granted Ven-Tel's motion for a stay of the injunction pending appeal. The court also granted Hayes' motion for prejudgment interest, awarded enhanced damages of twice the amount found and assessed by the jury, and awarded attorney fees. The court agreed to Hayes' requested accounting of infringing sales subsequent to December 31, 1990, and denied Hayes' motion for Rule 11 sanctions.

I. Standard of Review from Denial of JNOV on Issues of Infringement and Validity

This case is highly fact-specific and evidence-oriented. It is an excellent illustration of the difficulty appellants face in attempting to persuade a court to reverse a jury verdict involving questions of fact.

On appeal of a judgment entered on a verdict after denial of a motion for JNOV, Ven-Tel

must show that the jury's findings, presumed or express, are not supported by substantial evidence or, if they were, that the legal conclusion(s) implied from the jury's verdict cannot in law be supported by those findings.

Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 893, 221 U.S.P.Q. (BNA) 669, 673 (Fed.Cir.) (citation omitted), cert. denied, 469 U.S. 857, 105 S. Ct. 187, 83 L. Ed. 2d 120 (1984). Fact findings reviewed under the substantial evidence standard require affirmance unless appellant shows that no reasonable juror could have reached such a result.5 Id.; see Railroad Dynamics, Inc. v. A. Stucki Co., 727 F.2d 1506, 1512-13, 220 U.S.P.Q. (BNA) 929, 936 (Fed.Cir.), cert. denied, 469 U.S. 871, 105 S. Ct. 220, 83 L. Ed. 2d 150 (1984).

Ven-Tel argues that the '302 patent is invalid for failure to meet the written description requirement of 35 U.S.C. § 112. Specifically, Ven-Tel argues that the "timing means" and the "means, operative" were not described in the specification, and that Hayes maintained the timing means as a trade secret. According to Ven-Tel, Hayes attempted to benefit from the filing date of the 1981 parent application by claiming the escape mechanism without a written description of it in the divisional application filed in 1983, while simultaneously maintaining the escape mechanism as a trade secret. Hayes marketed the Hayes SmartModem in 1981, in which it implemented a software timer as the "timing means," the structure of which Ven-Tel asserts was held as a trade secret.

The first paragraph of § 112 requires that

[t]he specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art ...

35 U.S.C. § 112 (1988) (emphasis added). The standard for determining whether the written description requirement has been met has been stated as follows:

Although [the applicant] does not have to describe exactly the subject matter claimed, ... the description must clearly allow persons of ordinary skill in the art to recognize that [he or she] invented what is claimed.... The test for sufficiency of support in a parent application is whether the disclosure of the application relied upon "reasonably conveys to the artisan that the inventor had possession at that time of the later claimed subject matter."

Vas-Cath Inc. v. Mahurkar, 935 F.2d 1555, 1563, 19 USPQ2d 1111, 1116 (Fed.Cir.1991) (citations omitted). Whether the written description requirement has been met is a question of fact. Ralston Purina Co. v. Far-Mar-Co, Inc., 772 F.2d 1570, 1575, 227 U.S.P.Q. (BNA) 177, 179 (Fed.Cir.1985). Thus, we must determine whether substantial evidence supports the verdict that the disclosure adequately describes the invention of the challenged claims.

Ven-Tel first argues that "timing means" refers to a software timer, the structure of which is not disclosed. Hayes responds that the specification adequately discloses timing means and refers to several statements in the specification. Hayes points out that the "Summary of the Invention" section recites that "[t]he present invention provides a full duplex intelligent modem wherein the decision making capability preferably resides in a microprocessor ..." '302 patent, col. 2, lines 27-29 (emphasis added). Hayes also notes that the specification discloses the preferred type of microprocessor as a Z-8 type. The specification also states:

The details of the internal structure of the Z-8 processor used in the preferred embodiment will be known to those skilled in the art but are described in a publication[ ] entitled "Z-8 Microcomputer, Preliminary Technical Manual," publication No. 03-3047-02, published by Zilog Inc.

'302 patent, col. 5, lines 40-45.

We agree that the specification meets the requirements of section 112, first paragraph, recognizing that the specification is directed to one skilled in the art. See Chemcast Corp. v. Arco Indus. Corp., 913 F.2d 923, 926, 16 USPQ2d 1033, 1036 (Fed.Cir.1990) ("Of necessity, the disclosure required by section 112 is directed to those skilled in the art."). There is substantial evidence in the record to support the conclusion that one skilled in the art would understand that the timing means of the invention is incorporated into the structure of the microprocessor. For example, when shown a flowchart diagram for implementing the escape sequence, the inventor, Heatherington, testified that "if you had experience in doing microprocessor programming, you would know how to implement what's in that diagram." When further asked about being able to implement every detail of the invention from the patent disclosure, Heatherington responded that "you would get an escape sequence that required one second on the front end and one second on the back end and some number of characters in the middle, which is what I thought was important to disclose."

One skilled in the art would know how to program a microprocessor to perform the necessary steps described in the specification. Thus, an inventor is not required to describe every detail of his invention. An applicant's disclosure obligation varies according to the art to which the invention pertains. Disclosing a microprocessor capable of performing certain functions is sufficient to satisfy the requirement of section 112, first paragraph, when one skilled in the relevant art would understand what is intended and know how to carry it out.6

According to Ven-Tel, Hayes retained the firmware listing for implementing the timing means as a trade secret. Thus, Ven-Tel argues that Heatherington should have disclosed the firmware listing itself to meet the written description requirement. We disagree. The disclosure sufficiently recites the function of the firmware:

It should be understood that the program included in program ROM 112 of processor 55 includes instructions which are used to control the mode in which the modem of the preferred embodiment operates. In particular, two modes of operation of the preferred embodiment are defined: the command mode; and the transparent mode.

'302 patent, col. 5, lines 58-64. The evidence supports the conclusion that one of ordinary skill in the art would understand how to implement the timing means with a microprocessor without a firmware listing.

Ven-Tel further states that a microprocessor is defined by its programmable resources and, without programmed firmware, the microprocessor has no special functionality. The evidence of record supports the conclusion that all that was required for one of ordinary skill in the art to understand what the invention was and how to carry it out was the disclosure of a microprocessor having certain capabilities and the desired functions it was to perform. We disagree with Ven-Tel's contention that to satisfy section 112, a statement as to the specific function of a microprocessor is inadequate, that the actual program must be disclosed. While this may be true in some instances, this is not such a case.

Ven-Tel also focuses on the fact that the heart of the claimed invention of the '302 patent is described only in twenty-seven lines. Certainly no length requirement exists for a disclosure to adequately describe an invention. While some inventions require more disclosure, the adequacy of the description of an invention depends on its content in relation to the particular invention, not its length.

Ven-Tel also argues that the detecting "means, operative" portion of the claims lacks structure in the specification and fails to meet the requirements of 35 U.S.C. § 112, sixth paragraph, which states:

An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.

While Heatherington was required to disclose some structure in the specification for all "means" recitations in the claims, he was not required to disclose every means for implementing the stated function. See D.M.I., Inc. v. Deere & Co., 755 F.2d 1570, 1574, 225 U.S.P.Q. (BNA) 236, 238 (Fed.Cir.1985) ("Patentees are required to disclose in the specification some enabling means for accomplishing the function set forth in the 'means plus function' limitation. At the same time, there is and can be no requirement that applicants describe or predict every possible means of accomplishing that function."). The specification states that

[i]n the preferred embodiment, a predetermined sequence of data bits is the escape command accepted at port 50 to return the preferred embodiment to the command mode. Since any predetermined sequence of bits which is defined as an escape command has a finite probability of occurring in any file of data being transmitted by the preferred embodiment, an additional requirement is placed on the signal sequence which is defined as the escape sequence. The preferred embodiment of the present invention must experience one full second of no data being provided as input to data port 50, followed by the predetermined escape command, followed by a second full second of no data in order to interpret that input as the escape sequence.

'302 patent, col. 6, lines 16-29. This is what the disclosed microprocessor is programmed to do.

Mr. Enlow, Hayes' expert witness, testified that the "means" recitations were adequately supported by the specification:

Q: Do you have an opinion as to whether the means language in the '302 patent is supported by the specification?

A: Yes, it is. The specification discloses the microprocessor which the application clearly says is where the escape mechanism would be taken care of.

Q: Do you have an opinion as to whether or not the timing means is disclosed in the--the structure is disclosed in the specification?

A: Yes. That would be in the microprocessor.

Q: Do you have an opinion as to whether or not the requirements of section 112 of the patent act to provide structure for the timing means, instruction for the operative means, had been met in the '302 patent?

A: Yes.... it was met and it's in the microprocessor.

Thus, there is sufficient evidence in the record so that the jury could have determined that the specification reasonably conveyed to one of ordinary skill in the art that Heatherington invented the subject matter of the '302 patent.7

Likewise, we disagree with Ven-Tel that the '302 patent failed to include adequate drawings under 35 U.S.C. § 113 (1988). According to section 113, "[t]he applicant shall furnish a drawing where necessary for the understanding of the subject matter sought to be patented." Sufficient evidence exists to support the conclusion that, to the extent it was necessary, the drawings were sufficient for a skilled artisan to understand the subject matter of the claimed invention. The microprocessor is identified as element 55 in Figure 1B of the specification. On the facts of this case, no more needed to be included in the drawings to satisfy the description requirement.

Thus, substantial evidence exists to support a finding that the '302 patent was not invalid for failure to provide an adequate written description.

Ven-Tel argues that no reasonable jury could have reached the conclusion that the best mode of practicing the invention of the '302 patent was disclosed; it argues that the district court's judgment should be reversed. Specifically, Ven-Tel asserts that Heatherington considered firmware to be the best way of implementing his invention, but failed to disclose the details of the firmware which he considered to be part of his best mode.8 Ven-Tel recites eleven features of the Hayes SmartModem, a commercial embodiment of the claimed invention, which Ven-Tel considers to have been part of the best mode of implementing the firmware of the SmartModem, only one of which was disclosed. According to Ven-Tel, Heatherington kept several of these features as trade secrets and intentionally concealed various others. Those allegedly kept as trade secrets were all part of the specific software timer used in the SmartModem, which Ven-Tel considers to be part of the "timing means" of the claimed invention. Ven-Tel states that Heatherington provided only a bare functional description of the timing means and effectively concealed his best mode.

Determination whether the best mode requirement has been met is a question of fact. DeGeorge v. Bernier, 768 F.2d 1318, 1324, 226 U.S.P.Q. (BNA) 758, 763 (Fed.Cir.1985). As with other factual aspects of the jury verdict, we review the record for substantial evidence to determine if no reasonable juror could have found that the best mode requirement had been satisfied.

The specification ... shall set forth the best mode contemplated by the inventor of carrying out his invention.

A best mode analysis has two components. Chemcast Corp., 913 F.2d at 927, 16 USPQ2d at 1036. The first inquiry focuses on whether the inventor knew of a mode of practicing his invention at the time he filed his patent application which he considered to be better than any other. This determination is subjective, focusing on the inventor's state of mind at the time he filed his application. If he did have a best mode, the next question is whether he disclosed it and did so adequately to enable one of ordinary skill in the art to practice the best mode. This is an objective determination. There must be no concealment of a mode known by the inventor to be better than that which is disclosed. Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384-85, 231 U.S.P.Q. (BNA) 81, 94 (Fed.Cir.1986), cert. denied, 480 U.S. 947, 107 S. Ct. 1606, 94 L. Ed. 2d 792 (1987).

Ven-Tel asserts that Heatherington repeatedly admitted that he had a best mode of implementing his invention. Specifically, according to Ven-Tel, he considered firmware to be the best mode of implementation and this mode was concealed. Accordingly, Ven-Tel argues that the jury's verdict on best mode is wholly unsupported by the evidence. Hayes responds that substantial evidence supports the jury's findings that Heatherington had a mode of practicing his invention which he considered to be better than any other and that this mode was adequately disclosed in the '302 patent. We agree.

"Firmware" is a generic term used to describe any computer program permanently stored in ROM associated with a microprocessor. A "firmware listing" is a specific written computer program. Substantial evidence was introduced at trial in support of the conclusion that Heatherington did not consider the specific firmware listing he used to implement his invention, either in its entirety or in its subparts, to be better than any other firmware listing that implemented his invention. Rather, he believed that the best mode of his invention was to store a firmware listing in firmware, and not that the various details of the specific firmware listing used in the Hayes SmartModem were the best mode. Thus, he was not required to disclose the details of the Hayes SmartModem firmware listing for a person of ordinary skill in the art to practice the invention.

The Court of Customs and Patent Appeals addressed a similar issue in In re Sherwood, 613 F.2d 809, 204 U.S.P.Q. (BNA) 537 (CCPA 1980), cert. denied, 450 U.S. 994, 101 S. Ct. 1694, 68 L. Ed. 2d 193 (1981), where the inventor did not disclose the specific computer program used to implement a system for converting a seismic time section. In discussing whether the best mode requirement was met, our predecessor court stated:

... Whether failure to disclose a listing of that known computer program is fatal is the question.

In general, writing a computer program may be a task requiring the most sublime of the inventive faculty or it may require only the droning use of a clerical skill. The difference between the two extremes lies in the creation of mathematical methodology to bridge the gap between the information one starts with (the "input") and the information that is desired (the "output")....

The Cardwell affidavit outlines in a very reasonable manner the human-to-human elements necessary in communicating the concepts of a particular computer program to one having ordinary ... skill in this art. The specification provides the general mathematical equations used and teaches the further "trick" of chopping the physical input seismic traces into segments via a mathematical manipulation. For this reason, the specification in our view delineates the best mode in a manner sufficient to require only the application of routine skill to produce a workable digital computer program.

Id., 613 F.2d at 817, 204 U.S.P.Q. (BNA) at 544 (footnotes omitted). In Sherwood, as Ven-Tel notes, the specification taught general mathematical equations for implementing the invention, unlike the Heatherington invention. While we agree that the '302 patent only discloses the general function of the firmware without teaching mathematical formulas, flow charts, or a firmware program listing, no more was needed here.

Heatherington believed that the best mode of implementing the escape sequence with a guardtime mechanism was with a microprocessor which contained firmware that included instructions to execute the escape sequence described in the patent. That was disclosed. The evidence supports the conclusion that a person of ordinary skill in the microprocessor art could develop such a firmware listing. This is apparent from the following testimony:

Q: Now, Mr. Heatherington, in your patent ... you have a sentence in there which refers to: "experience one full second of no data being provided as input to data port 50," followed by: "the predetermined escape command," followed by: "a second full second of no data." Do you recall that sentence--

A: Yes.

Q: --in your application--in your patent? Mr. Heatherington, do you believe that that sentence accurately describes the entire escape sequence as we have here, as implemented in the firmware, and as shown in the flow chart?A: I felt that it was sufficient information for a person that was skilled in microprocessor programming to duplicate it.

Moreover, Heatherington testified that "I believe that [the patent disclosure] was an adequate description of how the escape sequence works so that someone reading the patent would understand how that worked and how they would go about implementing it." Heatherington also stated that

I feel that what was disclosed was sufficient for engineers at that time that were familiar with microprocessors and modems and when they look at the patent, they read the patent, they could have, with that information alone, designed a SmartModem that has an escape sequence that would operate correctly.

Furthermore, Mr. Enlow testified that the fact that the particular implementation of the Hayes escape sequence was in firmware was a trade secret did not necessarily mean that the best mode of the invention was not disclosed.

Ven-Tel specifically recites allegedly undisclosed features pertaining to the best mode of the claimed invention. First, Ven-Tel claims that Heatherington failed to disclose the use of a software timer loaded with guardtime, that Heatherington actually concealed the use of a software timer as a trade secret, and that the software timer was an element of the claimed invention. This is not correct. There is ample evidence to support a finding that a software timer was not part of the best mode of practicing the invention. Both hardware and software timers are disclosed in the specification, and Heatherington testified that neither type is necessarily better than the other. Hardware timers are disclosed where the specification states that "[a] pair of counter timers designated as 117 are also included in the hardware of the Z-8 processor." ' 302 patent, col. 5, lines 21-23. The specification also discloses software timers and states that "the program included in program ROM 112 of processor 55 includes instructions which are used to control the mode in which the modem of the preferred embodiment operates." Id. at col. 5, lines 58-61. When asked whether a software timer is better than a hardware timer, Heatherington testified, "No, it's not." Thus, substantial evidence supports the conclusion that a software timer was not the best mode of carrying out the claimed invention.

According to Ven-Tel, Heatherington failed to disclose specific details of how the timer operates. It argues that the software timer counts down one second to zero from the last keyboard interrupt, at zero the timer idles at zero until the next keyboard interrupt, and that the timer resets each time a character interrupt is received. Heatherington testified, when asked if anything in the patent referred to the manner in which the timer counted down, that

the language in the patent describes how it works. It doesn't get into the level of detail that you got there, but it implies that it was necessary to have a timer that counts down to zero.

Heatherington testified that the patent specification was descriptive enough to enable one of ordinary skill in microprocessor programming to implement these features.

Ven-Tel also argues that Heatherington failed to disclose specific details of the escape command such as that the escape command is a single escape character repeated three times, that the escape character and the guardtime may be programmable, that the amount of time separating the escape characters is measured by the same timer as the guardtime, and that, if more than the guardtime elapses between escape characters, the escape sequence must restart. We agree with Hayes and find substantial evidence in the record to support a determination that Heatherington did not consider these features to be his best mode.

Specifically, although the use of a single character repeated three times was used in the Hayes SmartModem, Heatherington testified that picking three plus signs "was a fairly arbitrary decision" and that "it could have been almost anything." He further testified:

Q: From the standpoint of the function of the device, did it matter which characters you selected?A: No.

Q: You could have selected anything?

A: Yes.

* * * * * *

Q: Were there any advantages to using three pluses?

A: Not ... not really. It was kind of a--I chose that as a default. I had to pick something for a default. And there are a large number of things I could have used. It just looked nice to me. There wasn't any particular reason.

Thus, the evidence of record indicates that the particular selection of characters did not matter. Referring to the amount of time separating escape characters, Heatherington testified that the "time outs between pluses" did not need to be measured for the escape sequence to work. Ven-Tel also asserts that a single character repeated three times would use less memory and that Heatherington's invention had serious memory constraints. While there may be some evidence to support this contention, ample evidence exists that Heatherington did not consider this feature to be part of his best mode.

The Hayes SmartModem firmware allowed the escape character and the guardtime to be programmable and Ven-Tel contends that this was also part of the best mode of carrying out the claimed invention. We disagree. The record evidence indicates that whether the escape sequence was pre-programmed or not was not part of Heatherington's best mode.

Whether the best mode requirement is met depends on the scope of the invention, the skill in the art, evidence as to the inventor's belief, and other surrounding circumstances. Wahl Instruments Inc. v. Acvious Inc., 950 F.2d 1575, 1580, 21 USPQ2d 1123, 1127 (Fed.Cir.1991). Substantial evidence supports the jury's finding that the best mode requirement was not violated.

According to Ven-Tel, the claimed invention of the '302 patent would have been obvious at the time the invention was made and therefore the patent is invalid under 35 U.S.C. § 103 (1988).9 Ven-Tel states that the only novelty, if any, in the claimed invention is in the particular escape code signal, using guardtime contiguously with a defined escape character. Ven-Tel asserts that the patent does not claim the escape code signal with guardtime, but claims a physical mechanism for responding to it. Ven-Tel then states that Heatherington admitted that once the escape code signal was known, it would have been obvious to a person skilled in the art to build a mechanism for detecting the signal.

Obviousness is a legal determination that may be submitted to a jury with proper instruction. Railroad Dynamics, 727 F.2d at 1514-15, 220 USPQ at 937-38. A finding of nonobviousness depends on a number of factors, including objective evidence. Specialty Composites v. Cabot Corp., 845 F.2d 981, 989, 6 USPQ2d 1601, 1607 (Fed.Cir.1988) (citing Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S. Ct. 684, 693-694, 15 L. Ed. 2d 545, 148 U.S.P.Q. (BNA) 459, 467 (1966)).

Ven-Tel's primary assertion is based on an alleged admission during prosecution that modems with means to respond to escape signals were known in the prior art, and that the only novelty in the claimed invention rested in the escape signal using guardtime. Ven-Tel then suggests that Hayes conceded at trial that the patent claims a physical mechanism to respond to an escape code signal and does not claim the signal itself. Accordingly, Ven-Tel asserts that Heatherington's testimony during the trial that one of ordinary skill in the art could have implemented the invention when told of the escape sequence with guardtime is an admission of obviousness. We disagree.

Ven-Tel relies on Constant v. Advanced Micro-Devices, Inc., 848 F.2d 1560, 1569, 7 USPQ2d 1057, 1063 (Fed.Cir.), cert. denied, 488 U.S. 892, 109 S. Ct. 228, 102 L. Ed. 2d 218 (1988), for the proposition that an inventor's admissions during prosecution are probative and meaningful. While this is true, the fact that prior art modems existed does not support Ven-Tel's allegation of obviousness. Ven-Tel points to the following testimony during trial:

Q: Good morning, Mr. Heatherington. Is it your testimony, then, that if you just tell somebody skilled in the art that you wanted to have an escape sequence with a guardtime, it would be obvious for them how to build it?

A: I believe it would be.

This testimony was hardly an admission of obviousness in the context of Heatherington's overall testimony. Heatherington was addressing the adequacy of the disclosure; he was not admitting that the invention was obvious.

Furthermore, the record supports the jury's conclusion that the invention was nonobvious. Ven-Tel presented no evidence of the similarities between the prior art and the claims in issue. In fact, during an in-chambers conference on jury instructions, counsel for Everex, speaking on behalf of the defendants, admitted that they had not presented any evidence on obviousness. Specifically, he stated:

We intended to pursue obviously necessary [sic] and decided not to, and we have not pursued obviously necessary [sic], and we have not put on evidence on that issue.

* * * * * *

I don't understand why the plaintiff is pressing for obviously necessary [sic] instructions when it's our burden of proof, and I am telling you on behalf of the defendants that we have not put in an obviously necessary [sic] case.

However, on appeal, Ven-Tel relies on the preamble of the Jepson-type claim as an admission of prior art and asserts that the escape code was in the public domain at the time of the invention. According to Ven-Tel, the escape code "-+++-" was introduced to the modem industry in 1981, the use of escape code signals with modems was known in the prior art, and the Heatherington invention was therefore obvious.

We find no support for that contention. First, the parent application of the '302 patent was filed on June 15, 1981. The Hayes SmartModem, utilizing the escape sequence with guardtime, was first introduced to the public on June 16, 1981. Thus, the escape signal "-+++-" was not available to the public before the effective filing date of the '302 patent. Second, even if the escape signal had been in the prior art, that would not have made the claimed invention, which relates to a mechanism for detecting the escape signal, obvious. Ven-Tel points to no prior art that suggests modifying prior art modems to detect the escape signal with guardtime. Finally, although the use of escape code signals generally with modems was not new at the time of the Heatherington invention, that did not make the claimed invention obvious. In fact, the commercial success of the invention, the failure of others to solve the problem addressed by the patented invention, and the fact that the escape code signal "-+++-" has become the industry standard is compelling objective evidence of the nonobviousness of the claimed invention.

Thus, the record evidence does not support an admission of obviousness or that the escape signal was in the public domain. Ven-Tel failed to prove by clear and convincing evidence any facts on which to base a conclusion that the subject matter as a whole would have been obvious to a person having ordinary skill in the art at the time the invention was made.

Ven-Tel further argues that Hayes presented insufficient evidence for the jury to find infringement. Specifically, Ven-Tel states that Hayes' case for infringement was based solely on the testimony of Hayes' expert, Dr. Charles Cliett, who it says failed to properly construe the claims and had no knowledge of the structure of the Ven-Tel product. Moreover, Ven-Tel argues, Hayes failed to meet its burden of proof to establish, under 35 U.S.C. § 112, sixth paragraph, that the structures in the accused device which perform the recited functions are the same as or equivalent to the structures disclosed in the specification for performing those same functions.

Literal infringement may be found when an accused device falls within the asserted claims as properly construed. Palumbo v. Don-Joy Co., 762 F.2d 969, 974, 226 U.S.P.Q. (BNA) 5, 7 (Fed.Cir.1985) (citation omitted). The patentee has the burden of proving infringement by a preponderance of the evidence. SmithKline Diagnostics, Inc. v. Helena Lab. Corp., 859 F.2d 878, 889, 8 USPQ2d 1468, 1477 (Fed.Cir.1988). The determination of literal infringement is a question of fact, Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 1269-70, 229 U.S.P.Q. (BNA) 805, 811 (Fed.Cir.1986), cert. denied, 479 U.S. 1030, 107 S. Ct. 875, 93 L. Ed. 2d 829 (1987), as is the determination of equivalent structure under 35 U.S.C. § 112, paragraph 6, D.M.I., 755 F.2d at 1575, 225 USPQ at 239. "The first step in determining infringement is to construe the claims." Fonar Corp. v. Johnson & Johnson, 821 F.2d 627, 631, 3 USPQ2d 1109, 1112 (Fed.Cir.1987), cert. denied, 484 U.S. 1027, 108 S. Ct. 751, 98 L. Ed. 2d 764 (1988). The second step is to decide whether each limitation in the properly construed claims is found in the accused device. Id. Although claim interpretation is a question of law, the jury's ultimate finding on infringement is an issue of fact. See H.H. Robertson, Co. v. United Steel Deck, Inc., 820 F.2d 384, 389, 2 USPQ2d 1926, 1929 (Fed.Cir.1987). To find the scope of a means plus function element of a claim, we look to see whether the "means in the accused device which performs the function stated in the claim is the same as or an equivalent of the corresponding structure described in the patentee's specification as performing that function." D.M.I., 755 F.2d at 1575, 225 USPQ at 239.

Ven-Tel alleges that Dr. Cliett did not perform a proper infringement analysis; he never read section 112, he had no knowledge of the internal structure of the accused products, and he never compared the disclosed structure in the specification with the structure of the accused device. Ven-Tel specifically focuses on the lack of structure in the patent for the "timing means" and the "means, operative." We find no error; Hayes presented substantial evidence that Ven-Tel's products contained structure identical or equivalent to that in the patent.

As we previously discussed in the context of section 112, there is substantial evidence to support the conclusion that one of ordinary skill in the art would understand that the "timing means" and the "means, operative" are incorporated into the structure of the microprocessor. The record shows that Ven-Tel's modems were microprocessor-controlled, that the microprocessors in those modems performed timing tasks and changed modes of operation, and that the microprocessors in the Ven-Tel modems had been programmed to change modes of operation upon receipt of one second of guardtime, three pluses, and a subsequent second of guardtime. Specifically, Dr. Cliett testified that the Ven-Tel modem has both a transparent mode and a command mode of operation. He described how he manually tested each modem:

In that testing, I would simply take my watch and I would, I have a second hand that jumps, and I would sit there and I would wait a certain period of time and then I would type the escape character and then wait another period of time and look to see whether I got okay from the modem which typically ... means that it is ready to receive commands, that it is in the command mode....

So, I tested it, the Ven-Tel modem by, I would allow a second before a plus, a plus, a plus and a second after and then determine if I got the okay on my screen.

I did.

Dr. Cliett also testified how he tried varying permutations to see if the Ven-Tel modem responded like the Hayes SmartModem:

... [W]ith the Ven-Tel modem if I failed to give it any leading time before I typed the escape characters, then it failed to escape. If I didn't give it any trailing time, if I went plus, plus, plus and then typed my name immediately without waiting for a trailing period of time, then it also failed to escape. Only when I allowed it one second before one second after the three escape characters did it in fact escape.

The evidence shows that Dr. Cliett tested the modems manually, in an automated computer-assisted test, and by reviewing the instructions in the Ven-Tel product manual.

Additionally, Dr. Cliett testified that the Ven-Tel modem had a timing mechanism in order to detect the one second period of time both before and after the escape code. Although Dr. Cliett did not disassemble a Ven-Tel modem to inspect exactly how it worked, the attorney for Ven-Tel stipulated that there was a microprocessor inside the Ven-Tel modem. Dr. Cliett further testified that Ven-Tel's modems included a timing mechanism.

Although Dr. Cliett was not a patent law expert, he had read and understood the claims of the '302 patent at the time he tested the accused infringing devices. With respect to the Ven-Tel modem, he stated that each element of the claims was included in the accused device. Specifically, the modem had two distinct modes of operation, a transparent mode and a command mode. The modem included a means which defined a predetermined sequence of data signals as an escape character and a timing means for detecting each occurrence of a passage of a predetermined period of time. The Ven-Tel modem also included a means that operated when the modem was in the transparent mode of operation for detecting the predetermined sequence of data signals and for causing the modem to switch to the command mode of operation if and only if the predetermined sequence of data signals occurred contiguously in time with at least one occurrence of guardtime. Furthermore, Dr. Cliett testified that Ven-Tel's modems were "functionally equivalent" to the claimed invention, which in the context of this case supports the jury's conclusion that the means were structurally equivalent.

In addition, Gavin Grant, the engineer in charge of Ven-Tel's modem design, testified that he was instructed by Richard McVicker, Ven-Tel's president, to design a Hayes compatible modem including the "AT" SmartModem command set.10 The Ven-Tel modem was advertised as being "Hayes compatible." Mr. Grant also stated that he designed the Ven-Tel modem to respond to the Hayes' command sequence:

Q: In your design criteria for Hayes compatibility did you also want to make sure the modem responds to the escape sequence originated by Hayes in the same manner that the Hayes modem does? ...

A: Yes....

Q: As far as you know, do all of the Ven-Tel modems, which are "AT" command compatible, accomplish that compatibility insofar as they respond to the Hayes' developed escape sequence with guardtime?

A: To the best of my knowledge.

* * * * * *

Q: As far as you know, does all the firmware require both the leading and trailing guardtime to be present on the escape sequence before the modem will escape in response to an escape sequence? ...

A: Yes ...

From Mr. Grant's testimony, the jury could reasonably have inferred that the firmware, and hence the microprocessor, was programmed to control the escape sequence of the Ven-Tel device.

Ven-Tel stipulated that its modems included a microprocessor and Dr. Cliett testified that the modems were functionally equivalent to the claimed invention. Additionally, Ven-Tel did not object to the jury instructions regarding infringement and stated that "the jury was properly instructed."Generally, a number of factors may be considered when determining the scope of a means-plus-function limitation, "including the language of the claim, the patent specification, the prosecution history of the patent, other claims in the patent, and expert testimony." Palumbo, 762 F.2d at 975, 226 USPQ at 8 (citation omitted). This list is not exhaustive and reasonable inferences by the fact finder are appropriate. Although Dr. Cliett did not compare the specific structure of the "timing means" and the "means, operative" of the Ven-Tel device to the language of the claims, he was able to testify that the accused device included a microprocessor and was functionally equivalent to the claimed invention. He was not required to be a patent law expert in order to so testify. The primary issue here is whether a reasonable juror could have concluded from the evidence that the Ven-Tel devices contained a structural equivalent to the "timing means" and the "means, operative" disclosed in the '302 patent. Our review of the record enables us to conclude that a reasonable juror could have done so.

Ven-Tel argues that its infringement was not willful. The jury found otherwise. A finding of willfulness is a question of fact, Underwater Devices Inc. v. Morrison-Knudsen Co., 717 F.2d 1380, 1389, 219 U.S.P.Q. (BNA) 569, 576 (Fed.Cir.1983), which must be proved by clear and convincing evidence. Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 628, 225 U.S.P.Q. (BNA) 634, 644 (Fed.Cir.), cert. dismissed, 474 U.S. 976, 106 S. Ct. 340, 88 L. Ed. 2d 326 (1985). To determine whether a finding of willfulness is warranted, a jury looks to the totality of the circumstances to see whether "a reasonable person would prudently conduct himself with any confidence that a court might hold the patent invalid or not infringed." Ryco, Inc. v. Ag-Bag Corp., 857 F.2d 1418, 1428, 8 USPQ2d 1323, 1331 (Fed.Cir.1988). Whether the infringer intentionally copied the ideas of another; whether the infringer, once on notice of the patented invention, investigated the scope of the patent to form a good-faith belief that it was invalid or not infringed; and the infringer's behavior as a party to the litigation, are relevant factors to a willfulness determination to merit an increase in damages. Bott v. Four Star Corp., 807 F.2d 1567, 1572, 1 USPQ2d 1210, 1213 (Fed.Cir.1986). All of these factors may well have played a role in the jury's conclusion in this case.

Ven-Tel primarily relies on the opinion letter of Mr. Call as evidence that it was not a willful infringer. Although reliance on competent counsel's opinion is evidence of good faith, it is not conclusive. Central Soya Co. v. Geo. A. Hormel & Co., 723 F.2d 1573, 1577, 220 U.S.P.Q. (BNA) 490, 492 (Fed.Cir.1983) (citation omitted). In this case, substantial evidence exists to support the conclusion of willfulness, irrespective of the opinion letter. Mr. Enlow, Hayes' expert witness, testified in detail that the letter did not give Ven-Tel a reasonable basis for believing in good faith that the '302 patent was invalid. He stated that the letter treated the claims of the '302 patent superficially, failed to set out a standard for one of ordinary skill in the art, failed to consider secondary considerations in determining obviousness, mischaracterized prior art, and failed to perform a proper best mode analysis. Additionally, the letter only addressed invalidity and did not discuss infringement.

Mr. Call's opinion letter makes broad and conclusory statements with little, if any, support. The tenor of the letter describes a scenario in which "the lawyers 'invented a patent' ... rather than 'patenting an invention' " made by an engineer. The letter characterizes Hayes' actions as an attempt to monopolize the personal computer modem market and discusses concepts rather than analyzing the patent claims. There is no recitation of specific claims whatsoever. Mr. Enlow testified to that effect:

Q: In fact, in the green light letter does Mr. Call analyze the elements of the claim as you've outlined them in determining whether the prior art cites his obviousness analysis as relevant? Has he analyzed it in the way you describe?A: No, he hasn't. That's one of the great deficiencies that I see in the letter. He doesn't tell the client how this prior art might relate to these claims so that you can then go and make the conclusion that there's some reason why the invention is obvious. He has never done that.

* * * * * *

There is a great lack in his letter of ever relating some of the art that he recites. There's very little relating that back to the claims of the invention.

The evidence shows that the advice of counsel was more of a protective device than a genuine effort to determine before infringing whether the patent was invalid. See Underwater Devices, 717 F.2d at 1390, 219 USPQ at 576 (Once a potential infringer receives notice of a patentee's rights, that potential infringer has a duty of due care, including "the duty to seek and obtain competent legal advice from counsel before the initiation of any possible infringing activity." (citations omitted)).

In 1986, Hayes sent letters to numerous manufacturers, including Ven-Tel, informing them of the Heatherington patent and requesting that they take licenses. Rather than do so, Ven-Tel reacted by inviting other Hayes-compatible modem manufacturers to a meeting in November 1986. This meeting was organized "to discuss the possibility of forming [an] industry group to fund the litigation to invalidate the [Heatherington and Eaton] patents."¹¹ Two meetings, in fact, were held--one in San Jose and a later one in Las Vegas. At these meetings in November 1986, Mr. Call gave an oral opinion of invalidity of the Heatherington patent.12 However, Mr. Call testified that this opinion was "based on less than all this information." According to Mr. Johnson, who attended the Las Vegas meeting, Mr. Call did not provide any of the material that formed the basis for his invalidity opinion. Instead, Mr. Call "stated that he had found some prior art, but he did not give that information to the people that were there." Although members of the MPDG were asked to contribute $10,000 each for Mr. Call's opinion letter, which was part of the quid pro quo of membership in the MPDG, the letter was not written until May 1, 1987, and was not distributed until June 1987.

The Ven-Tel modem was designed to be Hayes compatible, and was advertised as such. During an MPDG meeting in San Jose in January 1987, Robert Stitt, Director of Personal Computer Products at Ven-Tel and Ven-Tel's representative in the MPDG, stated that "If the Group loses it is fairly certain that willful infringement will most certainly apply." Ven-Tel, as a founder of the MPDG, was aware that litigation against Hayes would be more effective if other modem manufacturers did not take a license from Hayes. All of these factors could have influenced the jury. Thus, the evidence supports an inference by the jury that Ven-Tel made its decision to infringe the '302 patent without a good-faith belief in its invalidity, and that it used the opinion letter as a basis for forming a patent defense group rather than as a genuine basis for decision-making. Although Ven-Tel received advice of counsel on the invalidity of the '302 patent, advice of counsel alone cannot be used as a shield irrespective of the nature and timing of that advice in the context of the surrounding circumstances. Substantial evidence supports the jury's conclusion that Ven-Tel's infringement of the '302 patent was willful.

The district court granted Hayes' motion for a permanent injunction, but stayed the injunction pending appeal. In granting the stay, the court permitted continued sales by Ven-Tel and required Ven-Tel to pay a royalty of 1.75% into escrow. Additionally, the district court awarded Hayes double damages, granted Hayes' request for an accounting of infringing sales from December 31, 1990 until April 22, 1991, the date of the court's judgment and order, and awarded enhancement of this portion of the damages as well. Hayes cross-appeals, arguing that the district court abused its discretion by failing to enhance the interim royalty during the pendency of this appeal. Hayes asserts that during the appeal period, Ven-Tel is still a willful infringer, albeit one paying an interim royalty imposed by the district court. According to Hayes, it is inconsistent for Ven-Tel to pay an enhanced royalty rate on all damages except those during the appeal period.

A finding of willfulness may be a basis for an award of enhanced damages under 35 U.S.C. § 284 (1988).13 See, e.g., Underwater Devices, 717 F.2d at 1389-90, 219 USPQ at 576-77 (upholding finding of willful infringement and award of treble damages). This court will not disturb a trial court's refusal to award enhanced damages unless we determine that it abused its discretion. Rite-Hite Corp. v. Kelley Co., 819 F.2d 1120, 1126, 2 USPQ2d 1915, 1919 (Fed.Cir.1987).

In this case, the district court awarded Hayes enhanced damages. Hayes' objection is to the failure of the court to enhance the interim royalty imposed in lieu of the permanent injunction, which is stayed pending this appeal. Although we agree with Hayes that it seems somewhat illogical to have awarded Hayes double damages prior to the filing of the appeal, but fail to do so during the appeal, we cannot say that the court abused its discretion. Apparently, although not clearly stated, the district court did consider enhancement of the interim royalty. The district court, when addressing the motion to stay the injunction, stated that "[i]n granting enhanced damages and attorney fees against defendants, the court is imposing substantial penalties on defendants for their willful infringement." In re Hayes, 766 F. Supp. at 823, 20 USPQ2d at 1840 (Footnote omitted). Thus, the enhanced damages that were granted along with the attorney fee award appear to be all that the court intended to serve as a penalty and deterrent. No abuse of discretion is found.

Hayes cross-appeals from the refusal of the district court to impose sanctions under Fed.R.Civ.P. 11 against the defendants and their counsel for pursuing the defense of "fraud on the patent office" and inequitable conduct. The amount requested was $24,600, that which Hayes asserts it spent having Eugene S. Zimmer, the patent attorney who prosecuted the '302 patent, attend the trial. Hayes argues that the defendants knew that the defense of "fraud" and inequitable conduct could not have been supported factually and they presented no such evidence to support them.

The district court stated that "under Rule 11, a pleading, motion, or other paper is frivolous only where it is 'both baseless and made without a reasonable and competent inquiry.' " In re Hayes, 766 F. Supp. at 828, 20 USPQ2d at 1844 (quoting Townsend v. Holman Consulting Corp., 914 F.2d 1136, 1140 (9th Cir.1990)). The court denied sanctions because (1) it found that the conduct of the defendants and their counsel did not meet the threshold for Rule 11 sanctions, and (2) "Mr. Zimmer was an integral part of Hayes' 'trial team', testifying on many aspects of patent law, interpretation, and procedure far removed from simply rebutting an inequitable conduct claim." Id.

Defendants' claim of "fraud" and inequitable conduct was based on their assertion throughout the trial that Heatherington failed to disclose the best mode of his invention, the Hayes SmartModem. We have held that there was no failure to disclose the best mode. Moreover, failure to disclose the best mode may occur without the intent to deceive required for a finding of inequitable conduct. In any event, the proposition that failure to disclose the best mode results in inequitable conduct is legally incorrect. Inequitable conduct requires an intent to deceive as well as a threshold level of materiality. J.P. Stevens & Co. v. Lex Tex Ltd., 747 F.2d 1553, 1559-60, 223 U.S.P.Q. (BNA) 1089, 1092 (Fed.Cir.1984), cert. denied, 474 U.S. 822, 106 S. Ct. 73, 88 L. Ed. 2d 60 (1985). To demonstrate an intent to deceive, the defendants relied on Mr. Call's statement in his opinion letter that there was "fraudulent procurement," and on the statement of the defendants' expert, Warren Jessup, that there was "perhaps" a deliberate concealment of the best mode of the invention by the inventor of the '302 patent. Conjecture alone is not sufficient to show an intent to deceive to support the defense of inequitable conduct.

Although we deplore such unfounded accusations of inequitable conduct as occurred here, we nevertheless uphold the district court's denial of sanctions under Fed.R.Civ.P. 11. We review the failure to award Rule 11 sanctions under the abuse of discretion standard. Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 405, 110 S. Ct. 2447, 2460, 110 L. Ed. 2d 359 (1990). The district court denied the award of sanctions both because it determined that the threshold for Rule 11 sanctions was not met and because Mr. Zimmer, the patent attorney accused of inequitable conduct, was an integral part of Hayes' "trial team." We cannot say that the district court abused its discretion in failing to award Hayes sanctions under Fed.R.Civ.P. 11.

We affirm the district court's judgment that the '302 patent is not invalid and is infringed; we affirm the district court's partial denial of enhancement of damages and sanctions under Fed.R.Civ.P. 11.