Aguayo v. Universal Instruments Corp. (356 F. Supp. 2d 699, 2005)

MEMORANDUM AND OPINION

ROSENTHAL, District Judge.

This patent dispute arises from a longstanding problem of automated assembly processes: the people who operate the machines can make mistakes in loading components on the assembly machines. These mistakes are more likely, and the consequences more costly, when the components involved are tiny and visually indistinguishable. Such components increase the potential for two types of human errors: identifying what the component to be placed is and determining where to place it for automated assembly. A number of individuals and entities, including plaintiffs, have devised ways to address these problems. On August 31, 1992, plaintiffs Kermit Aguayo and Khanh Tran applied for United States Patent No. 5,283,943 (the “’943 Patent”). This patent claimed an invention for “ensuring the proper loading of assembly equipment” used in the automated assembly of multiple-component products. In 2002, plaintiffs sued Universal Instruments Corporation, alleging that two devices, the PSV — “Platform Setup Validation” — and the CVS 1 — “Component/Reel Verification System” — infringed most of the claims of the ’943 Patent. Universal designed and marketed the CVS + and the PSV as options for use on Universal’s general surface mount— “GSM” — automated assembly machine.

At trial, Universal did not dispute that the CVS + , which was discontinued in 1996, infringed claims of the ’943 Patent. Universal sold only two CVS+ units before introducing the successor device in 1997, the PSV. Universal designed the PSV specifically to avoid infringing the ’943 Patent and throughout this litigation vigorously denied any infringement. Universal also challenged the validity of the ’943 Patent, arguing that the ’943 Patent did not describe a new solution to the problem of operator error in placing components on automated assembly machines and that prior art made the ’943 Patent invalid.

This court held a hearing under Markman v. Westview Instruments, Inc., 52 F.3d 967 (Fed.Cir.1995), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996), and construed the terms that the parties submitted. After granting in part and denying in part the parties’ cross-motions for summary judgment, this court held a jury trial on the following issues: whether Universal’s PSV machine infringed the asserted claims of the ’943 Patent; whether the ’943 Patent was invalid based on three different items of prior art; what a reasonable royalty rate would be for Universal’s sales of its CVS+ and PSV products; and whether any infringement was willful. The parties timely made and renewed motions for judgment as a matter of law on infringement and validity.

In its answers to specific questions, the jury found that the PSV literally infringed the asserted claims of 'the ’943 Patent. Universal moves for judgment as a matter of law, asking this court to find nonin-fringement or, in' the alternative, order a new trial. This court grants the motion for judgment as a matter of law, finding no infringement.

The jury also found that three prior art references invalidated many of the claims of the ’943 Patent as anticipated, obvious, or both. The ’943 Patent has two independent claims, claim 1 and claim 13. The jury’s answers to the questions made all of claim l’s dependent claims invalid under 35 U.S.C. § 102(b) except for dependent claims 4 and 5. The invalidity finding was based on one of the three item's of prior art — the Philips Reel Verification System sold to Ford Motor Company (the “Philips RVS”). Although the jury found that dependent claims 4 and 5 were not anticipated by the Philips RVS, these two claims require virtually the same elements as claims 17 and 18, which depend from independent claim 13. The jury treated these two pairs of claims consistently in answering every other question. In answering the questions as to anticipation by the Philips RVS, however, the jury gave a different answer as to claims 17 and 18 than as to claims 4 and. 5. The jury found that claims 4 and 5 were not anticipated by the Philips RVS, while claims 17 and 18 were anticipated. Universal asserts that the jury’s answers are inconsistent and asks for judgment as a matter of law that claims 4 and 5 are anticipated by the Philips RVS. Plaintiffs respond that the ver.dict is not inconsistent and ask for judgment as a matter of law that no claim of the ’943 Patent is invalid as anticipated or, in the alternative, for a new trial. This court concludes that the verdict is indeed inconsistent and that as a matter of law, claims 4 and 5 are anticipated by the Philips RVS.

Universal also moves for judgment as a matter of law that the defenses of laches and equitable estoppel bar plaintiffs’ claims. This court finds that neither lach-es nor equitable estoppel are applicable.

The remaining sets of motions address damages and attorneys’ fees. At trial, the jury found a reasonable royalty rate of $2,000 for the sale of each infringing unit. Universal moves for judgment as a matter of law that plaintiffs may not collect such royalties for 65 units sold to Motorola, which were the subject of a settlement between plaintiffs and Motorola in a suit alleging that Motorola infringed the ’943 Patent. This court grants Universal’s motion. Finally, plaintiffs move to recover their fees and costs under 35 U.S.C. § 285 on the basis that the case is exceptional. Because this court found, as a matter of law, that the claims of the ’943 Patent are invalid and not infringed, this motion is not addressed.

These rulings, based on a careful review of the pleadings, the motions and responses, the record, and the applicable law, are explained in detail below.

I. Background

The ’943 Patent, issued on February 8, 1994, relates to “improving processes of automated assembly.” The “Summary of the Invention” states that “[t]he present invention contemplates a system for assembling multiple component products using automated assembly equipment,” by “ensuring the proper loading of assembly equipment.” An assembly machine operator loads components or compartments of components on the proper locations on the automated assembly machine. The machine then retrieves the components or compartments from the designated locations and places them in the desired locations on the product undergoing assembly. In the section entitled “Background of the Invention,” the ’943 Patent described the problem it was attempting to address, as follows:

Many mistakes occurring in modern automated manufacturing are caused by human error in the set up of the assembly process. Specifically, assembly operators are prone to load components in the wrong places because modern articles of manufacture have increasingly large numbers of components which are often visually similar. To prevent these set up errors, well trained operators are usually very careful and deliberate in their work. However, no degree of care can totally prevent mistakes. Furthermore, careful and deliberate behavior results in very slow set up which reduces productivity.

The aforementioned problem is particularly acute in the assembly of electronic products, where components are especially numerous and physically similar.... To compound matters, modern electronic assemblies are assembled by machines called pick and place assemblers. These assemblers are designed to retrieve components directly from industry standard reels or tubes. Therefore, the operator’s primary visual contact is with the package and not the component.

Electronic products typically require several dozen different components. In a pick and place system, each tube or reel of components must be loaded in a specific place. Given the similarity of visual appearance of the reels and tubes, it is extremely easy for an operator to load a, pick and place location with the wrong components.

The ’943 Patent disclosed a three-element system and a four-element system for “ensuring the proper .loading of assembly equipment.” Independent claim 1 recites a system “comprising” three primary elements: an “assembly machine,” a “component identifier,” and a “location indicator.” Independent claim 13, requires the same three primary elements and. adds an “information processor.” The use of the transitional word “comprising” in the preamble of a claim indicates that the scope of the claim is not limited to the recited elements. Additional elements may be present but still fall within the scope of the claim. See, e.g., AFG Indus. Inc. v. Cardinal IG Co., Inc., 239 F.3d 1239, 1244-45 (Fed.Cir.2001) (“When a claim uses an ‘open’ transition phrase, its scope may cover devices that employ additional, unrecited elements. We have consistently held that the word ‘comprising’ is an open transition phrase.”) (citations omitted).

Plaintiffs contend that the Universal PSV infringes independent claim 1 and dependent claims 3, 4, and 5, and independent claim 13 and dependent claims 14-18, 21-24, and 26-28. Independent claim 1 and dependent claims 3, 4, and 5 involve the three-element system. These claims read:

1. A system for assembling multiple component products, comprising:

an assembly machine which assembles products using components retrieved from a plurality of compartments located at a corresponding plurality of locations, each compartment located according to a type of component stored within;

a component identifier that identifies the type of component stored in each compartment by analyzing indicia of the components or component compartments; and

at least one location indicator, responsive to the component identifier, for producing an indication signal of the corresponding location of each component compartment or component after the type of component has been identified.

3. The system of claim 1, further comprising, an error indicator that indicates if a compartment has not been placed in a corresponding location indicated by the location indicator.

4. The system of claim 3, said error indicator indicating if a compartment has not been placed in said corresponding location indicated by the location indicator within a predetermined amount of time.

5. The system of claim 4, said predetermined amount of time been selected to correspond to a time required to load the compartment in said corresponding location.

The second set of allegedly infringed claims involves the four-element system. Independent claim 13 and dependent claims 14-18, 21-24, and 26-28 read as follows:

13. A system for assembling multiple component products, comprising:

a machine which assembles products using components retrieved from a plurality of compartments, each compartment being located in a corresponding one of a plurality of locations according to type of component stored within each compartment;

an information processor;

a component identifier, connected to the information processor, that identifies a type of component stored in a compartment, said information processor operating to determine a proper location of each identified compartment; and

at least one location indicator, connected to and controlled by the information processor, that indicates the proper location of each identified compartment.

14. The system of claim 13, further comprising at least one compartment sensor that senses whether a compartment is present in a previously identified location.

15. The system of claim 13, further comprising one compartment sensor per location.

16. The system of claim 13, further comprising an error indicator which indicates if a compartment has not been placed in a location indicated by the location indicator.

17. The system of claim 16, said error indicator indicating if a compartment has not been placed in a location indicated by the location indicator within a predetermined period.

18. The system of claim 17, said predetermined time period being selected to correspond to a time required to load a compartment in a corresponding location.

21. The system of claim 13, wherein the component identifier comprises an optical reader.

22. The system of claim 21, wherein the indicia of the components or compartments comprises optically readable indicia.

23. The system of claim 22, wherein the optically readable indicia comprises a bar code.

24. The system of claim 13, wherein the location indicator comprises an individual visual indicator corresponding to each location.

26.The system of claim 13, where the assembly machine comprises: a pick and place machine; and a tape and reel feeding system.

27. The system of claim 13, where the information processor comprises: a computer with memory; and a digital I/O circuit that facilitates communication between the computer and the component identifier and the location indicators.

28. The system of claim 27, further comprising a relay circuit connected to and controlled by the computer for disabling the assembly machine if all locations are not loaded with compartments containing the correct type of component.

Figure 1 of the ’943 Patent is a block diagram of the three-element and the four-element systems in an automated assembly apparatus. The figure is described as an illustration of “two embodiments comprising the minimum necessary features of the invention.”

The “Detailed Description” section of the ’943 Patent sets out the differences between the three- and four-element sys-terns claimed in the independent claims:

The three element embodiment contrasts with the four element embodiment both physically and conceptually. The four element embodiment may broadly be described as having a central information processor which is designed to accommodate all other elements of the invention. The information processor serves to prompt, interpret, compute, process and transfer information as necessary. In contrast, the three element embodiment distributes the functions of the information processor among the other elements. It is further contemplated that hybrid embodiments might also become preferable. A hybrid embodiment would involve information processing functions occurring both at the central information processor and at the peripheral elements. Such an embodiment could realize speed and efficiency advantages in complex or large assembly situations.

The “Detailed Description” of the three- and four-element systems states that both systems have the “identical” three primary parts: an assembly machine, a component identifier, and a location indicator. The four-element system adds an information processor.

The “Detailed Description” sets out the “typical” process using the three-element system, as follows:

The machine operator would begin the process of loading the assembly machine by identifying a compartment or group of components. That is to say that the machine operator uses component identifier 103 to identify the first type of component. The component type is then compared with reference information that allows for the determination of the proper location 107 for the identified components. For purposes of a three-element system, this comparison and determination may take place either in the component identifier 103 or in the position indicator 102.

The “Detailed Description” also sets out the “typical” operation of a four-element system:

[T]he machine operator uses component identifier 103 to identify the first type of component 104. The component identifier 103 then transfers the component type to the information processor 108. The component identifier ... may transfer this information without performing any computations or processing because the information processor 108 performs those tasks. The information processor ... then determines the proper location 107 of the identified components using preprogrammed information in addition to the data provided by component identifier. Next, the information processor ... informs the position indicator 102 of the position that should be indicated.... After the proper location ... has been determined it is conveyed to the machine operator.

Following a hearing, this court construed all the claim terms the parties requested under Markman. The construed terms are as follows:

A “component product” is a product assembled from component parts.

An “assembly machine” is an apparatus that assembles products by gathering components.

A “component” is a part stored in a compartment and used in the assembled product.

A “component identifier” is an optical or magnetic reader, or an equivalent thereof, which identifies, recognizes, or selects the component or component compartment being placed in a location during the assembly process.

“Indicia” is a marking.

A “location indicator” is a device that indicates a proper or corresponding location of an identified compartment or component before placement of the compartment or component in that location.

An “error indicator” is a device that indicates if a compartment has not been placed in a location, or if the compartment has not been placed in a location within a predetermined amount of time.

A “predetermined amount of time” is the amount of time determined before placement of a compartment in a location.

A “time required to load a compartment” is the time it takes to load a compartment in a corresponding location.

A “disabling device” in claims 6 and 7 (which depend on claim 1, the three-element version of the invention) is a relay, or equivalent structures, connected to the location indicator, the component identifier, and the assembly machine, which disables or enables the assembly machine in response to commands from either or both the location indicator and component identifier, depending upon the distribution of the functions of the information processor between the location indicator and the component identifier.

A “disabling device” in claims 19 and 20 (which depend on claim 13, the four-element version of the invention) is a relay, or equivalent structures, connected to the information processor and the assembly machine, which disables or enables the assembly machine in response to commands from the information processor.

An “information processor” is a computing system that processes information to determine a proper location of each identified component.

A “sensor” is a device that senses the presence of a compartment.

An “individual visual indicator” is a device that provides a visual indication corresponding to each location.

A “light emitting diode” is a semiconductor device that emits optical radiation when biased in the forward direction.

A “digital I/O circuit” is a digital input/output circuit.

A “relay circuit” is an electric device connected to and controlled by a computer or other information processor to cause contact operation in associated electric control circuits.

After the Markman ruling, the parties moved for summary judgment as to both infringement and validity. This court granted the motions in part and denied them in part. The parties tried the issues of infringement, validity, reasonable royalty, and willfulness to a twelve-person jury. In preliminarily instructing the jury before the parties presented evidence, this court added two points to the claim construction order. The first point clarified that as to a location indicator, “the proper or corresponding location for a particular component must be determined after the component is identified by the component identifier and before the location indicator indicates that location for that component.” The second point made it clear that a “corresponding location” is the correct or proper location in which to place the identified components or component compartments.

The evidence at trial described the work leading up to the patent and the patent claims. Both inventors testified, as did representatives of the company where they worked when they developed the invention. The testimony revealed that Aguayo, a process engineer, and Tran, a software engineer, developed the invention while working in quality control at a company called XeTel in Austin, Texas. XeTel was in the business of manufacturing printed electronic circuit boards for computers. In the 1980s, many assembly operations were using bar codes to identify component parts contained in reels. It was also common for assembly machine operators to use a bar code reader to identify a component part number and to consult a paper listing component part numbers and their corresponding locations on the assembly machine to determine where to place a component or reel of components on the assembly machine.

Aguayo and Tran attempted to reduce human errors in placing components on assembly machines during the production of printed circuit boards. In 1989, Aguayo and Tran built a prototype of an invention designed to give increased guidance to assembly machine operators in placing components on assembly machines. The prototype, known as the XeScan, used a bar code reader to read the label on a reel of components. This step avoided reliance on the human operator to figure out what a particular component or reel of components was. The XeScan then determined the proper location for. that scanned component and displayed the number of that slot location on the monitor. This step avoided reliance on the operator’s comparison of a component number with a preset reference list of component numbers and corresponding locations. The XeScan used a bar code reader to identify a component or a reel containing components, compared the component identification information provided by the bar code scan against a preprogrammed list of components and their proper locations, and displayed the number of the slot location where each identified component had to be placed on the assembly machine. Plaintiffs’ witnesses at trial described the XeS-can as a “preferred embodiment” of what became the ’948 Patent. XeTel determined not to pursue a patent on the XeS-can, but allowed Aguayo and Tran to do so.

Plaintiffs sought to patent both the automated assembly apparatus and the method for its use. The patent examiner required plaintiffs to elect between the apparatus and method claims. The examiner stated:

Inventions I and II are related to apparatus and process of use. The inventions can be shown to be distinct if either or both of the following can be shown: (1) the process for using the apparatus as claimed can be practiced with another materially different apparatus or (2) the apparatus as claimed can be used in a materially different process of using that apparatus. In the instant case the apparatus as claimed can be used in a materially different process of using that apparatus such as simply identifying components.

The patent issued on February 8, 1994. After the patent issued, plaintiffs unsuccessfully sought commercial applications. Although XeTel became bankrupt in 2001, plaintiffs followed technological developments in pick-and-place assembly machines, attending trade shows and reading industry publications. The parties dispute whether plaintiffs met their responsibility as patent holders by waiting until 2002 to notify Universal of their belief that the PSV infringed the ’943 Patent.

The evidence at trial described' the design and operation of Universal’s CVS + system and two versions of its PSV system. In 1990, Ford . Motor Company had purchased a reel verification system from Philips. In 1992, Ford asked Universal to design a reel verification system based on the Philips RVS. Universal’s response was the CVS + . Testimony at trial described in detail both the Philips RVS, which Universal asserts is prior art that invalidates the ’943 Patent, and the CVS +. Both the Philips RVS and the CVS + used a bar code scanner to identify a component or combination of components; used a computer to compare the identified components to preprogrammed information listing the proper slot location for each component or combination of components to determine the proper slot location for that component; and showed the operator the number for that slot location. Universal designed the CVS+ to operate like the Philips RVS. The CVS+ itself compared the bar code scan data with a prepro-grammed list of component numbers and slot locations and told the operator the number of the slot location where the component was to be placed.

On May 27, 1995, plaintiffs notified Universal by letter that the CVS+ infringed the ’943 Patent. Universal responded by a letter from outside patent counsel dated June 1, 1995. In the letter, Universal explained why it believed the CVS+ did not infringe. That belief was based on an erroneous and incomplete understanding of how the CVS+ functioned. At trial, Universal acknowledged that the CVS + did infringe the ’943 Patent claims. Although plaintiffs disagreed with the nonin-fringement conclusion stated in the 1995 letter from Universal, they took no further action until 2002.

Universal discontinued making and selling the CVS + in 1996, for reasons unrelated to the ’943 Patent. In 1997 Universal replaced the CVS + with the PSV. In 1996, during the development of the PSV design, Universal became aware that its planned design infringed the ’943 Patent. Universal changed the PSV design specifically to “design around” the ’943 Patent, by removing the step in which the system itself compared the component identification number obtained from the bar code scan data with the component identification numbers listed in the preprogrammed setup table, determined the slot where that component belonged, and displayed the number of that slot on the computer monitor.

Plaintiffs’ witnesses testified as to how the PSV could operate. Before the operator begins to load any components or compartments, the operator enters a pre-programmed setup list or table of the components needed and the particular slot locations for each of these components. There may be a large number of slot locations to be loaded with components, up to seventy-two. The operator uses a bar code scanner to “read” the label on a component or reel of components. The bar code labels on the reels contain an alphanumeric (and perhaps English) description of what component type it contains. When an operator scans a bar code label on a reel of components, the scanner sends the data obtained from the bar code scanner to a computer, which derives the component identification number from the data and stores that number.

Plaintiffs’ witnesses testified at trial that the setup information for a product may refer to a particular component or reel of components by an identification number that is different from the identification number used by the vendor of that component or reel. Defendants’ witnesses disputed that this occurred often, but agreed that it could happen. In that event, a software process turns the bar code data into a component identification number that matches the format of the identification numbers in the preprogrammed setup list. This process involves removing unnecessary characters and matching the scanned component number to the stored component identification number in the setup list. The PSV then displays that component identification number on a computer monitor. The monitor also displays the preprogrammed setup list or table of the component identification numbers and the proper slots for each.

All the witnesses agreed that the PSV system does not perform any comparison between the component identification number and the preprogrammed setup list or table. All the witnesses agreed that the PSV computer monitor does not directly or specifically tell the human operator the particular slot number for the identified component. Instead, the operator must scroll and read down the preprogrammed setup list to find the component identification number that matches the component identification number produced after the barcode scan. The operator must then read across the setup list to find the corresponding slot location number. As noted, the preprogrammed setup list or table of the identification numbers of the components to be placed and the corresponding slot numbers for each of those components can have as many as seventy-two entries.

Universal marketed the PSV in industry publications and at trade shows. Aguayo was familiar with and read the publications and attended some of the trade shows, but did not specifically recall learning about the PSV until 2002. In April 2002, Aguayo attended a trade show and visited Universal’s display booth. Aguayo testified that he saw a Universal device with light emitting devices (“LEDs”) at each slot location. He believed that the PSV had this feature, which was one reason he thought the PSV infringed the ’943 Patent. After the trade show, Aguayo and Tran sued, alleging willful infringement. Aguayo was incorrect: Universal does not offer a device that uses LEDs and the PSV did not have them.

At trial, plaintiffs argued that the PSV infringes the ’943 Patent because of the simultaneous display on the computer monitor of the component identification number and the preprogrammed setup table of component identification numbers and corresponding slot location numbers. Plaintiffs offered two experts — Dr. Thomas Rhyne, an electrical engineer, and J. Tipton Cole, a software programmer — to compare the claims of the ’943 Patent with the PSV. Both experts concluded that infringement is present because the PSV has an assembly machine (the GSM); a component identifier (the bar code scanner); an information processor (the embedded computer that translates the bar code scan data into a component identification number and stores that information); and a location indicator (the computer screen that simultaneously displays the component identification number and the prepro-grammed list of the component identification numbers and their corresponding slot numbers). Rhyne and Cole testified that Universal’s PSV system uses a bar code scanner to “read” a label on a component reel; uses an information processor to “translate” the bar code data into a component identification number that corresponds to the format of the component identification numbers in the prepro-grammed setup list; and displays the translated component identification number on a computer screen that also displays the previously-stored setup list of the numbers of the components to be placed and the corresponding number of the slot location where each of those components was to be placed. According to plaintiffs’ witnesses, the PSV infringes the ’943 Patent because the computer monitor’s simultaneous display of the component identification number and the pre-programmed setup table of component identification numbers and corresponding slot locations is a “location indicator.” Plaintiffs’ experts agree that the computer monitor provides a list that an operator must scroll through to find the component identification number matching the simultaneously displayed identification number for the just-scanned component, and read across the table to find the proper slot location for that component.

Universal did not dispute how the PSV could operate, but did dispute how the PSV was intended to operate. Universal argued that the PSV was intended to be a postplacement verification system, not a preplacement guidance system. Universal presented witnesses and manuals to show that the PSV was intended to have the operator first consult the preprogrammed setup list on the computer screen to locate a particular component and identify the slot in which it belonged; partially place the component in that slot; and then scan the component using the bar code scanner. The PSV shows the component number on the computer monitor. The operator fully mounts the component in the slot. If the right component is placed in the proper slot, the color of the field on the computer monitor changes to green and the feeder is “enabled.” If the incorrect component is mounted in the slot, the field turns red. Universal argued that the intended function of the PSV as a postplacement verification, in which the scan takes place after the operator identifies the correct slot location and partially inserts the component into that slot, does not infringe the ’943 Patent, which claims a preplacement location indicator system. Universal’s witnesses agreed that the PSV could simultaneously display on the computer screen the component identification number and the preprogrammed setup list or table, before the operator fully mounts the scanned component into a particular slot on the assembly machine.

Universal also argued that whether the PSV was used to guide an operator in the initial decision of where to load a component reel, or to confirm to the operator that an initial, tentative decision as to the proper placement was correct, it did not infringe the ’943 Patent. Universal argued that the ’943 Patent claims a system in which the invented device both identifies the component for the human operator and tells the operator where to place that component. The PSV, by contrast, was deliberately designed not to provide the second piece of information. The PSV performs no comparison or analysis of the information obtained by the bar code scan after whatever manipulation or translation may be required to produce a component identification number in the proper format. The PSV does not provide any specific indication to the operator of the proper location for an identified component. Instead, the PSV simultaneously displays the component identification number and a list or table of component identification numbers and their corresponding slot locations. The operator must scroll through the preprogrammed setup list, compare the component identification number provided after the scan with the preprogrammed setup list of component identification numbers, find the matching component number, and then read across the setup list to learn the proper slot location for that component. The ’943 Patent, according to Universal, claims a device that not only provides a component identification number, but also determines where that identified component will go and indicates that location to the operator. Universal argues that the device claimed under the ’943 Patent — not the human operator — must determine the proper location for each identified component after that component is identified. In the four-element system, that function is performed by the information processor. In the three-element system, that function is distributed among the three elements. Universal contends that it is reading the claims in light of the specification; plaintiffs contend that Universal is importing limitations from the specification not present in the claims.

Universal also defended the case on invalidity grounds. The critical date for the ’943 Patent is August 31, 1991. Universal presented evidence of three allegedly invalidating prior art sources. The first is an article written by Timothy Mennitt dated April 1991 and titled “Feeder Identification Assembly.” The article was featured in Motorola’s Technical Developments publication and mailed to over 200 recipients. In the article, Mennitt describes the Fuji Reel Verification System, an automated assembly machine in commercial use, and proposes a solution to improve the use of such machines. The second prior art source is a Japanese patent application, referred to as the Oki patent application, dated November 6, 1989 and published on June 25, 1991. The Oki patent application claims an automated assembly machine used to mount parts on printed circuit boards. The third source of prior art is the Philips Reel Verification System (“RVS”), an automated assembly machine developed by Philips and sold to Ford in July 1991. Universal presented documents describing the RVS and presented testimony from employees from both Philips and Ford involved in the development, sale, and use of the RVS. Edison Hudson, an engineer in the field of automated assembly machines, gave expert testimony on behalf of Universal as to how the Philips RVS operated and compared the RVS to each allegedly infringed claim of the ’943 Patent.

Plaintiffs presented testimony from Rhyne disputing that the Philips RVS anticipated the ’943 Patent claims or made them obvious. Plaintiffs also challenged aspects of each of the items of prior art, asserting that the Mennitt article was not enabling; that the Oki patent application was not enabling and did not disclose all of the elements in the ’943 Patent claims; and that the Philips RVS was not offered for sale and was not ready for patenting.

At the close of the evidence, Universal moved under Rule 50(a) of the Federal Rules of Civil Procedure for judgment as a matter of law that the asserted claims of the ’943 Patent are invalid because they are obvious or anticipated. Plaintiffs cross-moved for judgment as a matter of law that the prior art did not anticipate any of the ’943 Patent claims or make them obvious. Universal also moved for judgment as a matter of law that the PSV does not infringe any of the asserted claims of the ’943 Patent.

At the close of the evidence, plaintiffs asserted infringement under the doctrine of equivalents only as to dependent claim 28. Universal moved for judgment as a matter of law that the PSV did not infringe claim 28 under the doctrine of equivalents. This court granted Universal’s motion for judgment as a matter of law that there was no infringement under the doctrine of equivalents because plaintiffs had failed to provide sufficient proof as to the function, way, and result tests of the doctrine of equivalents. The expert testimony did not specifically compare the function, way, and result of the relay circuit limitation found in claim 28 with that of the PSV. This court found that without such testimony or other evidence, plaintiffs’ proof under the doctrine of equivalents was insufficient as a matter of law. Lear Siegler, Inc. v. Sealy Mattress Co. of Mich., Inc., 873 F.2d 1422, 1427 (Fed.Cir.1989).

The jury returned a verdict in plaintiffs’ favor on the literal infringement issue as to all the asserted claims of the ’943 Patent. In answering the questions on validity, the jury found that all the claims except claims 4 and 5 were invalid because they were either anticipated or obvious under the prior art. In response to Question Number 5, the jury found that the Philips RVS sold to Ford anticipated claims 1, 3, 13-23, and 26-28. The jury found that claims 4, 5, and 24 were not anticipated by the RVS. In response to separate questions, the jury found that claim 24 was anticipated by the Oki patent application and obvious in light of the prior art. In its posttrial renewed motions for judgment as a matter of law, Universal claims that the jury verdict is inconsistent. The jury found that claims 4 and 5 were not anticipated by the Philips RVS but found that claims 17 and 18 were anticipated. Universal argues that the two sets of claims have the same elements, making the verdict inconsistent. Universal moves for judgment as a matter of law that claims 4 and 5 are anticipated by the Philips RVS or, in the alternative, for a new trial. Universal also seeks judgment as a matter of law that all the asserted claims of the ’943 Patent are invalidated by the Philips RVS. Plaintiffs respond that the seemingly inconsistent verdict can be reconciled and, if not, should be the basis for a grant of a new trial. Plaintiffs renew their motion for judgment as a matter of law under Rule 50(b) that none of the ’943 Patent claims are anticipated or obvious, and, in the alternative, move for a new trial. Universal also renews its motion for judgment as a matter of law that the PSV system does not literally infringe the asserted claims of the ’943 Patent. Universal also moves for judgment as a matter of law that if the ’943 Patent does infringe, it is invalid for lack of a written description.

Each of the arguments as to the judgment that should be entered is examined below.

II. The Standard for Judgment as a Matter of Law

“The grant or denial of a motion for judgment as a matter of law is a procedural issue not unique to patent law, reviewed under the law of the regional circuit in which the appeal from the district court would usually lie.” Summit Tech., Inc. v. Nidek Co., 363 F.3d 1219, 1223 (Fed.Cir.2004). A motion for judgment as a matter of law is appropriate when, on reviewing the entire record, there is no legally sufficient evidentiary basis for a reasonable jury to find for the "nonmoving party on an issue. DP Solutions, Inc. v. Rollins, Inc., 353 F.3d 421, 427 (5th Cir.2003). In evaluating the record, the court must make all reasonable inferences for the nonmoving party. Id; Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000). Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge. DP Solutions, 353 F.3d at 427 (citing Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986)). In considering a Rule 50 motion for judgment as a matter of law following a jury verdict, the court must be “especially deferential” to the jury’s findings. Brown v. Bryan County, OK, 219 F.3d 450, 456 (5th Cir.2000). The Fifth Circuit’s standard for reviewing a jury verdict is whether “the state of proof is such that reasonable and impartial minds could reach the conclusion the jury expressed in its verdict.” Am. Home Assur. Co. v. United Space Alliance, 378 F.3d 482, 487 (5th Cir.2004) (quoting Liberty Mut. Ins. Co. v. Falgoust, 386 F.2d 248, 253 (5th Cir.1967)). A jury verdict must stand unless there is a lack of substantial evidence, viewed in the light most favorable to the successful party, to support the jury’s factual findings, or the legal conclusions implied from the jury’s verdict cannot, in law, be supported by those findings. Am. Home, 378 F.3d at 487.

“Judgment as a matter of law of no literal infringement is appropriate if .no reasonable fact finder could determine that the accused devices meet every limitation of the properly construed claims.” Elkay Mfg. v. EBCO Mfg., 192 F.3d 973, 980 (Fed.Cir.1999). Anticipation is a question of fact, and a jury determination of anticipation is reviewed for sufficient evidence. Minn. Mining & Mfg. Co. v. Chemque, Inc., 303 F.3d 1294, 1301 (Fed.Cir.2002). Obviousness is a legal question based on factual determinations. “In review of a jury verdict on the ground of obviousness, the underlying findings of fact, whether explicit or presumed as necessary to support the verdict, are reviewed for substantial evidentiary support; and the ultimate question of obviousness is reviewed for correctness in law, based on the factual premises.” Cardiac Pacemakers, Inc. v. St. Jude Med., Inc., 381 F.3d 1371, 1375 (Fed.Cir.2004). “When presented with patent validity and infringement issues, trial courts should ... decide both.” Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1540 (Fed.Cir.1983).

III. The Motions for Judgment as a Matter of Law as to Literal Infringement

Plaintiffs assert-that the PSV infringed both independent claim 1 and independent claim 13. Because the PSV consists of an assembly machine, a bar' code scanner, a computer monitor, and a computer, the bulk of the plaintiffs’ testimony and analysis focused on claim 13, the claimed four-element system. According to plaintiffs, the - bar code scanner is the component identifier; the embedded computer monitor is the location indicator; and the computer is the information processor. Universal disputes that the PSV has a location indicator or an information processor as claimed in' claim 13. As to claim 1, which does not require an information processor as a separate element, Universal disputes that the PSV has a location indicator as required in claim. 1. Universal generally argues that because the ’943 Patent was intended to avoid reliance on human operators to read identification numbers on components and match those numbers to numbers on lists and tables that showed corresponding slot locations, and because ■ the PSV requires human operators rather than the device .to perform the, second function, the PSV does not infringe.

A. The Applicable Legal Standard

Patent infringement claims involve two analytic steps. Scanner Tech. Corp. v. ICOS Vision Sys. Corp., N.V., 365 F.3d 1299, 1302 (Fed.Cir.2004); Novartis Pharm. Corp. v. Eon Labs Mfg., Inc., 363 F.3d 1306, 1308 (Fed.Cir.2004). First, the court determines the meaning and scope of the asserted claims. Scanner Tech., 365 F.3d at 1302; Novartis Pharm., 363 F.3d at 1308. Claim construction is a matter of law. Bai v. L & L Wings, Inc., 160 F.Sd 1350, 1353 (Fed.Cir.1998). Second, the claims as construed are compared to the allegedly infringing device to determine whether the claims encompass the accused structure. Scanner Tech., 365 F.3d at 1302; Novartis Pharm., 363 F.3d at 1308. Whether the accused device contains each element exactly, as properly construed, is a question of fact. Bai, 160 F.3d at 1353. Literal infringement of a claim requires that every limitation recited in the claim appears in the accused device, “i.e., that the properly construed claim reads on the accused device exactly.” Cortland Line Co., Inc. v. Orvis Co., Inc., 203 F.3d 1351, 1358 (Fed.Cir.2000). If even one limitation is missing or not met as claimed, there is no literal infringement. Id. at 1358.

The parties accuse each other of arguing for claim constructions inconsistent with the Federal Circuit’s approved approach. Plaintiffs accuse Universal of importing limits from the specification to cabin the claims to preferred embodiments. Universal accuses plaintiffs of ignoring the consistent and repeated statements in the specification in presenting the claims. After this court issued its Mark-man opinion, and very shortly before trial in this case began, the Federal Circuit granted an en banc rehearing in a case, to address the law of claim construction. In Phillips v. AWH Corp., 376 F.3d 1382 (Fed.Cir.2004), the court intends to clarify seven issues, including the following three: (1) Is the public notice function of patent claims better served by referencing primarily to technical and general purpose dictionaries and similar sources to interpret a claim term or by looking to the patentee’s use of the term in the specification? If both sources are to be consulted, in what order? (2) If dictionaries should serve as the primary source for claim interpretation, should the specification limit the full scope of claim language (as defined by the dictionaries) only when the patentee has acted as his own lexicographer or when the specification reflects a clear disclaimer of the claim scope? If so, what language in the specification will satisfy those conditions? What use should be made of general as opposed to technical dictionaries? How does the concept of ordinary meaning apply if there are multiple dictionary definitions of the same term? If the dictionary provides multiple potentially applicable definitions for a term, is it appropriate to look to the specification to determine what definition or definitions should apply? (3) If the primary source for claim construction should be the specification, what use should be made of dictionaries? Should the range of the ordinary meaning of claim language be limited to the scope of the invention disclosed in the specification, for example, when only a single embodiment is disclosed and no other indications of breadth are disclosed? Id. at 1383.

Although the forthcoming en banc decision will no doubt clarify how specifications and sources such as dictionaries are to be used in claim construction, the infringement analysis at issue does not depend on the outcome. To the extent the parties ask this court to revisit the construction of disputed claim terms and to compare the construed claims to the accused PSV, this court need not await the resolution of the precise relationship among ordinary and customary meaning, dictionary definitions, and the intrinsic record, because even under plaintiffs’ preferred methodology and claim constructions, the record reveals no literal infringement, as a matter of law.

B. The Information Processor: Claim 13

Claim 13, the four-element system, requires the presence of an information processor “operating to determine a proper location of each identified component.” In its opinion issued under Markman, this court adopted plaintiffs’ proposed construction of “information processor” as “a computing system that processes information to determine a proper location of each identified component.” This construction rejected Universal’s argument, renewed in its requested jury instructions and in its motion for judgment as a matter of law, that construing claim 13 in light of the consistent and exclusive description in the specification results in the following language: an information processor that performs the function of determining the proper location of the identified components by comparing the identified component to preprogrammed reference information that identifies the proper location. After hearing arguments from the parties, this court declined to instruct the jury that the information processor “operate[s] to determine” the location of the identified components only by comparing the component identification information against the prepro-grammed setup list or table. This court instructed the jury that if the instructions did not provide a particular definition for a term used in the patent claims, those terms were to be given their ordinary meaning.

In its motion for judgment as a matter of law, Universal seeks the following “implicit definition” of “operating to determine a proper location,” derived from the ’943 Patent specification:

performing operations to find out or come to a decision about the proper location for an identified component after the component is identified, by comparing the identified component to pre-programmed reference information that identifies the proper location.

Plaintiffs assert that the proposed revised claim construction is both too late and too limiting. Universal is correct that the only way the specification describes the four-element system is that after the component is scanned, the information processor compares the component identification number with the stored setup list in order to figure out the slot location of the just-scanned component. The testimony at trial confirmed that this was the only description of the operation of the claimed systems, providing a strong argument in favor of Universal’s proposed construction. See Bell Atl. Network Servs. Inc. v. Covad Communications, Group, Inc., 262 F.3d 1258 (Fed.Cir.2001) (finding that the term “mode” was defined “by implication” based on its consistent use throughout the specification to mean only the three operational “modes” in the specification). Plaintiffs are correct, however, that there is no language in claim 13 that requires comparison of the preprogrammed reference information — the setup list — with the component identification number as the only method for determining the proper location of the identified components. Acknowledging the fine line between reading the claim in light of the specification and reading a limitation into the claim from the specification, this court declines to revise the construction of “information processor” as Universal seeks.

In the claim construction briefing, the parties did not seek, and this court did not provide, a construction of “determine” as used in the claim 13 element of the “information processor.” At trial, the parties urged different “dictionary” meanings of this term. Universal contended that “determine” meant to “find out or come to a decision about by investigation, reasoning, or calculation.” Plaintiffs contended that “determine” meant to “resolve,” “fix conclusively,” or “remove doubt.” Regardless of which of these dictionary definitions is used, and applying the construction of “information processor” that plaintiffs themselves sought, this court concludes that as a matter of law, the PSV does not contain the information processor element of claim 13 and therefore does not infringe.

Universal changed the original design of the PSV specifically to “design around” the ’943 Patent by removing the step in which the device itself determines the number of the specific slot in which an identified component belongs and provides that number to the operator. The embedded computer in the PSV determines the component identification number, but does not do any calculation or analysis that determines the proper location of that identified component. As plaintiffs’ witnesses repeatedly acknowledged, the PSV embedded computer does not do anything with the component identification number once it is derived from the bar code scan data, except to store it and simultaneously display it on a computer screen with the preprogrammed list of the component identification numbers and their corresponding slot locations.

In the earlier version of the PSV, the OS2 version, a computer monitor displayed a validation status screen showing a preprogrammed table of component identification numbers and corresponding slot locations. The table could consist of as many as seventy-two components numbers and slot locations, although only four were visible on the computer monitor at any one time. After the bar code scan produced a number for the scanned component, the operator then scrolled down the preprogrammed setup table to find the same component identification number on the preprogrammed table, then read across the table to find the slot location for that component. In the later version of the PSV, the NT version, a validation status window showed the setup list or table before a component reel was scanned. Another window showed the bar code scan activity. A field labeled “scanned id.” displayed the component identification number after the operator scanned the component. The validation status window simultaneously displayed the preprogrammed setup list or table. To use the simultaneous display of information to learn the proper slot location for the identified component, the operator had to scroll through the setup list, find a component identification number in the list that matched the component identification number derived from the bar code scan data, then read across the table to find the number of the proper slot to place the component.

Plaintiffs’ witnesses testified at trial that the embedded computer in the PSV is the “information processor,” as called for in claim 13. Rhyne and Cole testified that this computer receives data as to the component type from the bar code scanner. The bar code scanner, according to plaintiffs, satisfies the “component identifier” element of claim 13. The bar code scanner then sends the data to the computer. The computer manipulates or translates the data to produce a component identification number, which is stored, as is the preprogrammed setup list, and displayed on the computer monitor. Rhyne and Cole explained that the PSV “operates to determine” the proper location of the identified component by having the bar code scanner — the “component identifier”obtain data about the component type; having the embedded computer — the “information processor” — translate the bar code scan data into a component identification number; then having the computer monitor— the “location indicator” — simultaneously display the component identification number with the preprogrammed setup list, which the operator can read, scroll through, match numbers, and read across, to learn the proper location for that component. According to Rhyne’s and Cole’s testimony, the embedded computer in the PSV performs no operation, calculation, or analysis other than translating the bar code scan data into a component identification number.

The embedded computer in the PSV “operates to determine” — whether defined as “figures out,” “calculates,” “resolves,” or “ends uncertainty about” — what the component identification number is, not where the identified component should be placed. The “information processor” in claim 13 must, according to plaintiffs’ own construction, “operate to determine” “a proper location of each identified component.” The embedded computer in the PSV does not meet this claim limitation. According to plaintiffs’ own witnesses, the embedded computer in the PSV merely refines or manipulates the bar code scan data about what the component is to produce a component identification number. The computer in the PSV does not obtain or provide any information as to where that identified component belongs. Nor does the computer in the PSV resolve uncertainty. about where that identified component belongs. According to Rhyne and Cole, it is the computer monitor — the location indicator — that performs this claimed function by simultaneously displaying the component identification number produced by the embedded computer with the pre-programmed setup list. The embedded computer in the PSV does nothing more than translate or manipulate bar code scan data to obtain a component identification number. It does not “operate to determine” a proper location for an identified component, as required for an information processor under claim 13 of the ’943 Patent.

Plaintiffs’ witnesses opined that the process of translating the bar code data into a component identification number that corresponds to the preprogrammed setup list in the PSV is the same thing as “operating to determine a proper location.” Plaintiffs’ witnesses testified that the PSV “operates to determine” a proper location for a component when it “translates” the bar code data for that component into a component identification number. Cole described one element of this step as the “alternate entry lookup” — a comparison of the scanned component identification number with the stored setup list of component numbers. The specification describes this process of taking information from the bar code scanner, transferring it to a computer, and comparing it against a “list of parts for the current product of manufacture,” a “pre-stored index of valid part numbers,” as component identification. In this portion of the specification, as in all other parts of the ’943 Patent, determining the component identification value or number is different from determining the proper location for the identified component. Throughout the ’943 Patent, and in claim 13 itself, it is clear that after a component is identified, the information processor then determines the correct location of the identified component, so that the location or position indicator may convey that information to the human operator. The embedded computer in the PSV uses pre-programmed information as to valid part numbers, in addition to the bar scan data, only to determine what the component is, not where it goes. The embedded computer in the PSV does nothing in addition to, or after, “translating” the bar code scan data into a component identification number that is stored. By contrast, the information processor required in claim 13 must “operate to determine a proper location of each identified component.” The ’943 Patent is clear that information as to what a component is differs from information about where it belongs. The embedded computer in the PSV “translates” the bar code scan data into a component identification number if the vendor’s label does not match the format used in the assembly process. This “translation” does not “determine” the proper location of that component; instead, it “determines” the component identification number for that component.

The language of claim 13, construed as plaintiffs themselves sought, consistent with the specification and the different dictionary definitions that could apply, does not all