Wyko Corporation appeals the final judgment of the United States District Court for the District of Arizona (Civil Action No. CIV 88-454-T-JLQ), which held Zygo Corporation’s United States Patent No. 4,201,473 (“the ’473 patent”) for an interferometer system not invalid and infringed by three models of Wyko interferometers, resulting in an award of $2,668,710.43 in damages calculated in part as lost profits and in part as a reasonable royalty. We hold that the district court clearly erred in finding that one model, the Wyko 6000 Redesign, infringed under the doctrine of equivalents. We affirm the finding of infringement on the Original Wyko 6000 designs.
In sum, for reasons discussed more fully below, we affirm-in-part, reverse-in-part, and remand for recalculation of damages.
I.
Background
Dissatisfied with existing interferometers, Zygo set out to design and patent its own devices. In 1977, Zygo designed the Zygo Mark II, the commercial embodiment of the invention claimed in the ’473 patent. The patented interferometer has two modes of operation, alignment and viewing. In the alignment mode, the internal TV camera is focused down the “alignment leg.” The reference and test wavefronts are collected by a series of optic elements and projected as two dots of light onto a stationary diffuse screen. Embedded into the stationary diffuse screen is a visible cross-hair reticle with a marked center indicating the optical axis. The image on the screen of the dots and reticle is picked up by a TV camera and displayed on a CCTV monitor. An operator moves the dots to the center of the alignment reticle at which point alignment is achieved.
To view the fringe pattern, the interferometer must be switched into the “view mode.” When the view mode is entered, a flip mirror is moved and a TV camеra is focused down the “view leg,” which permits the combined wavefronts to be imaged as a fringe pattern onto a second diffuse screen. This diffuse screen is movable with respect to the path of the wavefronts so as to destroy the coherence length of the beam 1 and to eliminate spurious fringe patterns. The view leg includes a continuous zoom lens positioned between the camera and the movable diffuse screen. The diffuse screen, together with the fringe pattern caused by the interference of the reference and test wavefronts in the recombined beam, is imaged onto the camera.
The inventors constructed a “breadboard” of the patented interferometer during the course of their development work. The breadboard, which was bufit on a 2’ by 2’ plate, consisted of the various optical components disclosed in the ’473 patent. In transforming the breadboard interferometer into a functioning commercial unit, Zygo made several necessary additions. For example, the entire device was encased in a metal box for protective purposes, and a spherical field lens, which functioned to make the alignment spots appear brighter on the television monitor, was built into the back of the stationary view screen. As a result of encasing the device, a mirror had to be added to illumi *1566 nate the view screen and improve visibility. The ’473 patent does not disclose the enclosure, the spherical field lens, or the illumination mirror.
On April 21, 1978, Zygo filed the patent application which matured into the ’473 patent. The claims were initially rejected under 35 U.S.C. § 103 as being obvious in view of a brochure describing a prior art interferometer, the Tropel reference. Zygo responded with argument that the claims as filed were patentably distinct over that prior art. In particular, Zygo noted that the Tropel reference failed to teach the use of zoom magnification, an alignment reticle, or means for destroying coherence length (i.e. the movable diffuse screen). The Examiner then allowed the claims, and the ’473 patent issued on May 6, 1980.
The Accused Device
Defendant Wyko Corporation was founded in 1982 by Dr. Wyant, a professor at the University of Arizona, and Mr. Chris Koliap-olis, a former doctoral candidate student of Dr. Wyant’s. This litigation centers around two Wyko optical interferometer systems: the original Wyko 6000 and the Wyko 6000 Redesign. The original Wyko 6000 interferometer was first introduced in January 1988. The Wyko 6000 Redesign made its debut about one year after initiation of this lawsuit.
The original Wyko 6000 included a stationary diffuse screen onto which the rеference and test wavefronts were focused as dots of light which were imaged onto a television camera, converted into electronic data, and appeared on a computer monitor along with a reticle for alignment that was generated by computer software. The two dots were tilted and tipped until they overlapped at the center of the cross-hair which meant they were aligned along the optical axis of the interferometer.
In the Wyko 6000 Redesign, no reticlе is displayed for alignment. Instead, a prism is used which is factory-positioned so that its axis is parallel to the optical axis of the interferometer. When so positioned, the prism deflects light that is not parallel to its axis. If the wavefronts are not in alignment, the prism creates a split-image of the test and reference wavefronts such that four dots (two for each wavefront) are projected onto the diffuse screen. This image is transferred to a TV monitor. Both pairs of dots are moved together. When рroperly aligned, the four dots coalesce into a single spot, which means they are aligned along the optical axis of the interferometer.
In the view mode, both the original and redesigned Wyko 6000 systems included a rotating ground glass screen positioned in front of a continuous zoom lens. Similar in purpose to the moveable screen in the Zygo interferometer, the rotating ground glass screen functioned to eliminate spurious fringe patterns. There is no dispute that the Wyko devices meet the limitations respecting the view mode.
Procedural History
The court entered its Findings of Fact and Conclusions of Law on March 3, 1993, holding the ’473 patent was neither invalid (1) for lack of an enabling disclosure or (2) for failing to disclose the best mode, nor was the patent unenforceable for inequitable conduct before the Patent Office. While none of the accused Wyko devices were found to literally infringe the ’473 patent, the court concluded that all versions of the accused Wyko devices infringed under the doctrine of equivalents.
The damаges phase of the trial was similarly tried to the bench, after which the court entered its Findings of Fact and Conclusions of Law and awarded Zygo over $2.1 million in total lost profits and a reasonable royalty. The court also awarded prejudgment interest in the amount of $532,755.04.
On appeal, Wyko alleges the district court erred in the finding that the accused devices infringed the ’473 patent under the doctrine of equivalents, and in concluding that the ’473 patent was not invalid for failure to disclose the best mode. Wyko also challеnges the award of damages calculated as lost profits.
II.
Best Mode
Compliance with the best mode re
*1567
quirement of 35 U.S.C. § 112 ¶ 1
2
is a question of fact.
Scripps Clinic & Research Found. v. Genentech, Inc.,
Prior to filing the patent application, the inventors made an optical breadboard embodiment of the invention. Also prior to filing, Zygo made a commercial embodiment which encased the interferometer in a box. To make the encased interferometer work, it was necessary tо add an illumination mirror and a spherical field lens. Only the interferometer, that is, without any enclosure together with the add-ons thereby made necessary, is disclosed and claimed.
Wyko argues that the encased interferometer — the commercial embodiment — was the best mode contemplated by the inventors for carrying out their invention. Zygo argues that the court properly concluded that the best mode requirement of section 112 ¶ 1 was met here because the claimed invention is not an encased interferometer and that section 112 does not mandate disclosure of activities done in furtherance of commercialization of the claimed invention.
The phrases “best mode” and “carrying out the invention” are not statutorily defined. What is a “mode” of the “invention”? What acts or ideas are meant to be encompassed by the phrase “carrying out the invention”? These questions are not answered by a mechanical rule.
See Wahl Instruments, Inc. v. Acvious, Inc.,
Wyko’s best mode defense amounts to an attack on the nondisclosure of a particular embodiment, a commercial mode for a particular type of use of the claimed invention. At least one of the inventors contributed to the commercial design. The failure to disclose the commercial mode, however, does not
ipso facto
result in a section 112 violation. The focus of a section 112 inquiry is not what a particular user decides to make and sell or even in what field the invention is most likely to find success. Rather, in keeping with the statutory mandate, our precedent is clear that the parameters of a section 112 inquiry are set by the claims.
Engel Indus., Inc. v. Lockformer Co.,
Here, the claimed invention is an interferometer system which has a number of applications. The inventors disclosed the mode for carrying out that claimed invention: a light source, beamsplitters, various optical elements, a diffuse screen with a reticle, a camera, and a monitor arranged in a particular manner. It is undisputed by the parties that the disclosure is adequate to enable one of skill in this art to practice the invention of the claims.
Chemcast,
With respect to the subject matter of the claims, an enclosed mode may or may not be “best” for a particular commercialization. But it is clear that an enclosure is not a necessary part of this invention and there is no evidence that either inventor considered an enclosed interferometer to be the best mode of carrying out the invention.
Cf. Dana Corp. v. IPC Ltd. Partnership,
Under all of the circumstances of this case, the trial court’s finding that Wyko failed to establish a violation of the best mode requirement of section 112 ¶ 1 is not clearly erroneous. Accordingly, the court’s conclusion of law that the ’473 patent is not invalid is affirmed.
III.
Infringement Under the Doctrine Of Equivalents
The infringement dispute in this case centers on the claim language “means for collecting said two wavefronts and focusing them as spots onto a diffusе SCREEN Containing AN INTEGRAL ALIGNMENT RETICLE HAVING A marked Center.” The court construed the emphasized language to mean the alignment reticle and the diffuse screen were “one piece.” Literal infringement is not an issue. No Wyko device contains a unitary screen and an alignment reticle as required by the claim. The only issue is infringement under the doctrine of equivalents. We affirm the finding of infringement by the original Wyko 6000 models but reverse with respect the Wyko 6000 Redesign.
Infringement under the doctrine of equivalents may be found where those limitations of a claim not found exactly in the accused device are met equivalently. In
Pennwalt Corp. v. Durand-Wayland, Inc.,
It is well settled that each element of a claim is material and essential, and that in order for a court to find infringement, the plaintiff must show the presence of every element or its substantial equivalent in the accused device. Lemelson v. United States,752 F.2d 1538 , 1551,224 USPQ 526 , 533 (Fed.Cir.1985).
See Athletic Alternatives, Inc. v. Prince Mfg., Inc.,
A.
In the original Wyko 6000 devices, a reticle which coincided with the location of the optical axis, in the words of the claim, an “alignment reticle,” was generated by hard *1569 ware or software and appeared on a monitor together with the two dots imaged by the camera from the diffuse screen. 3 Thus, the image on the Wyko monitor duplicated the required image on the monitor in the alignment mode of the invention. We have no difficulty agreeing that the image of the reti-cle generated on the monitor is the equivalent of the physical reticle on the screen which is also imaged on the monitor in the Zygo invention. While the alignment reticle has beеn moved within the alignment leg of the accused interferometers, the function of alignment is accomplished in the same way, by moving the two spots to coalesce at the center of a reticle which coincides with the optical axis of the interferometer. Similarly, the point at which the crosshairs cross is at least equivalent to a marked spot.
In sum, the evidence supports the trial court’s finding of equivalency between the alignment reticles of the accused and patented devices both of which are the means for alignment. Inasmuch as all other elements of the claims are admittedly met in Wyko’s original devices, this ultimate finding of infringement is not clearly erroneous.
B.
Turning to the Wyko 6000 Redesign, we reach the opposite conclusion. In these devices, an image inverting prism is placed in the path of the wavefronts that emerge from the diffuse screen. Each portion of the wavefronts that passes around the prism is projected as a spot on the camera. Each part of the wavefrоnt passing thRough the prism is deflected and projected as a separate spot on the camera. The spots are mirror images. When these four spots coincide, the positions of which can be adjusted by the operator, alignment is achieved. In the Redesign, there is no substitute “alignment reticle,” that is, crosshairs defining the optical axis of the interferometer. The “alignment aid” is the prism structure which splits each of the beams on its way to the camera into two so that four dots appear on thе monitor. These four dots are moved so as to coalesce into one, at which point the beams are centered on the axis of the interferometer. Thus, to find infringement, the prism structure must be found an equivalent of an alignment reticle placed on the optical axis, a finding which is not supportable. That both can be labelled “alignment aids,” as Zygo terms them, does not answer the question whether they are equivalent within the concept of infringement. That the sub-stituent performs the same general function tо achieve the same result as the required element also does not establish their equivalency. The result must be achieved in substantially the same way.
Dolly,
Here, there is simply no dispute that the alignment aids in the two devices are not only structurally different but also operate in a different way. Making four beams coalesce without a reticle for centering on the optical axis is obviously not the same principle of operation as making two beams coalesce at the center of an alignment reticle. A finding of equivalency just because the same result is achieved is “a flagrant abuse of the term ‘equivalent’.”
Burr v. Duryee,
One complicating factor here is that a software generated reticle can be made to appear on the computer monitor of the Wyko Redesign and the four alignment spots happen to coalesce very near the center of that reticle. Zygо asserts that because of this reticle in the Wyko Redesign, these devices also meet the limitations of “an alignment reticle integral with the screen.” Wyko counters that the reticle on the computer monitor is not positioned exactly on the optical axis and, therefore, unlike in the original Wyko 6000, this reticle cannot serve as a meaningful reference. The record establishes that the software in the original was replaced in the Redesign. The new software creates a reticle which is used оnly to assist factory setup of the view mode intensity display. It serves no purpose on the comput *1570 er monitor in the alignment mode, and the computer must be misadjusted for it even to be seen.
Zygo argues that under the precedent of
Intel Corp. v. United States Int’l Trade Comm’n,
In any event, the alignment aid in the Wyko Redesign is the prism, not an alignment reticle, ie., one placed on the cеnter of alignment. The reticle in the Redesign is not positioned as an “alignment reticle,” does not function as one, and is present in the device for an entirely different purpose. Further, using spots through and around the prism is a distinctly different “way” of achieving alignment from imaging spots directly from the diffuse screen to a monitor for alignment on a displayed reticle. Because the Redesign operates on different optical principles, the same “way” prong of the tripartite test has not been satisfied. 4
Finally, fоr purposes of infringement under the doctrine of equivalents, the differences between the claimed device and the accused device must be insubstantial.
Graver Tank & Mfg. Co. v. Linde Air Prods. Co.,
Wyko’s patent, against which the ’473 patent was cited and considered as prior art, is thus presumed nonobvious in view of the ’473 patent until proven otherwise. The nonobviousness of the accused device, evidenced by the grant of a United States patent, is relevant to the issue of whether the change therein is substantial.
Roton Barrier, Inc. v. Stanley Works,
In sum, the accused interferometer with the prism alignment system simply can not be deemed an insubstantial change from Zygo’s claimed interferometer with the reticle alignment system.
Hilton Davis,
IV.
Damages
In view of the reversal of the finding that the Wyko 6000 Redesign infringed the ’473 patent, the damages award is vacated. The trial court’s analysis did not distinguish between the Original and the Redesign Wyko models in addressing the issues of lost profits, convoyed sales, and reasоnable royalty, and it is unclear that the court’s findings are applicable to the Original Wyko alone. Un *1571 der these circumstances, we remand with instructions to recalculate the damages due Zygo for the infringement by the Original Wyko models.
Lost Profits
The central damages issue on appeal is whether the court erred in concluding that Wyko’s SIRIS interferometer was not an acceptable noninfringing alternative to the Zygo interferometers. The record indicates that Wyko stopped marketing the SIRIS interferometer when it bеgan marketing the Wyko 6000 interferometer. Wyko argues that the award of lost profits was error on the theory that they would have continued making and selling the SIRIS had they not made and sold the Wyko 6000 interferometers.
It is axiomatic, however, that if a device is not available for purchase, a defendant cannot argue that the device is an acceptable noninfringing alternative for the purposes of avoiding a lost profits award. A lost profits award reflects the realities of sales actually lost, not thе possibilities of a hypothetical market which the infringer might have created. Thus, whether or not the SIRIS was an acceptable noninfringing alternative is relevant only for the period of time that the SIRIS interferometer was being marketed by Wyko.
The court’s finding that the SIR-IS interferometer was not an acceptable non-infringing alternative is reviewed for clear error.
See Minnesota Mining and Mfg. Co. v. Johnson & Johnson Orthopaedics, Inc.,
The SIRIS interferometer with the FIZ4 adapter, like the Wyko 6000 and the Mark IV interferometers, was a phase-shifting Fizeau type interferometer. The software that was used in the Wyko 6000 was “basically the same packаge” as used in the SIRIS interferometer and the SIRIS was “repackaged” and renamed the 6000. Furthermore, there was ample testimony regarding the capabilities of the SIRIS interferometer with the FIZ4 adapter, the advantages of the SIRIS and the disadvantages of the Mark IV interferometer. For example, whereas the SIRIS used a “powerful Hewlett-Packard computer,” the Mark IV had a Zygo-built computer that lacked an internal hard drive, was “very slow” and had “very poor computer graphics.” There was also testimony that the SIR-IS and Mark IV interferometers could test the same components, served the same applications, and that there was nothing the Mark TV could do that the SIRIS with the FIZ4 adapter could not.
While we do not, on this record, have a definite and firm conviction that a mistake has been made warranting reversal on this issue,
United States v. United States Gypsum Co.,
Reasonable Royalty
In contrast, the fact that Wyko could have continued marketing the SIRIS is a factor relevant to the determination of a proper royalty during hypothetical negotiations. Wyko would have been in a stronger position to negotiate for a lower royalty rate knowing it had a competitive noninfringing *1572 device “in thе wings.” On remand, the court shall reconsider its award of a 25% royalty rate in light of Wyko’s ability to market the noninfringing SIRIS in lieu of marketing the infringing Original Wyko 6000.
V.
Conclusion
The court’s conclusion that the ’473 patent is not invalid for failure to set forth the best mode is affirmed, as is the finding that the original Wyko 6000 infringed claim 1 of the ’473 patent under the doctrine of equivalents. The court’s finding that the Wyko 6000 Redesign infringed claim 1 under the doctrine of equivalents is reversed. The case is remanded to the district court for a rede-termination of the damage award consistent with this opinion.
Costs
Each party shall bear its own costs.
AFFIRMED-IN-PART, REVERSED-IN-PART, AND REMANDED.
Notes
. "Coherence length” is defined as the maximum tolerable optical path length difference between two energy beams which are forming an interference pattern. Handbook of Chemistry and Physics, F-83 (Robert C. Weast, Ph.D. ed., 63rd ed.1982-83)
. Section 112 reads in pertinent part:
The specification ... shall set forth the best mode contemplated by the inventor for carrying out his invention.
. In the first four original units, the alignment crosshair was displayed on the computer monitor only. In the 14 modified original units, the reticle also was generated on the TV monitor.
. We note that the trial court's original opinion contained language in agreement with our disposition:
The Wyko Redesign utilizes a different way, that is, a prism, to create a split image alignment aid focused onto a diffuse screen, with no alignment aid physically appearing on the television monitor, which is a different way of indicating the spot where the optical axis is.
However, the court issued an Amended Opinion from which this language was stricken and replaced with a contrary conclusion.
. That the SIRIS was noninfringing was not, as noted by the court, in dispute below, nor does Zygo assert the SIRIS infringed the '473 patent on appeal.