BOWNES, Circuit Judge.

Milgo Electronic Corporation and International Communication (Milgo) appeal from an adverse declaratory judgment in a patent validity action brought by Codex Corporation and Yellow Freight Systems, Inc. (Codex). Codex cross-appeals because of the failure of the district court to grant all of the relief it sought and on the ground that the amount of attorney fees awarded it was too low. The district court opinion is reported, Codex Corp. v. Milgo Electronic Corp., 534 F.Supp. 418, 432 (D.Mass.1982).

The dispute revolves around three patents, all owned by Milgo as assignee:

(1) No. 3,524,023, Sang Y. Whang, inventor, Band Limited Telephone Line Data Communication System Whang 023);

(2) No. 3,619,503, Robert G. Ragsdale, inventor, Phase and Amplitude Modulated Modem (Ragsdale ’503); and

(3) No. 3,783,194, Viesturs V. Vilips, inventor, Data Modem Having a Fast Turnaround Time Over Direct Distance Dialed Networks (Vilips ’194),

I. Background

All computer machine languages operate on a binary number system. This system or language involves only two elements: positive or negative or, most commonly, 1 or 0.

Computers do not always operate alone; some are built to communicate with other computers. In order for one computer to talk with another some sort of communication link has to be established. At the time the patents in this case were issued the primary communication link was direct distance dialed telephone lines (DDD). Unfortunately for talking computers, binary or digital data cannot be transmitted over DDD lines rapidly and reliably. To transmit information through DDD lines the transmission must be made in an analog form. This analog transmission is best thought of in terms of a sinusoidal waveform (sine wave) as illustrated below:

[[Image here]]

The above illustration shows a sine wave with an amplitude of one and a cycle duration of T seconds. Both of these terms will be explored in some detail later.

For computers to talk one with the other it is necessary to transform digital signals to analog at the transmitting computer and back again at the receiving computer. This transformation is accomplished in both instances by a modem.

The word “modem” is short for modulator-demodulator. Modulation is the alteration of the sine wave in some manner so as to impart some information to it. For the *625 purposes of this litigation modulation can take one of three forms: phase modulation, amplitude modulation, or a combination of both phase and amplitude modulation.

Phase modulation, or more specifically in this case differential phase modulation, is accomplished by sending out a signal pulse during a modulation period followed by another modulation period with a signal pulse of a different phase. The receiving modem detects this phase shift which contains the information in each signal pulse.

Amplitude modulation is accomplished in the transmitting modem by changing the amplitude of the sinusoidal sine wave from one modulation period to the next. The receiving modem detects this change in amplitude thus reading the information contained in the signal pulse. The combination of phase and amplitude modulation is accomplished by changing both the phase and amplitude of the sine wave from one modulation period to the next.

The speed at which information can be transmitted depends primarily on two factors. First, it depends on the number of signal pulses (bauds) that can be transmitted per second, and secondly, it depends on the number of discrete “bits” of binary data (1 or 0) that can be encoded in each baud (bits per baud).

A problem with analog signal transmission is that some distortion of the signal is bound to occur; the received signal is not going to be identical to the one transmitted. This distortion is a function of two characteristics, one of the signal, the other of the DDD line itself.

The problem with the DDD line is that at the upper and lower ends of the frequency range available the analog signal is susceptible to amplitude and delay distortion. These distortions are fatal to fast and accurate transmission of information. It was well-known in the early 1960’s that at the center of this available frequency range there existed a “sweet spot.” The sweet spot is a band of about 1000 Hz 1 in a range between the frequencies of 1200 Hz and 2200 Hz which has the characteristic of being relatively free from distortions.

The problem with the signal occurs when it is modulated. Modulation causes a dispersal of energy up and down the frequency spectrum. This energy, outside of the sweet spot, is susceptible to delay and amplitude distortion. The delayed reception of this energy by the receiving modem will distort the apparent phase and amplitude of the received signal causing inaccurate decoding of the information.

To solve this problem of energy dispersal over the frequency spectrum a composite filter to filter out energy at all but the desired frequencies is used. This solution, however, causes its own problem, it spreads the signal out over time. The energy representing the signal in one baud is smeared in time so that some of it is still “ringing” or echoing in the bauds that succeed it. This ringing causes intersymbol interference or confusion from one baud to the next. The next step was to solve this problem.

In 1928 Harry Nyquist published his discovery that an information carrying pulse, or baud, of duration in time of T seconds required a bandwidth of 1/T Hz for accurate transmission. This basic constraint on the bandwidth is crucial to mitigating the intersymbol interference.

Nyquist went on to describe a filter which would accomplish this. The ideal Ny-quist filter, called “brickwall,” passes only this 1/T bandwidth. It was recognized that the actual construction of such a brickwall filter was not possible. Practical filter design requires the use of filters with a roll-off of greater than zero. The diagram and text which follows illustrates both the brickwall filter and the concept of roll-off.

*626 [[Image here]]

The carrier frequency (fc) is the frequency of the sine wave which is modulated by the modem to produce the information-carrying signal pulse. The portions of the bell-shaped curve that extend beyond the brick wall are known as the skirts. One way of defining the roll-off of the filter is by taking the ratio of the width of the skirt along the abscissa to the width of the Ny-quist ideal of 1/T Hz. A composite filter exhibits 50% roll-off when the sum of the skirts is 1/2T Hz. It should also be noted that to meet Nyquist’s criteria a roll-off of 100% is the maximum allowable.

One further necessary element of modem design is a mechanism for determining when a baud begins and ends. Concomitant with that determination is the need to know when or where to read the information contained in a baud. The mechanism for determining when to sample or read the information contained in a baud is called “clock” in the present art. It is a characteristic of the 1/T bandpass filter that the energy of the signal pulse will be minimum at the beginning of the pulse, build to a maximum at approximately the middle of the pulse, and then decay to a minimum point at the end of the pulse. It is this characteristic that permits clock recovery by the receiving modem. This allows the receiving modem to read the information at the center of each signal pulse. Nyquist revealed that with a 1/T bandpass filter the interference or ringing caused by one baud would be zero at the center of the succeeding baud. Thus, the concept of center sampling was born; the idea being to read the information encoded in a baud as close as practicable to the center of each baud.

The presence of echo suppressors on the DDD lines present yet another problem to modem designers. Echo suppressors permit signals to be transmitted in only one direction at a time, the preference being given to the stronger signal. It is a characteristic of DDD lines and associated amplifiers that if echo suppressors were not present the telephone user would hear echoes of his own voice. The problem with the echo suppressors is that they take 100 milliseconds to reverse direction or turn around. This means that once the first party stops talking and the second party starts it takes 100 milliseconds before the echo suppressor will pass on the second party’s speech energy. For spoken communication a 100 millisecond turnaround delay is no problem. For data communications a 100 millisecond turnaround delay is intolerable.

The solution to this problem is to disable or turn off the echo suppressors by putting a tone on the line. It was well recognized in the early 1960’s that echo suppressors once disabled would remain disabled as long as there was energy on the line. This could be accomplished by a signal transmitted at a frequency other than the frequency of the main data channel. This had been accomplished, up to the time of application for the patents in this case, by utilizing a reverse channel. A reverse channel transmits a signal in a direction opposite to that of the main data signal and at a different frequency. This reverse channel could be used for limited data transmission or solely for the purpose of disabling the echo suppressors in order to reduce the turnaround time.

II. The Whang ’023 Patent

Whang admitted at trial that the only novel feature to be found in his patent, Whang ’023, was “narrow skirts.” That is, a roll-off of 50% or less. The other salient features of Whang ’023 such as differential phase modulation, limiting the passband to 1/T Hz and center sampling were already known both separately and in combination. *627 We agree with the district court’s conclusion that the asserted novelty, a composite filter roll-off of 50% or less, is nowhere to be found in any of claims 1, 19, or 25 either expressly or by implication. Since the asserted novelty does not, in fact, exist there is no invention.

The ultimate question of patent validity is one of law. Graham v. John Deere Co., 383 U.S. 1, 17, 86 S.Ct. 684, 693, 15 L.Ed.2d 545 (1965). A determination of validity, however, requires certain factual inquiries, such as anticipation of prior art. Carter-Wallace, Inc. v. Gillette Co., 675 F.2d 10, 15 (1st Cir.1982). The findings of fact by the court below can be set aside only if they are clearly erroneous. Forbro Design Corp. v. Raytheon Co., 532 F.2d 758, 763 (1st Cir.1976); Fed.R.Civ.P. 52(a). This is a recognition of the trial court’s superior position with respect to factual determinations. Any resolution of factual questions “requires a balancing of credibility, persuasiveness and weight of evidence.” Graver Mfg. Co. v. Linde Co., 339 U.S. 605, 609, 70 S.Ct. 854, 857, 94 L.Ed. 1097 (1949). “A finding is ‘clearly erroneous’ when, although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” United States v. Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1947).

Milgo urges that because claims 1, 19, and 25 are “means plus function” claims we must turn to the specification for explanation. We agree that this is the correct approach, 35 U.S.C. § 112, but contrary to Milgo’s assertion we do not find any language in the specification describing narrow skirts. It is true that “a patent should not be struck down ... where a reasonable construction of the specifications and claims will protect the invention .... ” Corning Glass Works v. Anchor Hocking Glass Corp., 374 F.2d 473, 478 (3d Cir.1967), and that a patentee may be his own lexicographer, choosing his own words, “so long as he remains consistent in their use and makes their meaning reasonably clear.” . Ellipse Corp. v. Ford Motor Co., 452 F.2d 163, 167 (7th Cir.1971), cert. denied, 406 U.S. 948, 92 S.Ct. 2041, 32 L.Ed.2d 337 (1972). But it is also true that the law requires the specification to be written in “such full, clear, concise, and exact terms as to enable any person skilled in the art ... to make or use the same [invention] .... ” 35 U.S.C. § 112. The inventor may not transform the claim into “a nose of wax which may be turned and twisted in any direction, by merely referring to the specification, so as to make it include something more than, or something different from, what its words express.” White v. Dunbar, 119 U.S. 47, 51, 75 S.Ct. 72, 74, 30 L.Ed. 303 (1886).

The 50% roll-off requirement is nowhere explicitly stated in the specification, nor is it implied. In fact, the specification can be read to teach the opposite. As Whang admitted, and the district court found, it was left up to the design engineer to decide what roll-off to use, with the proviso that it meet the Nyquist criteria. This would put the roll-off anywhere from 0 to 100%. This is not the degree of specificity required by 35 U.S.C. § 112 so as to make 50% roll-off part of the invention.

Milgo seeks to justify its position through a tortured reading of Eibel Process Company v. Minnesota & Ontario Paper Company, 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523 (1923). In Eibel the critical factor of pitch in the invention was defined in the claims by the terms “high” and “substantial.” The figure which accompanied the specification illustrating this improvement to pitch indicated an angle of 4% or an elevation of 12 inches. References to small elevations shown in prior art devices indicated that the patentee had in mind elevations substantial as compared to them.

In Eibel the critical factor was pitch; in Whang ’023 it is the 50% roll-off. Whang ’023, unlike the patent in Eibel, gives no numerical representations of the critical factor. Neither does it discuss the percentage roll-off in prior art so as to indicate to those skilled in the art that a roll-off of 50% or less is an essential element in the function of Whang ’023.

*628 Since 50% roll-off is not included in claims 1, 19, or 25 of Whang ’023 the question becomes whether the named claims are valid without the narrow skirts. We first state the legal principles involved.

The claims of a patent in order to be patentable and valid must, among other things, not have been “described in a printed publication in this or a foreign country ... more than one year prior to the date of application for patent in the United States .... ” 35 U.S.C. § 102(b). In order for a publication to anticipate an invention as defined in a patent claim under 35 U.S.C. § 102(b), the publication “must disclose all the elements of the claimed combination, or their equivalents, functioning in substantially the same way to produce substantially the same results.” Decca Limited v. United States, 420 F.2d 1010, 1027, 190 Ct.Cl. 454, cert. denied, 400 U.S. 865, 91 S.Ct. 102, 27 L.Ed.2d 104 (1970).

To give the grant of a patent substance and value there is a rebuttable presumption that the patent is valid. 35 U.S.C. § 282. Evidence of prior art not considered by the Patent and Trademark Office (PTO), especially in combination with evidence of omissions or inaccuracies in prior art presented to the PTO, eviscerates the presumption of validity. Parker v. Motorola, Inc., 524 F.2d 518, 521 (5th Cir.), cert. denied, 425 U.S. 975, 96 S.Ct. 2175, 48 L.Ed.2d 799 (1975). The presumption of validity can be strengthened by a prior judicial determination of validity. American Home Products Corp. v. Lockwood Mfg. Co., 483 F.2d 1120, 1125 (6th Cir.), cert. denied, 414 U.S. 1158, 94 S.Ct. 917, 39 L.Ed.2d 110 (1973). However, that which strengthens the presumption and even the presumption itself can be rebutted by a showing of clear and convincing evidence to the contrary. Saf-Gard Products, Inc. v. Service Parts, Inc., 532 F.2d 1266, 1271 (9th Cir.), cert. denied, 429 U.S. 896, 97 S.Ct. 258, 50 L.Ed.2d 179 (1976). It should be noted that while the presumption of validity can be rebutted, the burden of persuasion “is and always remains upon the party asserting invalidity .... ” Solder Removal Co. v. United States International Trade Commission, 582 F.2d 628, 633, 65 CCPA 120 (1978).

The district court found that the presumption of validity which attached to Whang ’023 was rebutted by three pieces of prior art not before the PTO when Whang ’023 was issued. Further, the district court accorded no comity to the prior adjudication of validity of Whang ’023 by the district court of Kansas in Milgo Electronics Corp. v. United Telecommunications, Inc., 189 USPQ 160 (D.Kan.1976), aff’d, 623 F.2d 645 (10th Cir.), cert. denied, 449 U.S. 1066, 101 S.Ct. 794, 66 L.Ed.2d 611 (1980). We first discuss the question of comity and then the prior art.

As discussed earlier, a patentee can be his own lexicographer only if the specification clearly indicates the unique meaning that he wishes to attach to a word or phrase. Unfortunately, the Kansas District Court misapplied this rule of law. The Kansas court allowed Whang to define the term “passband width” as found in claims 1, 19, and 25 so as to incorporate 50% roll-off. This is contrary to the lexicographer rule because Whang’s definition was not one used by those skilled in the art nor is the unique meaning of “passband width” made reasonably clear anywhere in Whang ’023 to those skilled in the art. This is a case “where comity, being a rule of convenience intended to persuade, not to command ..., should not prevail against an opposite judgment when based upon clear conviction.” New York Scaffolding Co. v. Liebel-Binney Const. Co., 243 F. 577, 581 (3d Cir.1917), aff’d, 254 U.S. 24, 41 S.Ct. 18, 65 L.Ed.2d 112 (1920); see, Barr Rubber Products Co. v. Sun Rubber Co., 425 F.2d 1114, 1120 (2d Cir.), cert. denied, 400 U.S. 878, 91 S.Ct. 118, 27 L.Ed.2d 115 (1970). We agree with the district court’s conclusion that no comity should be given to the findings of the District Court of Kansas.

We also agree that Whang ’023 is invalid for lack of novelty. Absent the narrow skirts, which are not part of the patent, Whang ’023 lacks any novel improvement over the prior art. The district *629 court’s finding that any of the three references cited fully anticipated claims 1, 19, and 25 of Whang ’023 was correct. 2

Milgo, by a footnote in its brief, would have us remand this action to the district court for a determination of the validity of dependent claims 2-5, 10, 11, 20, 27-29, 36, and 37 of Whang ’023. Codex, who originally brought the declaratory judgment action in the district court, specifically asked the court to be silent on the validity of the dependent claims. Milgo had every opportunity in the court below to assert the dependent claims in their cross-claim for infringement. It failed to do so and will not be heard to complain on appeal. Moreover, in any future action Milgo would be collaterally estopped from asserting the validity of the dependent claims on the basis that they incorporate narrow skirts, 50% roll-off, or 2% energy at the passband edges, all of which are functionally equivalent. “It is the issues litigated, not the specific claims around which the issues were framed, that is determinative.” Westwood Chemical, Inc. v. United States, 525 F.2d 1367, 1372, 207 Ct.Cl. 791 (1975). See Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, et al., 402 U.S. 313, 91 S.Ct. 1434, 28 L.Ed.2d 788 (1971).

III. Ragsdale ’503

Milgo contends, on two grounds, that it was error for the district court to have entered summary judgment for Codex finding claims 1 and 5 of Ragsdale ’503 invalid. First, they argue that summary judgment may not be entered on an oral motion at trial, and second, they assert that claims 1 and 5 of Ragsdale ’503 are valid.

Milgo made no objection to the action of the district court at the time of the ruling on Codex’s motion for summary judgment, although it had an opportunity to do so. This forecloses our review of this issue. Fed.R.Civ.P. 46. “The purpose of Rule 46 is to inform the trial judge of possible errors so that he may have the opportunity to consider his rulings and correct them if necessary.” Stone v. Morris, 546 F.2d 730, 736 (7th Cir.1976); C. Wright & A. Miller, Federal Practice and Procedure § 2472, at 454-56 (1971). The rule that issues not raised in the trial court cannot be considered by the Court of Appeals as a basis for reversal is adhered to save in exceptional circumstances where the obvious result “would be a plain miscarriage of justice.” Hormel v. Helvering, 312 U.S. 552, 558, 61 S.Ct. 719, 722, 85 L.Ed. 1037 (1941); Johnston v. Holiday Inns, Inc., 595 F.2d 890, 894 (1st Cir.1979). We find no miscarriage of justice here.

We agree with the district court’s finding that claims 1 and 5 of Ragsdale ’503 are invalid. It is uncontroverted that Rags-dale ’503 teaches, and claims 1 and 5 cover, a modem containing (a) phase modulation encoding; (b) amplitude modulation encoding; and (c) a composite filter means of a passband width of 1/T Hz. The alleged novelty is the combination of all three elements. The district court found, and we agree, that the combination was not novel. Ragsdale ’503 does not pass muster under 35 U.S.C. § 102(a); it was squarely anticipated by a single prior art reference. 3

IV. Vilips ’194

The district court found Vilips ’194 invalid as obvious based on two prior art references not before the examiner. 4 Milgo asserts that this was error. The Supreme Court has laid out the factual analysis to be *630 followed in passing on obviousness in light of 35 U.S.C. § 103: “Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved.” Graham v. John Deere, 383 U.S. at 17, 86 S.Ct. at 694.

The findings resulting from this factual analyses are reviewed against the clearly erroneous standard. Carter-Wallace, Inc. v. Gillette Co., 675 F.2d at 13. As already discussed in the background section, it was well-known in the prior art that echo suppressors, once disabled, would remain disabled as long as there was energy on the line. In the prior art this was accomplished by a reverse channel which transmitted in a direction opposite to that of the main data signal. This reverse channel was at a different, much lower frequency. While in some prior art references the reverse channel was used to carry data it was recognized that the reverse channel need not carry data but could serve the singular purpose of disabling the echo suppressors. The district court found as follows: “The device described in the Vilips ’194 patent eliminates the data carrying aspect of the secondary channel and causes the continuing tone to be activated by the modem rather than the business machine to which the modem is attached.” Codex Corp. v. Milgo Elec. Corp., 534 F.Supp. at 432. The court went on to describe the prior art’s concept of having the mechanism for applying this tone in the business machine, versus the Vilips ’194 concept of making the device smaller and eliminating the data receiving components. The court found that the elimination of the data receiving components was not an invention but an “improvement on the order of a. mechanic’s expedient.” Id. at 432.

We find no errors in the findings of fact of the district court. Applying the legal standard of obviousness, we affirm the district court’s conclusion that Vilips ’194 was obvious in light of the prior art.

V. Award of Attorney Fees

The district court, pursuant to 35 U.S.C. § 285, 5 awarded attorney fees to Codex, finding that this was an “exceptional case” within the purview of the statute.

A reviewing, court will not interfere with the determination by the district court of attorney fees in a patent case absent an abuse of discretion or an erroneous conception of law. St. Regis Paper Co. v. Royal Industries, 552 F.2d 309, 316 (9th Cir.), cert. denied, 434 U.S. 996, 98 S.Ct. 633, 54 L.Ed.2d 490 (1977); Graham v. Jeoffroy Mfg., Inc., 253 F.2d 72, 78 (5th Cir.), cert. denied, 358 U.S. 817, 79 S.Ct. 28, 3 L.Ed.2d 59 (1958).

Attorney fees are usually awarded under § 285 “only where the district court finds strong evidence of unfairness and bad faith on the part of the losing party.” Colortronic Reinhard & Co. v. Plastic Controls, 668 F.2d 1, 8 (1st Cir.1981); Plymouth Rubber Co. v. Minnesota Mining & Manufacturing Co., 203 F.Supp. 595, 601 (D.Mass.1962), aff’d, 321 F.2d 151 (1st Cir.1963), cert. denied, 375 U.S. 969, 84 S.Ct. 489, 11 L.Ed.2d 417 (1964). To support the award of attorney fees in a patent case there must be precise findings that clearly show the necessary prerequisites of inequitable conduct. Dow Chemical Co. v. Dart Industries, Inc., 475 F.2d 124, 125 (9th Cir.), cert. denied, 414 U.S. 1039, 94 S.Ct. 540, 38 L.Ed.2d 330 (1973).

We agree with Milgo that the proper standard of proof in assessing whether the necessary prerequisites of inequitable conduct exist is the “clear and convincing” standard. Barr Rubber Products Company v. Sun Rubber Company, 425 F.2d at 1120. The relevant inquiry is not whether the district court expressly stated that it was applying the “clear and convincing” stan *631 dard but whether its findings do meet this standard.

The district court made the following findings:

Plaintiffs seek their attorneys’ fees on the ground that this is a special case because defendants have deliberately fabricated the narrow skirt theory and have attempted to assert the Ragsdale ’503 patent which the inventor himself admitted had no novelty in the claims asserted in this trial. I regretfully come to the conclusion that the plaintiffs are correct. The evidence is very strong that Whang knew that the WU 2247 modem had wide skirts, and quite persuasive that he also knew that a number of Milgo modems labeled as covered by the ’023 patent did also. Mr. Whang’s present posture of ignorance of the meaning of terms in his own field of specialty is not at all persuasive. Both Mr. Whang and Attorney Jones are highly trained and sophisticated people. I cannot escape the conclusion that both of these men have deliberately misrepresented the narrow skirt issue to both the District Court of Kansas and to this court. One of the unavoidable hazards of patent litigation is the fact that district judges are likely to have no background in the technology involved. This places a heavy burden on patent lawyers and their expert witness to do their best to mitigate the situation rather than exploit it. Defendants’ post-trial brief strikes me as doing just the reverse; in fact, it verges perilously close to doubletalk.

Codex Corp. v. Milgo Electric Corp., 534 F.Supp. at 433-34.

We hold that these findings, which were supported by the evidence and were clear and precise, meet the “clear and convincing” standard.

Milgo also contends that they were denied a fair opportunity to defend against the charge of deliberate misrepresentation. We believe that the district court’s reliance on Campbell v. Spectrum Automation, 601 F.2d 246 (6th Cir.1979), on this point was well taken. There, as here, the district court found that the party against whom attorney fees were adjudged had acted in bad faith based on the evidence presented at trial. The Court of Appeals for the Sixth Circuit held that where factual issues supporting a finding of “exceptional” circumstances are brought out at trial, the court was authorized to decide a motion under 35 U.S.C. § 285 based solely on the trial record. Id. at 252.

VI. Amount of Attorney Fees

We now turn to the question of the amount of attorney fees awarded Codex. The district court used the “lodestar” approach set forth in Furtado v. Bishop, 635 F.2d 915 (1st Cir.1980), which was based on the analysis used in Copeland v. Marshall, 641 F.2d 880 (D.C.Cir.1980), and Lindy Brothers Builders, Inc. v. American Radiator & Standard Sanitary Corp., 540 F.2d 102 (3d Cir.1976).

The starting point is to calculate the “lodestar”: “The number of hours reasonably expended multiplied by a reasonable hourly rate.” Copeland, at [891]. This would involve separating out work done in relation to a firm’s hierarchy, from senior partner to junior associate (and, we would add, including work that was or ought to have been assigned to a non-lawyer); eliminating time beyond that consistent with a standard of reasonable efficiency and productivity; and, after receiving documentation and possibly holding a hearing, assigning appropriate hourly rates for the kinds of work done by those at different levels of expertise. This results in a “lodestar” fee that then is adjusted upward or downward to reflect the contingent nature of any fee (if such is not reflected in the hourly rate), delay in payment, quality of representation (i.e., an unusually good or poor performance above or below the skill already reflected in the hourly rates), exceptional (and unexpected) results obtained, etc.

Furtado v. Bishop, 635 F.2d at 920 (footnote omitted).

Codex argues that we should modify this approach in a case such as this where the litigation was caused in large part by the *632 unethical conduct of Mugo and its attorneys. It contends that under such circumstances it should be saved completely harmless from the cost of the litigation. It further urges that due to the arms length relationship with its attorneys, great weight should be given to the bargained-for fees. We do not agree.

The fee charged a client by its attorneys is a private matter in which the court, barring unusual circumstances, will not get involved. When, however, a court is compelled by the nature of the case or statutory mandate to award attorney fees to a party, the determination of such award is not only a matter of public record, it becomes part of the great body of our law. A court would be shirking its responsibility to render a principled decision were it to accept without scrutiny and close examination the fees agreed upon by client and counsel. Although Furtado was a civil rights case, its applicability is not limited to that type of case. Indeed, the origin of the “lodestar” analysis was a private, multidis-trict plumbing fixtures antitrust case, Lindy Brothers Builders, Inc. v. American Radiator & Standards Sanitary Corp., 540 F.2d 102. As we noted in Furtado, the “lodestar” approach is one “that can be applied by trial courts in all cases and can also lend itself to meaningful review.” Furtado v. Bishop, 635 F.2d at 920.

The district court found, and we agree, that Codex’s submissions were inadequate. They submitted time sheets, expense sheets, and bills prepared by their attorneys, with monthly totals by attorney. The critical element missing was a categorization of the time spent according to what type of activity was engaged in and by whom.

In line with the first part of the “lodestar” analysis the district court made specific reductions for poor documentation, time unreasonably spent, duplication of effort, and for time spent on issues other than those which were infected by Milgo’s misconduct.

We have examined each of these specific reductions and find no errors of law or abuse of discretion by the district court. Nor did the district court err in refusing to adjust the award upward. We affirm the amount of attorney fees as awarded by the district court.

VII. Post Trial Motions

We will review the denial of a Rule 59 motion only for abuse of discretion. Nimrod v. Sylvester, 369 F.2d 870, 873 (1st Cir.1966); C. Wright & A. Miller, Federal Practice and Procedure § 2818 at 118 (1973). Likewise, the denial of a Rule 60(b) motion will be reversed only where the district court has abused its discretion. Wilkin v. Sunbeam Corp., 466 F.2d 714, 717 (10th Cir.1972), cert. denied, 409 U.S. 1126, 93 S.Ct. 940, 35 L.Ed.2d 258 (1973).

It follows from our discussion of the liability issues that the district court did not abuse its discretion in denying Milgo’s motions for a new trial and relief from judgment.

VIII. Unclean Hands

Codex urges us to find all three patents involved in this action wholly unenforceable because of unclean hands. This issue was not addressed by the district court, although it was presented by Codex in its trial and post trial briefs. Because the facts and the law are clear we think it can be decided without the benefit of findings by the district court.

Codex finds support for its contention in the Supreme Court’s decisions in Precision Co. v. Automotive Co., 324 U.S. 806, 65 S.Ct. 993, 89 L.Ed. 1381 (1945), and Keystone Driller Co. v. General Excavator Co., 290 U.S. 240, 54 S.Ct. 146, 78 L.Ed. 293 (1933).

In Keystone Driller, the patent applicant paid off another inventor to suppress his prior art pertinent to one of a group of related patents. The case emphasizes two points in finding that five related patents were unenforceable due to inequitable conduct with respect to one of them. First, the devices covered by the five patents must be “important, if not essential, parts of the same machine.” Id. at 246, 54 S.Ct. at 148. Second, the inequitable act must have “immediate and necessary relation to the equity that he seeks in respect of the matter in litigation.” Id. at 245, 54 S.Ct. at 147.

*633 In Precision Co., 324 U.S. 806, 65 S.Ct. 993, 89 L.Ed. 1381, there was misconduct involving at least two of the three patents in suit in the Patent Office and all three were tied together by contracts which were consummated as an integral part of the unsavory conduct.

The patents at issue here do not have the strong interconnections which the Supreme Court found so compelling in the cited cases. There is neither the interdependence of all claims in all the patents which the Supreme Court found in Keystone Driller, nor the inequitable conduct touching all the patents which was found in Precision Co.

The maxim of “he who comes into equity must come with clean hands” of necessity gives wide range to a court’s use of “discretion to withhold punishment of behavior which it considers not to warrant so severe a sanction.” Norton Co. v. Carborundum Co., 530 F.2d 435, 442 (1st Cir.1976).

Based on our review of the facts and the applicable law, we refuse to apply the unclean hands doctrine to the three patents as a whole. The findings of the district court as to the invalidity of the specific claims embodied in the three patents is affirmed.

Affirmed. No costs to either party.

Notes

. A Hertz (Hz) is one cycle per second.

. The references are: Bennet & Davey, Data Transmission, McGraw-Hill, N.Y., N.Y. (1965) Chapters 5, 8, 10, 11; Widl, W., An Experimental Data Transmission System, Ericsson Review, Vol. 39, No. 3, pp. 62-71 (1962); Evans, G.L., Enriquez, E., and Wilson, Q.C., A High Speed Serial, Four-Phase Data Modem for Regular Telephone Circuits, Convention Record Global Communications, pp. 100-104 (May 1961).

. Luck, R., (Thesis) Digital Phase and Amplitude Modulated Systems (1961).

. Disablement of Echo Suppressors, CCITT Supplement No. 85, Extract from AT & T Contribution Com. Sp. A/No. 75, pp. 674-688 (July 1963). Data Sets 402C and 402D Interface Specification, Bell System Technical Reference (November 1964).

. 35 U.S.C. § 285 provides: “The court in exceptional cases may award reasonable attorney fees to the prevailing party.”