One of the claims in Ex parte Fischer was directed to a plasma deposition apparatus, and included "a controller ... adapted for changing a sense of direction of said flexible substrate in said chamber ..." The Examiner asserted that the "adapted for" language was intended use, and found that the controller in the reference would have been capable of performing the function. On appeal, the Applicant argued that "adapted for" was not intended use, and that the reference must teach a controller that is programmed to support the claimed function. The Board agreed. The Board looked to the Applicant's specification and found it described a controller that was programmed to control the valves in relation to the substrate direction. The claimed controller thus becomes "a special purpose computer structured via programming to perform the control ... recited in the claims" (citing In re Alappat). Comparing that controller to the prior art controller, the Board explained that the "capable of" test was a much higher bar than that used by the Examiner:
While it might be possible to install software that would allow Lu's controller to control the valve opening based on the sense of direction of the substrate, the "capable of" test requires that the prior art structure be capable of performing the function without further programming. Typhoon Touch Techs., Inc. v. Dell, Inc., 659 F.3d 1376, 1380 (Fed. Cir. 2011) (discussing Microprocessor Enhancement Corp. v. Texas Instruments, Inc., 520 F.3d 1367 (Fed. Cir. 2008)). When the functional language is associated with programming or some other structure required to perform the function, that programming or structure must be present in order to meet the claim limitation. Id. While in some circumstances generic structural disclosures may be sufficient to meet the requirements of a "controller", see Ergo Licensing, LLC v. CareFusion 303, Inc., 673 F.3d 1361, 1364 (Fed. Cir. 2012) [*22] (citing Telcordia Techs., Inc. v. Cisco Sys., Inc., 612 F.3d 1365, 1376-77 (Fed. Cir. 2010)), that is not the case here.
Ex parte Stengelin included claims to a fuel cell system, including a cell voltage monitoring sub-system "configured to" perform various recited functions such as "determine if a wire connection has failed by measuring the voltage of each fuel cell in the fuel cell stack". The Examiner dismissed the functional language as not adding any structure to the monitoring sub-system, finding that neither the specification nor the claims themselves explain "what in the fuel system actually performs said function." The Board found error in the Examiner's application of the "capable of" rule.
The Board first noted that the Applicant's specification described an algorithm for determining whether various components in the fuel system have failed. The Board then explained the proper standard for determining whether a prior art apparatus is "capable of" performing a claimed function.
Although it is well established that claims directed to an apparatus must be distinguished from the prior art in terms of structure rather than function, see, e.g., In re Schreiber, 128 F.3d 1473, 1477-78 (Fed. Cir. 1997), in order to satisfy the functional limitations in an apparatus claim, however, the prior art apparatus must be capable of performing the claimed function. Id. at 1478. As such, to establish such capability recited in claim 1, the Examiner must demonstrate that the prior art cell voltage monitoring sub-system possesses the necessary structure, hardware or software, for example, the programming, to function as claimed. The prior art structure must be capable of performing the function without further programming. Typhoon Touch Techs., Inc. v. Dell, Inc., 659 F.3d 1376, 1380 (Fed. Cir. 2011) (discussing Microprocessor Enhancement Corp. v. Texas Instruments, Inc., 520 F.3d 1367 (Fed.Cir.2008)). When the functional language is associated with programming or some other structure required to perform the function, that programming or structure must be present in order to meet the claim limitation. Id.The Board then reversed, because the Examiner admitted that "neither reference is programmed or otherwise structured to function as claimed," nor did the Examiner "rely upon any reasoning why one of ordinary skill in the art would have configured the cell voltage monitoring subsystem to function as
The claims in Ex parte Wheat were also directed to a fuel cell system, including "a controller that controls said hydrogen supply and said air supply to power said heater to warm said fuel cell stack and said water supply when said vehicle is not running" (emphasis added). The reference taught using a programmable logic controller to maintain the temperature of the fuel cell at its optimum operating temperature while the vehicle is running. Examiner took the position that the controller in the reference nonetheless satisfied the limitation at issue, because the structure of the controller did not depend on whether the vehicle is running. The Board found error in the Examiner's position.
The Board discussed the line of case law that requires an apparatus to be distinguishable on structure, then explained how this case law is applied to controllers, where algorithms define structure:
The concept behind the “capable of” test is that an apparatus that is inherently capable of performing the function recited in the claim is an apparatus no different in structure from the apparatus of the claim. For instance, a parachute defined as opening in a certain functional way, does not have a different structure than one not disclosed as so opening if the prior art parachute is inherently capable of opening as claimed. See In re Ludtke, 441 F.2d 660, 664 (CCPA 1971). A funnel-like structure that functions to control the flow of oil is really no different in structure from a funnel-like structure defined as shaped to dispense popcorn kernels if the oil funnel is capable of dispensing popcorn in the same way. See In re Schreiber, 128 F.3d 1473, 1478 (Fed. Cir. 1997). Defining a structure in terms of its function does not necessarily distinguish it structurally from a prior art structure capable of performing the function.
However, the situation is more complicated when what is being defined by the claim limitation is a controller. The function of a controller serves to define structure by defining software and/or hardware for performing the function.
Therefore, to be "capable of" performing the recited functions, the prior art controller must "contain some mechanism or software program for performing the required control". Since it did not, the Board reversed. Moreover, the Board explained that the “capable of” test requires that the prior art structure be capable of performing the function without further programming, so it is irrelevant to anticipation that "it might be possible to install software that would allow the prior art controller to warm the fuel cell stack and the water supply when the vehicle is not running."
The claims in Ex parte Kamihara were also directed to a fuel system, including "a control unit configured to control the electric power or electric current extracted from the fuel cell stack in accordance withas not adding any structure." The Examiner dismissed the "configured to" phrase as not adding any structure, asserting that “nothing in the disclosure . . . is drawn to being programmed or having programs” and that “only a general purpose computer is supported by the disclosure.” On appeal, the Applicant argued that the claimed control unit was a special program computer “programmed or otherwise configured to perform” the claimed functions. The Board reversed the rejection, citing In re Schreiber and using the same reasoning found in Ex parte Wheat:
Although it is well established that claims directed to an apparatus must be distinguished from the prior art in terms of structure rather than function, see, e.g., In re Schreiber, 128 F.3d 1473, 1477-78 (Fed. Cir. 1997), in order to satisfy the functional limitations in an apparatus claim, however, the prior art apparatus must be capable of performing the claimed function. Id. at 1478. As such, to be capable of performing the functional limitations in claim 1, the control unit must possess the necessary structure, hardware or software, for example, the programming, to function as claimed. The Examiner indeed de facto admits that neither reference nor their combination is programmed or otherwise structured to function as claimed, nor does the Examiner rely upon any reasoning why one of ordinary skill in the art would have configured the control unit to function as claimed.
The Examiner in Ex parte Dewey was similarly reversed. The claims were, once again, directed to a fuel cell system, and included "a controller for controlling the amount of heat provided by the heater ... " The Examiner asserted that the functional language did not distinguish over the prior art apparatus since the reference taught "a fuel cell stack, a heater circuit including a heater and a controller, which are the same structural features as those claimed by Applicant." The Board reversed, explaining as follows:
However, contrary to the Examiner's apparent belief, the mere fact that Breault's fuel cell system includes a controller is not an acceptable reason to believe that the controller inherently possesses the capability of performing the function required by Appellant's claim 1 controller. For the reasons detailed in the Appeal Brief and Reply Brief, the claim 1 controller is patentably distinct from a controller which is not programmed or otherwise designed to perform the claim 1 function. (Internal citations omitted.)
Similarly, as correctly argued by Appellant, the record provides no acceptable reason for the Examiner's above presumption "that [in Breault's system] the current and voltage flow through the heater are also being monitored by the controller."
My two cents: Good stuff for Applicants. This sort of broad brush treatment of the "capability" of a reference isn't common in the art units where I do most of my work. But it appears to be all too common in Tech. Center 1700. I realize that none of these decisions are precedential. Even so, if faced with such a rejection, I would borrow the reasoning used by the Board in these cases. It's a good way to school the Examiner in how case law such as In re Scheiber is properly applied to computer or controller limitations.
Note that the Typhoon and Microprocessor Enhancement cases are infringement suits. However, the stuff about capabilities and functional language relates to claim construction, not infringement analysis. And while a different standard is used for claim construction standard during prosecution, I say the "capabilities" rule used by the Federal Circuit in these cases applies equally to prosecution.