BETH LABSON FREEMAN, District Judge.
This consolidated action consists of two underlying cases. In one case, Power Integrations, Inc. ("PI") sues ON Seminconductor Corporation and Semiconductor Components Industries, LLC (collectively, "ON"), alleging infringement of seven patents, including five that are at issue in the present claim construction dispute: U.S. Patent Nos. 6,249,876 ("the '876 Patent"); 8,115,457 ("the '457 Patent"); 7,239,119 ("the '119 Patent"); 7,538,533 ("the '533 Patent"); and 6,297,623 ("the '623 Patent"). In the other case, ON alleges that PI infringes claims of eight of its patents, including three that are at issue in the present claim construction dispute: U.S. Patent Nos. 6,429,709 ("the '709 Patent"); 7,447,601 ("the '601 Patent"); and 6,333,624 ("the '624 Patent"). The Court held a tutorial on September 13, 2018 and a Markman hearing on September 20, 2018 for the purpose of construing ten disputed terms in the above listed patents.
The Asserted Patents are directed generally to power supply controller circuits and switching power supplies. Each side alleges that the other party's products infringe the Asserted Patents. See generally PI. Compl., ECF 1; ON Third Am. Compl., ECF 94; ON Second Am. Ans, ECF 96; PI Ans., ECF 100. Each patent is summarized below.
The '624 Patent is titled "Circuit and Method for a Switching Power Supply with Primary Side Transformer Sensing" and was issued on December 25, 2001. Ex. 1 to ON Br. ("'624 Patent"), ECF 157-1. This patent claims a switching power supply using a voltage regulation compensation circuit, which receives information from the primary side of a transformer to compensate for load variation and provide a regulated power supply output. See id. at 5:25-27. The invention aims to address the issues associated with having a feedback mechanism on the secondary side of the transformer, which requires secondary-side feedback devices that make the power supply more complex and costly. See id. at 1:65-2:7; 5:28-34.
The '601 Patent is titled "Power Supply Controller Method and Structure" and was issued on November 4, 2008. Ex. 2 to ON Br. ("'601 Patent"), ECF 157-2. This patent claims a structure and method for preventing damage to a power controller under large load conditions by determining the value of an input power and using that value to regulate a value of the output voltage. See id. at 1:21-23; id. at 6:47-50. The invention aims to protect the power controller in scenarios such as when the output voltage decreases due to a short on the output and a switch current attempts to regulate the output by increasing the load current. See id. at 1:15-26.
The '709 Patent is titled "Power Converter Circuit and Method for Controlling" and was issued on August 6, 2002. Ex. 3 to ON Br. ("'709 Patent"), ECF 157-3. This patent claims a power converter circuit and method for operating a power supply, specifically by providing a soft start circuit for reducing the on-time of a power switch of the power supply during initial start-up. See id. at 1:66-2:3. The invention can prevent transformer saturation and potential failure of the power switch during start up, as well as reduce power consumption of the power converter circuit when the load at the power supply output draws little or no power. See id. at 1:51-58; 2:3-11.
The '876 Patent is titled "Frequency Jittering Control for Varying the Switching Frequency of a Power Supply" and was issued on June 19, 2001. Ex. 1 to Decl. of Howard G. Pollack ("Pollack Decl.") ISO PI Br. ("'876 Patent"), ECF 160-1. The patent claims an invention for jittering the switching frequency of a switched mode power supply oscillator within a narrow range to reduce electromagnetic interference ("EMI") noise by spreading the energy over a wider frequency. See id., abstract, 1:66-67. The invention aims to alleviate problems associated with EMI in switched mode power supplies, which can have issues due to the high frequency operation of the power supply relative to the frequency of an alternating current power line. Id. at 1:19-25. The invention also aims to improve the prior art in several ways, including by allowing for jittering operation at higher temperatures and in a smaller area on the regulator chip. See id. at 3:59-4:5.
The '457 Patent is titled "Method and Apparatus for Implementing a Power Converter Input Terminal Voltage Discharge Circuit" and was issued on February 14, 2012. Ex. 2 to Pollack Decl. ("'457 Patent"), ECF 160-2. The patent claims methods and apparatuses for implementing a circuit for discharging a capacitance between input terminals of a power system when an electrical energy source is uncoupled from the input terminals. Id. at 3:15-19. Many power supplies use capacitors, such as X capacitors, to filter EMI generated by a power supply during operation. These capacitors can stay charged after the energy source is disconnected, which can pose a safety risk to anyone who touches the terminals. Id. at 1:29-47. The patent claims a circuit and method to discharge this capacitance to safe levels once the energy source is disconnected, while also eliminating the power dissipation that occurred in the prior art when the energy source was still connected. Id. at 3:58-67.
The '623 Patent is titled "Off-line Converter with Digital Control" and was issued on October 2, 2001. Ex. 5 to Pollack Decl. ("'623 Patent"), ECF 160-5. The invention claims a regulator circuit which allows for on/off control, or digital feedback, that controls the regulation of the circuit. See, e.g., id. at 2:57-67, 6:43-58. The claimed invention aims to solve problems associated with pulse width modulation used in power supplies operating at low power levels, including the need for compensation of the regulation loop. See id. at 2:15-32. The invention allows for a minimum amount of time spent out of the regulation cycle, creates a power supply that is more cost effective for low power solutions, and creates a power supply that uses the minimum number of components possible. Id. at 3:17-28.
The '119 Patent is titled "Method and Apparatus to Provide Temporary Peak Power from a Switching Regulator" and was issued on July 3, 2007. Ex. 3 to Pollack Decl. ("'119 Patent"), ECF 160-3. The invention claims an apparatus and method for increasing the maximum switching frequency of a switched mode power supply to deliver temporary peak power in the face of high load demands. See id. at abstract, 1:25-34. The invention aims to improve the efficiency, size, and cost associated with power supplies in the prior art that meet the requirements for continuous moderate power, while also providing short duration peak power. Id. at 1: 33-38.
The '533 Patent is titled "Method and Apparatus to Provide Temporary Peak Power from a Switching Regulator" and was issued on May 26, 2009. Ex. 4 to Pollack Decl. ("'533 Patent"), ECF 160-4. The '533 Patent is a continuation of the '119 Patent, discussed above, and their specifications are identical. See '533 Patent at 1:7-12.
Claim construction is a matter of law. Markman v. Westview Instruments, Inc., 517 U.S. 370, 387 (1996). "It is a `bedrock principle' of patent law that `the claims of a patent define the invention to which the patentee is entitled the right to exclude," Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed. Cir. 2005) (en banc) (internal citation omitted), and, as such, "[t]he appropriate starting point . . . is always with the language of the asserted claim itself," Comark Commc'ns, Inc. v. Harris Corp., 156 F.3d 1182, 1186 (Fed. Cir. 1998).
Claim terms "are generally given their ordinary and customary meaning," defined as "the meaning . . . the term would have to a person of ordinary skill in the art in question . . . as of the effective filing date of the patent application." Phillips, 415 F.3d at 1313 (internal citation omitted). The court reads claims in light of the specification, which is "the single best guide to the meaning of a disputed term." Id. at 1315; see also Lighting Ballast Control LLC v. Philips Elecs. N. Am. Corp., 744 F.3d 1272, 1284-85 (Fed. Cir. 2014) (en banc). Furthermore, "the interpretation to be given a term can only be determined and confirmed with a full understanding of what the inventors actually invented and intended to envelop with the claim." Phillips, 415 F.3d at 1316 (quoting Renishaw PLC v. Marposs Societa' per Azioni, 158 F.3d 1243, 1250 (Fed. Cir. 1998)). The words of the claims must therefore be understood as the inventor used them, as such understanding is revealed by the patent and prosecution history. Id. The claim language, written description, and patent prosecution history thus form the intrinsic record that is most significant when determining the proper meaning of a disputed claim limitation. Id. at 1315-17; see also Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996).
Evidence external to the patent is less significant than the intrinsic record, but the court may also consider such extrinsic evidence as expert and inventor testimony, dictionaries, and learned treatises "if the court deems it helpful in determining `the true meaning of language used in the patent claims.'" Phillips, 415 F.3d at 1318 (quoting Markman, 52 F.3d at 980). However, extrinsic evidence may not be used to contradict or change the meaning of claims "in derogation of the `indisputable public records consisting of the claims, the specification and the prosecution history,' thereby undermining the public notice function of patents." Id. at 1319 (quoting Southwall Techs., Inc. v. Cardinal IG Co., 54 F.3d 1570, 1578 (Fed. Cir. 1995)).
Paragraph 6 of 35 U.S.C § 112 provides for means-plus-function claiming: "An element in a claim for a combination may be expressed as a means . . . for performing a specified function . . . and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof."
If a court concludes that a claim limitation is a means-plus-function limitation, "two steps of claim construction remain: 1) the court must first identify the function of the limitation; and 2) the court must then look to the specification and identify the corresponding structure for that function." Id. The claim limitation will then be construed to cover that corresponding structure and equivalents thereof. 35 U.S.C § 112 ¶ 6.
Under 35 U.S.C. § 112, ¶ 2,
The Court first discusses the disputed terms in ON's patents.
The parties dispute one term in the '624 Patent, which is addressed below.
The parties dispute the following phrases in the '624 Patent:
The two disputed phrases are identical except that claim 6 recites "current reference signal" instead of "compensation signal." The Court addresses the two phrases together as the parties have done in their briefs. Claim 6 is representative and recites:
'624 patent at 6:7-18 (emphasis added).
As an initial matter, in its response, PI accepts ON's proposed clarification of current reference signal and compensation signal, see PI Resp. at 1 n.2, ECF 170, so this Court adopts that clarification.
The key dispute between the parties is whether the claim should refer to a value of the current reference signal or the value of the current reference signal. In its opening brief, ON "agree[d] to the substance of [PI's] construction," understanding it to require that "at a given time, there is a unique correspondence between the inductor current and the current reference signal," in accordance with Ohm's Law (V = IR; or voltage = current x resistance). See ON Br. at 1-2. ON also proposed what it deemed a purely "grammatical" rewording. Id. at 1. PI opposes ON's rewording because ON's construction refers to "a value" of "the current reference signal," whereas PI's proposal refers to "the value of the current reference signal." Id. at 2. According to PI, this difference is a meaningful one because ON's proposed claim would be satisfied so long as "at least one" value of the current reference signal represents a unique value of the inductor current, whereas PI's proposal requires that "every value" of the current reference signal represent a unique value of the inductor current. Id. Like ON, PI relies on Ohm's Law, arguing that the "value of the voltage drop across the resistor (`the current reference signal') represents exactly one unique value of current through the resistor (`a unique value of the inductor current')." Id. at 1.
In its reply, ON argues that PI's requirement that "every value of the current reference signal must represent a unique value of the inductor current" imports a new limitation into the claim, namely that the current reference signal must always represent a unique value of the inductor current. ON Reply a 1, ECF 171; see Claim Constr. Tr. 110:7-9, ECF 192. It argues first that the intrinsic record does not define "representative of" to require such a limitation because this limitation will not be true at start-up when the start-up circuit is operating and the current reference signal is zero. Second, it argues that PI's limitation would always have to be satisfied to find infringement, which is contrary to governing law. As to method claim 24, ON says that PI's construction is precluded by Bell Communications Research, Inc. v. Vitalink Communications Corp., 55 F.3d 615, 622-23 (Fed. Cir. 1995), which holds that "an accused product that sometimes, but not always, embodies a claimed method nonetheless infringes." ON Reply at 1. As to apparatus claims 6 and 15 and their dependent claims, ON argues that those claims include function tied to a recited structure such that the functional language is interpreted as something "the structure can do, but need not actually (much less always) do." Id. at 1-2 (citing MasterMine Software, Inc. v. Microsoft Corp., 874 F.3d 1307, 1313-16 (Fed. Cir. 2017)).
The Court agrees with PI that the claim requires that "the value of the current reference signal [compensation signal] represents a unique value of the inductor current." The purpose of the invention is to regulate the power supply output to compensate for load variation. See '624 Patent at 5:25-27; see also id. at 3:11-13 ("The output voltage of switching power supply
Given that the invention is aimed at regulation, the Court agrees with ON that at start-up, when the start-up circuit governs, the current reference signal might not represent a unique value of the inductor current. See Claim Constr. Tr. 110:7-9 (PI's counsel agreeing that start-up is not at issue here). But where the circuit is in "normal operating conditions" (ON's language in its claim construction hearing presentation), PI's construction is correct. Because the claim does not apply to start-up conditions, there is no problem for the apparatus claims under MasterMine— during start up, the structure (the compensation circuit) need not perform the recited functions. As to the method claim, the Court is not convinced that Bell Communications has any applicability to this argument at all, as that case had nothing to do with claim interpretation, but rather involved claim infringement. See Kaneka Corp. v. Zhejiang Med. Co., No. CV 11-02389-SJO, 2018 WL 2718036, at *27 (C.D. Cal. Apr. 5, 2018) ("It is only once the claim scope is understood that a court may determine if a defendant infringes those claims. [Plaintiff's] argument goes to the first of these steps—claim interpretation—while the court in Bell Commc'ns addressed the second."). But in any event, even were Bell Communications relevant, the power supply need not practice the claimed method at start-up, as discussed.
The Court thus adopts PI's construction: "the value of the current reference signal [compensation signal] represents a unique value of the inductor current."
The parties dispute one term in the '601 Patent, which is addressed below.
The disputed term "form drive pulses to regulate the output voltage" appears, among other places, in independent claim 10, which recites:
'601 Patent at 8:11-123 (emphasis added).
The parties' constructions seemingly differ in two ways: (1) "form drive pulses" (ON) vs. "generate drive signal pulses" (PI); and (2) "regulate the output voltage" (ON) vs. "maintain the output voltage substantially at a desired value" (PI). The Court discusses each difference in turn.
PI concedes that there is no substantive difference between the verbs "form" and "generate," but it argues (without citation) that "`form' in the context of an electrical signal is less clear than `generate.'" The Court sees no need to rewrite the claim to accommodate PI's verb preferences. Indeed, to do so would be contrary to ordinary claim construction principles. Cf. Phillips, 415 F.3d at 1313 (Claim terms "are generally given their ordinary and customary meaning"); GE Lighting Sols., LLC v. AgiLight, Inc., 750 F.3d 1304, 1309 (Fed. Cir. 2014) ("[T]he specification and prosecution history only compel departure from the plain meaning in two instances: lexicography and disavowal."). As such, the Court adopts ON's construction of the plain and ordinary meaning.
Turning to the more substantial dispute, the Court finds that the parties are not actually at odds. As an initial matter, ON essentially agrees with PI that the term "regulate" is equivalent to "maintain . . . substantially at." See ON Br. at 3 ("[T]he controller regulates that output voltage so that it remains within accepted tolerances, such as plus or minus ten percent of a desired value. Importantly the regulated output voltage is merely substantially constant. . . ." (citing '601 Patent at 3:27-31; 4:51-54)); id. at 4 ("[T]he disputed phrase . . . allows variation of the output voltage within acceptable tolerances that may change." (emphasis added)).
By contrast, ON challenges PI's construction because ON interprets it to require a "constant" desired value that cannot change. See ON Br. at 3; ON Reply at 3. But PI makes clear that its addition of "a desired value" should not be interpreted in this way. PI admits that "the `desired value' can change during operation." PI Opp. at 4. According to PI, its addition of "a desired value" aims instead to tether the output voltage to some desired value, unlike ON's construction which PI argues "does not require targeting a `desired value' of the output voltage." Id. at 3. That is, PI believes ON "attempts to twist the teaching of changing the known desired value during operation into the absence of a target at all." Id. at 4. This assertion clearly misconstrues ON's argument; ON never argues that no desired value is necessary, but instead argues only that a fixed desired value is not necessary. Thus the parties agree that the output voltage is maintained substantially at a desired value that can change.
The Court agrees with the parties that the output voltage must be targeted at some desired value, but that the desired value can change. See, e.g., '601 Patent at 4:7-12 ("The change in the output voltage represents a transition from the condition of the desired steady state output voltage value or first value to another steady state regulated voltage condition."). The desired value is neither a constant (as ON interprets PI's construction to require), nor nonexistent such that "the output voltage fluctuates without regulation" (as PI interprets ON construction to require), PI Opp. at 4. The Court finds that PI's proposed construction best reflects this understanding. To be clear, though the Court adopts PI's construction (as to this portion of the disputed term), inclusion of the term "a desired value" "does not require that the output voltage be maintained at a constant value, but instead allows variation of the output voltage within acceptable tolerances [around a desired value] that may change," ON Br. at 4.
As such, the Court adopts the following construction: "form drive pulses that are used to maintain the output voltage substantially at a desired value."
The parties dispute one term in the '709 Patent, which is addressed below.
The disputed term "comparing the input voltage to a second reference signal after receiving the first compare signal" appears in independent claim 12, which recites:
'709 Patent at 13:1-11.
The dispute revolves around the definition of the word "after." ON argues that after is a purely "sequential" term: the input voltage comparison to the second reference signal need only occur at some time after (i.e., later than) receipt of the first compare signal. ON Br. at 5. ON argues that this sequential reading is supported by the structure of the claim, which recites "three steps performed in sequential order." Id. ON also points to Figure 2 of the patent, which illustrates a power supply that performs an embodiment of claim 12. According to ON, the specification describes the three steps of claim 12 sequentially, wherein (1) "[c]omparator 71 compares bias voltage Vcc to reference VREF1 to provide a compare signal, i.e., signal UVL," ("first compare signal") id. at 6 (quoting '709 Patent at 8:15-16); (2) "[a]fter amplifier 81 receives [compare] signal UVL having a logic high voltage, amplifier 81 compares bias voltage Vcc to reference voltage VREF2 to provide a compare signal to the gate electrode of FET 82," ("second compare signal"), id. (quoting '709 Patent at 8:60-63); and (3) FET 82 passes a control signal from node 87 to node 88 after receiving the compare signal from amplifier 81," id. (quoting '709 Patent at 8:63-65). See also id. (citing prosecution history distinguishing prior art where the comparisons were executed simultaneously, not sequentially). At the same time, ON recognizes that the specification describes that, during operation of the embodiment, "signal UVL is at a logic high voltage and amplifier 81 is enabled." Id. (quoting '709 Patent at 8:63-65). Still, ON argues that this is not enough to warrant importing such a limitation into the claim.
PI retorts that this is exactly why the limitation is required, because the specification elucidates that the term "after" is one of causation—the second comparison is "enabled by" receipt of the first compare signal. PI explains that the patent describes a functionality known as "under voltage lockout" ("UVL"), a process by which the power supply can effectively establish the operating bias voltage necessary for normal operation. See PI Opp. at 6. When in UVL mode, the UVL signal (i.e., "first compare signal") has a logic low value. See '709 Patent at 4:27-29. According to PI, this low value UVL signal "disables amplifier 81," the component that performs the second comparison, such that the second comparison cannot occur. Id. (quoting '709 Patent at 8:25-27). In confirmation of this reading, the specification uses the term at issue "after" to describe this process: "After amplifier 81 receives signal UVL having a logic high voltage, amplifier 81 compares bias voltage Vcc to reference voltage VREF2 to provide a compare signal to the gate electrode of FET 82." Id. at 6-7 (quoting '709 Patent at 8:60-63). PI also contrasts this causal function with a hysteretic comparison performed by comparator 71 as described in the specification. According to PI, though this hysteretic comparison requires only sequential functionality, the specification never uses the term "after" to describe it, instead reserving that term for the causal functionality intended in claim 12. See id. at 7.
The Court agrees with PI that the term "after" in claim 12 connotes a causal relationship. ON's own reference to the specification makes clear that the second comparison occurs not simply "after" receipt of the first compare signal, but rather only after receipt of a compare signal "having a logic high voltage." '709 Patent at 8:60-63. That is because when the first compare signal "is at a logic high voltage . . . amplifier 81 is enabled." Id. 8:57-59. Amplifier 81 can then perform the second comparison. By contrast, where the first compare signal has a logic low voltage, the second comparison cannot occur because the low voltage "disables amplifier 81." Id. at 8:25-27. So it is only "[a]fter amplifier 81 receives [the first compare signal] having a logic high voltage" that amplifier 81 is enabled such that it performs the second comparison. Id. at 8:60-63 (emphasis added). The term "after" thus encapsulates that enablement. And though the Court does not look to the accused product when construing this claim, the fact that the hysteretic comparator 71 is not described using the term "after," whereas amplifier 81 is so described, further suggests that the enablement component is an important aspect of the claim. Compare id. at 5:57-6:10 with id. at 8:26-29, 8:57-63.
As such, the Court adopts PI's substantive construction of the term. However, given the language and structure of the other claim elements, the Court modifies PI's construction to better mirror those elements and to ease understanding. The Court's modification is not meant to substantively differ from PI's construction. As such, the Court adopts the following construction: "comparing the input voltage to a second reference signal after receiving and being enabled by the first compare signal."
The Court next discusses the disputed terms in PI's patents.
The parties dispute one term in the '876 Patent, which is addressed below.
The disputed term "means coupled to the control input for varying the switching frequency, including: one or more current sources . . . and a counter coupled to the output of the oscillator and to the one or more current sources" appears in claim 21, which recites:
'876 Patent at 9:55-65 (emphasis added).
This term was previously construed in the District of Delaware in a case between PI and Fairchild
This is a means-plus-function term, governed by 35 U.S.C. § 112, ¶ 6. As such, the Court construes the claim element "to cover the corresponding structure, material, or acts described in the specification and equivalents thereof." Id. Because the law contemplates that the element will cover equivalents of the structure, the Court adopts PI's proposed addition of "and equivalents thereof" to "a counter having at least four bits." To construe this means-plus-function limitation, the Court must conduct two claim construction steps: "(1) the court must first identify the function of the limitation; and (2) the court must then look to the specification and identify the corresponding structure for that function." Biomedino, 490 F.3d at 950.
The parties agree that the function of the limitation is "varying the switching frequency," but PI believes the function should include "of the oscillator," such that the whole function would be "varying the switching frequency of the oscillator." PI Br. at 6. ON notes there is "little substantive difference between the proposals because the antecedent basis for `the switching frequency' is already recited in the oscillator element." ON Resp. at 4. PI notes that its interpretation is meant to maintain clarity and consistency with the Delaware court's interpretation. PI Reply at 4, ECF 172. Because there is no dispute that the function is varying the switching frequency of the oscillator, the Court adopts PI's construction, for clarity and consistency with the Delaware Court.
The next step is for the Court to identify the corresponding structure responsible for varying the switching frequency of the oscillator. The parties agree that the structure is a digital-to-analog converter ("DAC") and a counter having at least four bits, but ON proposes that it must be a DAC "with current sources," while PI's proposal includes no such limitation.
PI argues that ON's proposal is incorrect for two reasons. First, the claim cannot include a limitation of multiple current sources because the claim's plain language says "one or more current sources." PI Br. at 5. And second, the one or more current sources need not be a part of the DAC itself. Id. PI notes that the specification contemplates a means structure with only a single current source, id. (citing '876 Patent at 3:15-22), and also contemplates a voltage-based structure, id. (citing '876 Patent at 3:9-16). PI thus reads the claim, in light of the specification, to require means including "one or more current sources (per the express claim language), and a DAC and counter (having at least four bits)." Id. at 6 (emphasis omitted).
In response, ON argues that the means for varying the switching frequency must include two things: (1) a counter and (2) one or more current sources. ON. Resp. at 5. As such, the corresponding embodiment must have both a counter and one or more current sources. According to ON, though there are three different structural embodiments for varying the switching frequency, only one embodiment has both a counter and one or more current sources. These embodiments include (1) the embodiment in Figure 1 ('876 Patent at 4:32-39), which includes multiple current sources in DAC 150 as well as a counter; (2) an embodiment in which the DAC is implemented by voltage sources (id. at 4:63-66) and has a counter; and (3) an embodiment that uses a single current source, but without a counter (id. at Fig. 3; 6:6-29).
Given the two structural limitations, in ON's view, only embodiment 1 corresponds to claim 21. ON Resp. at 5. ON forecloses any combination of these embodiments for implementing DAC 150, even though it recognizes that DAC 150 may be implemented using voltage sources or current sources. Id. at 6 (citing '876 Patent at 4:62-66). According to ON, because the specification does not describe how to combine voltage sources and current sources into a single DAC, the DAC must be either voltage-based or current-based—that is, contain either solely voltage sources or solely current sources. See ON Resp. at 6 (citing ON Resp. Ex. 5 ("Baker Decl.") ¶¶ 23-27, ECF 168-6). ON's expert avers that this conclusion is supported by the "technical realities" of these types of sources—"frequency jittering voltage sources and frequency jittering current sources would be electrically incompatible," such that "DACs (including those known in the art at the time of the '876 Patent) are generally configured" to use either voltage- or current-based outputs only. Baker Decl. ¶¶ 26-27.
PI notes in reply that ON concedes that a voltage-based DAC is a corresponding structure, but that ON's expert "incorrectly alleges that a voltage-based DAC structure would not require any current sources." PI Reply at 5. PI argues that this is incorrect. PI's expert argues that a person of ordinary skill in the art ("POSITA") would understand that a voltage-based DAC uses a current source, and also that a voltage-based DAC could not function without a current source. Decl. of William Bohannon ISO PI Reply ¶¶ 12-13, ECF 172-1. According to PI, claim 21 thus covers voltage-based DACs ("one . . . current source[]") and current-based DACs ("more current sources"). PI Reply at 6.
As with all claim construction disputes, the Court starts with the plain language of the claim. The disputed term requires, in part, "one or more current sources." To construe this claim in ON's favor would mean reading out "one" from "one or more." Each side agrees that the corresponding structure includes a DAC. The specification makes clear that the DAC can have one or more current sources. See, e.g., '876 Patent at 3:15-22 ("The means for varying the frequency may include . . . a current source. . . ." (emphasis added)). Though means-plus-function limitations are limited to their corresponding structure, the claim language defines what the corresponding structure can be.
With this language in mind, ON's argument that only the first embodiment encapsulates the two required structures of claim 21 is unavailing. The second embodiment states that "[c]ounter 140 has a plurality of outputs . . . which are not used. The remaining outputs . . . are connected to [DAC] 150, which may be implemented as a series of frequency jittering voltage sources or current sources." Id. at 4:62-66. By its own terms, it encapsulates both a counter and DAC 150 with current sources. ON attempts to moot this description of a DAC with current sources by arguing that the specification "does not suggest combining these different embodiments" and that voltage and current sources would be "electrically incompatible if coupled to the same control input of the oscillator." Baker Decl. ¶¶ 25-26. But PI's expert makes clear that voltage-based DACs use at least one current source, such that a POSITA would understand claim 21 to cover the voltage-based DAC as well. See Bohannon Decl. ¶¶ 12-13. Thus, to the extent the voltage-based DAC includes a current source, the embodiment ON attempts to distinguish constitutes a corresponding structure of claim 21 because it contains a counter and one current source.
At the hearing, ON noted that PI's construction (and the Delaware Court's construction) cannot be right, since it does not include the limitation of a current source at all—that is, nothing in the language of PI's proposal requires any current sources. See Claim. Constr. Tr. at 24:4-10. To the extent this is a fair reading of PI's construction, the Court makes clear that the claim covers "one" current source (e.g., as in a voltage-based DAC), "or more" than one current source (e.g., as in a current-based DAC), but it does not cover a DAC with no current sources. Accord Fairchild II, 2009 WL 4928029, at *24 n.12 ("[T]he DAC cannot be implemented through means lacking current sources (such as embodiments using only voltage sources or capacitors), because the claim expressly limits itself to means having one or more current sources.").
Accordingly, the Court finds that the disputed term is a means-plus function term, with a function of "varying the switching frequency of the oscillator," and a structure of a digital-to-analog converter and a counter having at least four bits and equivalents thereof.
The parties dispute two terms in the '457 Patent, which are addressed below.
The disputed term "less than a maximum period of time" is recited in claims 1, 23, 29, and 40. Claim 1 is representative:
'457 Patent at 16:48-65 (emphasis added).
The parties dispute whether the term is indefinite under Nautilus, 134 S. Ct. at 2124— whether "[the patent's] claims, read in light of the specification delineating the patent, and the prosecution history, fail to inform, with reasonable certainty, those skilled in the art about the scope of the invention."
PI argues that the term is not indefinite. PI first explains that the purpose of the invention is to allow electrical energy stored in an X-capacitor to be discharged in less than a maximum period of time as mandated by international safety standards. See PI Br. at 6. It then explains that this maximum period of time equals the "sum of two distinct independent time periods: first, the time period that it take for the circuit to detect the AC input voltage has been disconnected and, second, the time it takes to discharge the X-capacitors to a safe level." Id. at 7 (citing '457 Patent at Figure 9). The patent then describes how to complete these steps within a "predefined maximum period of time." Id. (citing specification). Recognizing that Nautilus governs, PI argues that ON cannot show by clear and convincing evidence that the term is indefinite because in previous litigation involving this patent ON's now-subsidiary Fairchild never suggested that the term was indefinite, and Fairchild's expert was able to interpret the term without issue. See id. at 8.
ON argues that the "maximum period of time" contemplates an upper limit on the time in which the capacitor can discharge, but that "the '457 Patent offers no guidance on what that upper limit is." ON Resp. at 7 (citing Baker Decl. at ¶¶ 31-33). ON cites first to a Patent and Trial Appeals Board ("PTAB") Final Written Decision, which stated that the claim merely recites "some undefined maximum period of time" that is "not specified." Ex. 7 to ON. Resp. at 11, ECF 168-8. ON next points to the specification, which refers "to only a single example standard, EN60950-1, which discloses a `time constant' of 1 second." ON Resp. at 7 (citing '457 Patent at 1:55-58). ON argues that this example is not incorporated into the claimed maximum period of time, and in any event, this example does not comport with PI's proposed interpretation because it does not take into account one of the two time periods PI says must be included in the construction. See id. at 7-8 (citing PI argument in inter partes review). Finally, ON submits that PI's proposed construction "would only further compound the indefiniteness problem by splitting one indefinite maximum period into two equally indefinite maximum period of time." Id. at 8. ON also argues that PI's construction is simply incorrect, as the PTAB rejected "an essentially identical construction" and PI's construction imports several limitations that are not justified by the intrinsic record. Id.
In reply, PI asserts that the "maximum period of time" need not be known with specificity; "the claim is broader and does not require a specific time, such as the one second example given in the specification." PI Reply at 7. At the Claim Construction hearing, PI's counsel confirmed that the claim was not limited to the one second standard in the specification because "at any given time, the standard could change." Claim Constr. Tr. at 33:20. Instead, the maximum period of time is just some period of time that is "predefined before you start the discharge," meaning it is not merely "coincidental." Id. at 34:11-13. Under the claim, "you have to choose a value, and once you've chosen a value, you have to meet it. And if you do that, you meet the claim." Id. at 36:9-11. Finally, PI argues that its construction is correct because the two distinct time periods are necessary components of the maximum period of time.
The Court agrees with ON that the term "less than a maximum period of time" is indefinite. A POSITA, looking to the specification and the prosecution history, could not with reasonable certainty determine the scope of the claim. Nautilus, 134 S. Ct. at 2124. PI all but concedes this, stating at the hearing that the predefined time required by the term could be "ten minutes," "half a second," "one second," or "maybe it differs by country." Claim Constr. Tr. at 33-35. And it agreed with the Court that the time could be the time in which the capacitor "would naturally discharge," the one second set by the standard, or some time required by an older standard. Id. at 34:16-35:3. Under PI's definition, a user could predefine any time period, including a time that was as close as possible to the natural discharge time (i.e., a non-predefined time). But if the user chose that time period, the invention would not perform its intended function, namely discharging the energy from the capacitor in a "safe" period of time. See '457 Patent at 1:43-58. Ultimately, PI could not provide a single time period that would be too great to be considered above the "maximum period of time." Without an objective boundary, the claim is indefinite. Cf. Tinnus Enterprises, LLC v. Telebrands Corp., 733 F. App'x 1011, 1018 (Fed. Cir. 2018) (rejecting indefiniteness claim because patent "provide[d] objective boundaries for understanding the disputed term"). PI's splitting of this maximum period into two discrete periods does nothing to solve the issue—those periods are likewise indefinite; nothing in the specification indicates how long each period can be.
Because the specification and prosecution history do not provide an upper bound for "maximum period of time" the Court finds the term "less than a maximum period of time" to be indefinite under Nautilus.
The disputed term "wherein a current flow through the circuit is less than 30 μA in the first operating mode" is recited in claim 10, which is dependent on claim 1. As noted above, claim 1 recites:
'457 Patent at 16:48-65 (emphasis added).
Claim 10 recites:
'457 Patent at 17:22-24 (emphasis added).
The differences between the two proposed constructions present three questions the Court must resolve: (1) can the number of microamps be less than zero (i.e., negative); (2) does "the circuit" include both the control circuit and the switch; and (3) does "30 µA" represent the average of the current flowing through the circuit, or does it instead represent the instantaneous value of that current? There is no actual dispute on the first question: PI and ON agree that "less than zero" means between 0 and 30 µA. See Claim Constr. Tr. at 49:12 (PI agreeing that the current cannot have a negative value). The Court adopts this construction.
The second question is also easily resolved. The plain language of independent claim 1, on which claim 10 depends, defines "the circuit" as comprising both "a control circuit" and "a switch." PI argues that the embodiments in the specification elucidate that the current "does not necessarily flow through the switch" because the "the switches . . . are generally off, i.e., having a high average impedance." PI Br. at 9 (citing '457 Patent) at 7:45-51; 8:54-61. But PI also concedes that "little if any current flows through the [switches]" (that is, not no current) and also that "both the current sources and the switch are between terminals 225 and 226," the area in which the current flow is less than 30 µA. Id. at 9 & n.4. These citations and concessions directly belie PI's argument (and comport with the claim's plan language). PI is concerned that ON's construction requires that current be flowing through the switch in this mode. See id. ON rejects this reading, arguing that its construction requires only that any current flowing through the switch (to the extent there is any) must be included in the total current calculation. The Court agrees with ON. Reading "the circuit" to comprise both the control circuit and the switch does not require that current flow through the switch; it requires only that if current is flowing through the switch, that current be included in the total current calculation.
The final question is whether the 30 µA figure represents the average value of the current flow or the instantaneous value of the current flow (i.e., the value of the current flow at any one time, or peak value). ON argues that the claim must recite an average current flow, "because the electrical energy source may provide an alternating current (`AC')." ON Resp. at 11. Relying on its expert, ON explains that "[f]or an AC voltage source . . . a POSITA would understand the relevant measure of power is the average (e.g., [root mean square] rms) power." Id. (citing Baker Decl. ¶ 40). Moreover, the magnitude of the current flow is proportional to the power dissipated by the circuit. Id. (citing Baker Decl. ¶ 37). Thus, for an AC power source, a POSITA would know that because the power is measured as an average, the current too is measured as an average. See id. ON also notes that using the rms average does not change the outcome for DC circuits because the rms average value and the instantaneous value are identical in DC circuits. See Baker Decl. ¶¶ 43, 46.
ON also argues that this construction is supported by the specification, where both voltage and impedance (which have known relationships to current under Ohm's Law) are described in terms of averages. See '457 Patent at 3:55-57 ("In one example, the electrical energy source is a mains ac voltage source have an rms voltage level in the 85 to 264 Vac range."); id. at 8:54-61 ("In one example, the impedance . . . is such that a current flow . . . is less than 30 µA corresponding to an average impedance of typically greater than 3MOhms. . . ."). When the current is calculated using these average values (i.e., 85 volts/3 Mohms), the result is 28.33 µA, which is fully consistent with the claim's requirements. See ON. Resp. at 11. By contrast, using these averages to calculate the instantaneous current value gives a result of 40.1 µA (85 volts/3 Mohms x 0.707), which does not comport with the claim. See generally Claim Constr. Tr. at 47-50.
In its opening brief, PI argues only that reading in an average "deviat[es] from the claim language, which merely requires that `a current flow through the circuit' be `less than 30 µA.'" PI Br. at 9. In reply, it also notes that the specification never requires an average current. PI Reply at 8. It argues additionally that the specification's reference to average impedance is irrelevant to this dispute because no current is flowing through the switch in this mode, such that discussing the current flow through the switch is irrelevant. PI Reply at 8 (citing Bohannon Decl. ¶¶ 17-18). Finally, at the claim construction hearing, PI argued that this claim concerns a DC current, not an AC current, because "you only power a chip with a DC current." Claim Constr. Tr. at 50:3-7.
The Court finds that ON has the better argument. PI does not engage with ON's argument, supported by expert testimony, that a POSITA would understand that the current, like the power, in an AC circuit is measured in terms of the rms average. Instead, it argues that this fact is beside the point, because the average impedance discussed in the specification relates to the switch, which is off during the first operating mode (the mode at issue in this claim). But the cited specification language is not limited to the switch. Instead, it states: "In one example, the impedance between terminals
For these reasons, the Court adopts ON's construction. The term "wherein a current flow through the circuit is less than 30 µA in the first operating mode" means "wherein the RMS
The parties dispute two terms in the '623 Patent, which are addressed below.
The parties dispute the relationship between the two terms "control signal" and "duty cycle signal." These terms are recited in claim 1, which states:
'623 Patent at 12:2-13 (emphasis added).
As discussed at the hearing, the Court believes PI's construction to be improper, as it asks the Court to define the relationship between two claim elements. The parties are instructed to file a joint stipulation construing the term control signal by November 9, 2018. If no stipulation is submitted, the Court adopts the plain and ordinary meaning of the term.
The disputed term "said control signal being provided when no feedback signal is provided at said feedback input and said duty cycle signal is in said high state" is recited in claim 1, which recites:
'623 Patent at 12:2-13 (emphasis added).
ON proposes that these claims are indefinite under the Federal Circuit's decision in IPXL Holdings, LLC v. Amazon. com, Inc., 430 F.3d 1377 (Fed. Cir. 2005). In IPXL, the Federal Circuit addressed the question of "[w]hether a single claim covering both an apparatus and a method of use of that apparatus is invalid" for indefiniteness. Id. at 1383. The claim at issue recited both a system and a method for using that system (requiring that the user use a system element in a certain way). Id. at 1384. The Court held the claim indefinite, reasoning that mixed apparatus and method claims make it unclear as to when infringement occurs: when one makes the apparatus or when the user performs the claimed method. See id. Such ambiguity "does not apprise a person of ordinary skill in the art of [the claim's scope]." Id.
PI argues that this term is not indefinite under IPXL because it is an "apparatus claim element that reads on a regulator circuit structure that provides the `control signal' when the stated conditions are met." PI Br. at 12. PI notes that Federal Circuit law allows claims to "recit[e] physical [apparatus] elements in terms of the function they perform." Id. PI then argues that IPXL is limited to scenarios in which "infringement occurs only when the device is used in a particular manner by a user." Id. at 12-13. By contrast, where the claim merely describes how the apparatus functions "under a given set of conditions"—that is, where the "claim element describes a condition that is present or an aspect of the apparatus environment"—the claim is not indefinite. Id. at 13 (citing UltimatePointer, L.L.C. v. Nintendo Co., 816 F.3d 816, 827 (Fed. Cir. 2016); HTC Corp. v. IPCom GmbH & Co., KG, 667 F.3d 1270 (Fed. Cir. 2012)). PI argues that the recited claim here is the latter type because it "maps to a known set of digital logic circuit structures that could accomplish providing the control signal when the recited conditions are met." Id.
In rebuttal, ON contends that PI ignores an entire line of case law after IPXL that makes clear that functional language in an apparatus claim is indefinite not only where it recites a step performed by a user, but also where "the functional language is not specifically tied to a recited structure." ON Resp. at 13 (citing MasterMine Software, Inc. v. Microsoft Corp., 874 F.3d 1307 (Fed. Cir. 2017); Rembrandt Data Techs. v. AOL, LLC, 641 F.3d 1331 (Fed. Cir. 2011)). According to ON, claim 1 recites a function ("said control signal is provided when . . .") without reciting the structure that provides that signal, in clear violation of IPXL. Though the claim does recite three structures ("the feedback input," "the switch," and "the oscillator"), ON argues that none of these is capable of providing the control signal, as confirmed by the specification. Id. at 13-15 (citing '623 Patent at Figure 6).
In reply, PI abandons its user-action theory, and instead argues that a person of ordinary skill in the art would understand that the structure being described is "a standard MOSFET switch . . . that is capable of receiving a control signal." PI Reply at 9. With that understanding, it becomes clear that "the disputed element simply describes a circuit structure that determines when the switch should be turned on and off, i.e., it should be turned on `when no feedback signal is provided at said feedback input and said duty cycle signal is in said high state.'" Id.
The Court is persuaded that ON is correct, and that this term is indefinite. As an initial matter, the law is clear that user-action is not the only cause of indefiniteness under IPXL. In Rembrandt, the Federal Circuit held a claim invalid where it recited apparatus elements and a method element requiring that the claimed device perform the method, but not requiring user action. 641 F.3d at 1339 (citing IPXL). Such a claim is at issue here. The key question the Court must answer is whether this disputed element constitutes "a method of use of th[e] apparatus," IPXL, 430 F.3d at 1383, or whether instead the claim "is clearly limited to [an apparatus] possessing the recited structure and capable of performing the recited functions." Microprocessor Enhancement Corp. v. Texas Instruments Inc. ("MEC"), 520 F.3d 1367, 1375 (Fed. Cir. 2008).
The Federal Circuit has held invalid on numerous occasions claims reciting both an apparatus and a method of using that apparatus. As discussed above, in IPXL, the Court held a claim indefinite because it recited an apparatus element and required the user to use the apparatus element. 430 F.3d at 1383. In Rembrandt, the Court extended IPXL's holding to invalidate a claim that recited both apparatus elements and a method element that was performed by the apparatus (though not tied to a recited structure). 641 F.3d at 1339. Finally, in In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1318 (Fed. Cir. 2011), the Federal Circuit invalidated a system claim that also included method steps "directed to user actions, not system capabilities."
Another line of cases has distinguished IPXL where the claims at issue were "clearly limited to [an apparatus] possessing the recited structure and capable of performing the recited functions." MEC, 520 F.3d at 1375. In MEC, the Federal Circuit found definite a method claim with recited structural limitations because it was "clearly limited to practicing the claimed method in an [apparatus] possessing the requisite structure." Id. In MasterMine, the Court rejected the district court's finding of indefiniteness where the claim included "functional language used to describe the capabilities of [an apparatus element]." 874 F.3d at 1315. The court distinguished the holding in Rembrandt in finding that "the functional language here does not appear in isolation, but rather, is specifically tied to structure: [the apparatus element]." Id. at 1316; see also HTC Corp., 667 F.3d at 1277 (finding claim definite where the claim did not "recite [an apparatus] and then have the [apparatus] perform the six enumerated functions," but instead "merely establish[ed] those functions as the . . . environment in which the [apparatus] operate[d]."). Finally, in UltimatePointer, the court reversed the district court's finding of indefiniteness because the claims reflected the capability of the claimed apparatus. There, the "limitations only indicate[d] that the associated structures ha[d] [the functional] capability." 816 F.3d at 827. Each of these cases thus emphasized the fact that the functional language was not free-floating, but was instead tied to a recited structure, as opposed to being tied to the apparatus as-a-whole or to a user.
A review of this relevant case law makes clear that the disputed term is indefinite because it recites functional language divorced from any recited structure. As written, the claim requires that the control signal "be[] provided" under certain conditions. '623 Patent at 12:12-13. There is no recited structure in the claim that arguably provides that signal. Indeed, PI does not argue that any of the three recited structures in the claim provides the signal. As such, the functional language is not tied to the capability of any "associated structures." UltimatePointer, 816 F.3d at 827. Instead, the disputed term must be read as functional language constituting a method of the apparatus. See Rembrandt, 641 F.3d at 1339. In fact, PI's opening brief can be read to admit this exact point. Though it claims that the element is describing a "set of conditions" in which the regulator circuit operates, it also argues that this element "reads on a regulator circuit structure that provides the `control signal'" and "maps to a known set of digital logic circuit structures that could accomplish providing the control signal." PI Br. 12-13 (emphases added). These statements could be read to suggest that PI understands the regulator circuit (the apparatus) to be the thing performing the function. This would run headlong into IPXL.
Of course, in reply, PI changes tactics, arguing that this element "simply describes a circuit structure that determines when the switch should be turned on and off, i.e., it should be turned on `when no feedback signal is provided at said feedback input and said duty cycle signal is in said high state.'" PI Reply at 9. This choice quotation conveniently ignores the key language in the term, namely "said control signal being provided. . . ." PI's choice to ignore this language makes sense because its argument would require the Court to rewrite the claim language. PI wants this element to be read as denoting when the control signal is received by the switch (i.e., "when the switch should be turned on and off"). But the element mentions neither the switch nor receipt of the control signal. Instead, the claim recites only the provision of the control signal, without reciting an attendant structure to provide that signal. As such, the claim recites a function performed by the apparatus itself—a method element among apparatus elements.
Thus, the term "said control signal being provided when no feedback signal is provided at said feedback input and said duty cycle signal is in said high state" is indefinite under IPXL. Because the Court finds the term indefinite, it need not address the disputed term "no feedback signal."
The parties dispute one term in the '119 Patent, which is addressed below.
The disputed term "temporarily increasing" is recited in claim 26, which recites:
'119 Patent at 9:28-10:6 (emphasis added).
The parties' briefing makes clear that this dispute is somewhat of a mirage—the parties do not actually disagree in any meaningful way. PI's construction targets scenarios in which the external load demand is constant (i.e., "in the absence of any external load changes"). PI argues that in such scenarios, the increase in the maximum switching frequency must be terminated by something other than the load level. See PI Br. at 13-14. Where the load level is constant and greater than the power level threshold, such as when a fault occurs, the circuit must be able to cease increasing the maximum switching frequency to avoid overheating the power supply or causing the circuit critical damage. See id. Because the load is unchanging, PI argues, it must be the internal circuitry that causes the switching frequency to cease increasing. See id. (citing specification). By contrast, PI reads ON's construction to mean that the increase is terminated by mere "happenstance" based "solely on the load level." Id. at 14. Finally, at the hearing, PI admitted that "it certainly can be the load" that causes the switching frequency to stop increasing and that its construction deals only with the scenario in which "the load doesn't change." Claim Constr. Tr. 96:14-18.
ON reads PI's construction as requiring that the internal circuitry be the only means for controlling the switching frequency increase, "irrespective of external load conditions." ON Resp. at 16. ON notes that the intrinsic record describes "at least two mechanisms for limiting the duration of the maximum switching frequency": (1) external load changes, see id. at 16 (citing specification); and (2) internal circuitry in the form of an "optional time limit circuit," see id. at 17 (citing '119 Patent at 7: 1-14). Importantly, ON concedes that its construction is not meant to define "temporarily increasing" to be a "mere coincidence." Id. at 16. ON does not expressly admit what happens in the absence of external load changes, but given that it recognizes only two mechanisms for limiting the increase, ON seemingly acknowledges that in the face of a constant external load, the increase must be limited by the internal circuitry, lest the switching frequency increase forever (i.e., not "temporarily").
The parties thus do not disagree. They both recognize that the increase in the switching frequency can be terminated by either a change in the external load or by the patent's internal circuitry. See, e.g., '119 Patent at 4:56-61 ("the maximum frequency is varied depending on the output power demand"); 3:12-15 ("the oscillator in controller
Given this agreement, the Court need only determine which of the two constructions best embodies this agreement. The Court finds that PI's construction is the better, more limited choice. On its face, ON's construction does not speak to internal circuitry or external load change, such that PI's reading of the construction as relying on mere happenstance is not an unfair one. By contrast, PI's construction is limited to the specific scenario in which there is an "absence of any external load change" and makes clear that in that scenario, internal circuitry must be what terminates the increase of the maximum switching frequency. Contrary to ON's assertion, PI's construction does not also limit what happens when the external load changes. In that scenario, as both parties agree, the switching frequency's increase can be limited by either the internal circuitry or external load limit changes.
As such, the Court construes the term "temporarily increasing" to mean "increasing for a period of time that is limited by the internal circuit in the absence of any external load changes."
The parties dispute one term in the '533 Patent, which is addressed below.
The disputed term "moderate power level threshold"/"moderate power level threshold value" is recited in claim 1, which recites:
'533 Patent at 7:20-39 (emphasis added).
The parties dispute whether the term is indefinite under Nautilus, 134 S. Ct. at 2124— whether "[the patent's] claims, read in light of the specification delineating the patent, and the prosecution history, fail to inform, with reasonable certainty, those skilled in the art about the scope of the invention."
PI argues that this term is not indefinite under Nautilus because "moderate" is a term of degree, and the specification provides sufficient guidance to "measure that degree." PI Br. at 15 (quoting Biosig Instruments, Inc. v. Nautilus, Inc. ("Nautilus Remand"), 783 F.3d 1374, 1377 (Fed. Cir. 2015)). According to PI, the specification distinguishes the moderate threshold from "maximum or peak" and "very low" power output levels. PI Br. at 15-16 (citing '533 Patent at 1:27-33; id. at 2:37-39 ("moderate power level is much higher than very low output power operating conditions")). The specification also provides two examples of the moderate power level threshold. First, it notes that a DVD player could have a peak output of 20 watts, a moderate output of 10 watts, and a very low output of 0.5 watts. See '533 Patent at 3:48-52. Next, it provides a graphical representation of the moderate power level (PM) in Figure 2:
'533 Patent at Figure 2; see also id. at 3:31-33.
ON argues that the specification lacks any standard for measuring the degree of "moderate." ON notes that PI argues that the specification provides the two bounds between which moderate falls—namely, it is greater than "very low" and less than the "maximum"—but that this "Goldilocks-style guidance" would tell a POSITA only that the "`moderate power level threshold' falls somewhere within a potentially enormous range," which amounts to a "zone of uncertainty" under Nautilus, 134 S. Ct. at 2130. ON Resp. at 18 (citing Baker Decl. ¶ 51). Moreover, ON notes that the distinction drawn between the moderate and maximum levels is one of length, not value. See id. (quoting '533 patent at 1:26-32 ("Typically, a moderate continuous output power is required for a long duration . . . However, a maximum or peak output power is usually required for a relatively short duration and infrequently . . .")). Finally, ON notes that the DVD application PI cites is not included in the claim and provides no real guidance beyond the values in a DVD player. Id. at 18 (citing Baker Decl. ¶¶ 49-50).
In reply, PI agrees with ON that a "`moderate' power level lies between the[] extremes" of the "very low" or "maximum or peak" power level. PI Reply at 10-11 (citing Bohannon Decl. ¶ 20). Indeed, PI's expert avers that a POSITA "would understand that `moderate' refers to the range of values between these extremes." Bohannon Decl. ¶ 20. Then, it argues that the patent provides the two examples to help "quantify the difference between the power levels." PI Reply at 11.
The Court finds that the term is indefinite because the patent's specification does not provide a POSITA sufficient guidance for determining the scope of the invention with reasonable certainty. See Nautilus, 134 S. Ct. at 2130. Under the claim's terms, the recited controller behaves one way when the load is above the moderate power level threshold value and behaves another way when it is below that value. See '533 Patent at 7:25-34. Thus, the moderate power level threshold value cannot simply fall anywhere within a range of values between "maximum peak" value and a "very low" value. Instead, the specification must provide some meaningful way to determine what the moderate value is. The only two examples that might help a POSITA determine the actual value are the DVD example and Figure 2. While both examples provide some guidance (e.g., both demonstrate that the moderate power level threshold is closer to the very low value than the maximum or peak value), neither provides any standard as to how to calculate the moderate value in a different context. PI does not, for example, argue that the relationship between the maximum value and the moderate threshold value is always 2:1, as in the DVD example, and indeed the graphic depiction in Figure 2 would undermine such an argument. Thus, at most, the specification teaches that the moderate power level threshold value falls somewhere within a wide range, but it does not provide specific guidance on where in that range the value falls.
Accordingly, the term "moderate power level threshold"/"moderate power level threshold value" is indefinite.
As set forth above, the Court construes the disputed terms as follows:
The parties did not agree to any additional claim constructions in their Amended Joint Claim Construction Statement. See ECF 155 at 2.
The Court has found three terms in three patents indefinite. 35 U.S.C § 282(a) states that "[e]ach claim of a patent (whether in independent, dependent, or multiple dependent form) shall be presumed valid independently of the validity of other claims." Though it is clear that the independent claims in which these terms appear are indefinite, neither party briefed the issue of whether the associated dependent claims are likewise indefinite. That said, each party recognized in the headings of its briefs that each term is associated with certain independent and dependent claims. The Court ORDERS the parties to submit a joint statement
