`
`
`
`IN THE UNITED STATES PATENT & TRADEMARK OFFICE
`______________________
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`VESTAS-AMERICAN WIND TECHNOLOGY, INC.
`and VESTAS WIND SYSTEMS A/S,
`Petitioner,
`v.
`GENERAL ELECTRIC CO.,
`Patent Owner.
`______________________
`Case IPR2018-01015
`U.S. Patent No. 7,629,705
`______________________
`PETITIONER’S REPLY BRIEF IN SUPPORT OF ITS
`PETITION FOR INTER PARTES REVIEW OF
`U.S. PATENT NO 7,629,705
`______________________
`
`
`
`10659019
`
`
`
`
`
`
`
`
`
`TABLE OF CONTENTS
`
`Page
`
`I.
`
`II.
`
`THE DEPENDENT CLAIMS STAND OR FALL WITH THE
`INDEPENDENT CLAIMS. ................................................................... 1
`GROUND 1 TEACHES AND RENDERS OBVIOUS
`CLAIMS 1 AND 7. ................................................................................ 2
`A.
`The Art Teaches the “configuring . . .” Element of Both
`Claims. .......................................................................................... 2
`1.
`Prior Testimony of the Same Expert Patentee
`Hired for This Proceeding Establishes That ZVRT
`Does Not Require Accommodating Changes in
`Grid Phase or Frequency ................................................... 4
`Patentee’s New Position About Grid-Fault
`Characteristics Is Not Supported By Relevant
`Evidence. ............................................................................ 6
`The Deng Systems Teach Zero-Voltage Faults And
`Can Handle “Unpredictable” Post-Fault Grid
`Waveforms. ........................................................................ 9
`(a)
`Zero-Voltage Faults ................................................. 9
`(b)
`“Unpredictable” Post-Fault Waveforms ................ 12
`POSAs Would Have Combined the Static Power
`Converter of Deng ‘988 With the Wind Turbine of
`Janssen, Which Itself Uses a Static Power Converter. .............. 13
`III. GROUND 2 ANTICIPATES CLAIMS 1 AND 7. .............................. 16
`A.
`Teodorescu Does Not Disconnect During Stand-Alone
`Control Mode. ............................................................................ 17
`Teodorescu Does Not Disconnect During Idle Mode. ............... 18
`Teodorescu Concerns Zero-Voltage Grid Faults. ...................... 20
`Teodorescu Does Not Require Division by Zero During
`Zero-Voltage Faults. .................................................................. 21
`- i -
`
`2.
`
`3.
`
`B.
`
`B.
`C.
`D.
`
`10659019
`
`
`
`
`
`
`
`Page
`
`V.
`
`IV. GROUND 3A TEACHES AND RENDERS OBVIOUS
`CLAIMS 1 AND 7. .............................................................................. 21
`EON AND FERC MOTIVATE THE COMBINATIONS OF
`GROUNDS 1 AND 3A-5. .................................................................... 23
`A.
`Janssen and Erdman Disclose Wind Turbines Subject to
`EON and FERC. ......................................................................... 24
`Patentee Cites No Evidence about EON’s or FERC’s
`Applicability to Deng. ................................................................ 25
`Patentee’s Reliance on Distribution Systems Is
`Inapposite. .................................................................................. 26
`VI. CONCLUSION .................................................................................... 28
`
`
`B.
`
`C.
`
`10659019
`
`
`
`- ii -
`
`
`
`
`
`TABLE OF AUTHORITIES
`
`
`
` Page(s)
`
`Cases
`In re Am. Acad. of Sci. Tech Ctr.,
`367 F.3d 1359 (Fed. Cir. 2004) ...................................................................... 9
`Johns Manville Corp. v. Knauf Insulation, Inc.,
`IPR2015-01402, Paper 45 (PTAB Oct. 9, 2016) ............................................ 8
`Masabi Ltd. v. Bytemark, Inc.,
`IPR2017-01449 (PTAB), Paper 38 (Dec. 3, 2017) ........................................ 6
`
`10659019
`
`
`
`- iii -
`
`
`
`
`
`
`
`Upon instituting trial, the Board found a reasonable likelihood that each
`
`of the Challenged Claims—claims 1-8 of the Challenged Patent—are
`
`unpatentable.
`
`Trial afforded Patent Owner (“Patentee”) an opportunity to supplement
`
`the record with evidence of patentability. Passing on that opportunity, Patentee
`
`instead takes positions based entirely on unsubstantiated attorney argument,
`
`misapprehensions of the prior-art references at issue, and direct contradictions
`
`of representations that it and its expert previously made to the Board, to the
`
`Patent Office, and to a federal court.
`
`None of these positions, even if accepted, overcome Petitioner’s prior
`
`art. Through the trial, both parties’ experts have confirmed key factual
`
`underpinnings of Petitioner’s account of unpatentability, which facts stand
`
`unrebutted. The record now before the Board mandates—with even greater
`
`weight than the record at the time of institution—that the Board cancel each of
`
`the Challenged Claims. Petitioner therefore respectfully requests their
`
`cancellation.
`
`I.
`
`The Dependent Claims Stand or Fall with the Independent Claims.
`
`Patentee attempts no specific arguments about claims 2-6 and 8
`
`discussed in Grounds 4 and 5, leaving those claims to stand or fall with
`
`extraordinarily broad independent claims 1 and 7, the only claims substantively
`
`10659019
`
`
`
`- 1 -
`
`
`
`
`
`
`
`discussed in the Response. Paper 20 [hereinafter “POR”], 67. Patentee does
`
`not differentiate claims 1 and 7 in its argument.
`
`Patentee previously disclaimed Claims 10-12 and 15-17 upon receiving
`
`the Petition. Ex. 2001; Paper 6, 50-51. Claims 9, 13, and 14 were cancelled in
`
`earlier proceedings. Paper 2 [hereinafter “Pet.”], 1 n.1.
`
`II. Ground 1 Teaches and Renders Obvious Claims 1 and 7.
`
`Patentee argues that Ground 1 does not teach the “configuring” element
`
`of claim 1 or its claim-7 analog, see POR, 28-36, and that persons of ordinary
`
`skill in the art (“POSAs”) would not have combined the references of Ground
`
`1. Id., 23-28. Both arguments fail for multiple reasons as discussed below.
`
`A. The Art Teaches the “configuring . . .” Element of Both
`
`Claims.
`
`Petitioner has shown with evidence that Deng ‘988 and Deng ‘650 teach
`
`every element of claims 1 and 7—including the “configuring” element of claim
`
`1 and its analog in claim 7, which relate to a wind turbine that remains
`
`“electrically connected to the electric power system during and subsequent to”
`
`a voltage amplitude of the electric power system decreasing to “approximately
`
`zero volts” for a period of time.. E.g., Pet., 17-20, 23-26, 37-40, 44-47.
`
`Patentee disputes whether the art teaches the “configuring” element.
`
`POR, 35-36; see generally id., 28-36. According to Patentee, the art “describes
`
`only a solution for scenarios when its system loses sight of the normally
`- 2 -
`
`10659019
`
`
`
`
`
`
`
`
`
`operating grid voltage” because of a “condition of zero current between two
`
`points in an open circuit.” Id., 28, 30 (emphases omitted). Patentee
`
`misdescribes the art. Moreover, Patentee does not explain any import of this
`
`attempted distinction, which even if true, would be of no consequence.
`
`Instead, “Deng ‘988 teaches a specific synchronization solution that works
`
`even when the grid’s voltage amplitude drops to approximately zero volts.”
`
`Pet., 18.1
`
`Nevertheless, in other parts of the Response not directed to any claim
`
`element, Patentee argues that three-phase zero-voltage faults allegedly result in
`
`“an unpredictable waveform ‘subsequent to’ a zero-voltage event,” and that the
`
`art “simply follows the same stable, consistent voltage phase and frequency
`
`waveform during an open circuit that it synchronizes with during a ‘normal,’
`
`closed circuit condition.” POR, 22-23. To the extent Patentee contends that
`
`Petitioner’s art does not account for this alleged characteristic of zero-voltage
`
`events, extensive prior testimony of the same expert that Patentee hired for this
`
`proceeding directly belies the contention. Apart from the conclusory—and
`
`contradictory—declaration of that expert, Patentee provides no relevant
`
`evidence that could support its allegations about “unpredictable” waveforms.
`
`
`1 All emphasis in quotations is added unless otherwise noted.
`
`10659019
`
`
`
`- 3 -
`
`
`
`
`
`
`
`Moreover, Petitioner’s art clearly teaches how to handle such grid
`
`conditions should they, in fact, occur.
`
`1.
`
`Prior Testimony of the Same Expert Patentee Hired for
`
`This Proceeding Establishes That ZVRT Does Not
`
`Require Accommodating Changes in Grid Phase or
`
`Frequency.
`
`Although it is unclear whether Patentee’s allegations about zero-voltage
`
`events relate to its position about claim construction or the alleged factual
`
`content of Petitioner’s art, neither could be sustained. These allegations are
`
`directly contrary to positions Patentee and its expert here took in prior
`
`proceedings to successfully defend the Challenged Patent.
`
`That expert, Dr. Grady, previously testified that the ZVRT capabilities
`
`covered by the Challenged Claims do not require wind turbines to handle
`
`changes to grid frequency or phase at all. To the contrary, he testified that a
`
`ZVRT-capable wind turbine consistent with the Challenged Claims must not
`
`stray from the pre-fault frequency and phase of the grid if the wind turbine is to
`
`remain connected subsequent to the fault.
`
`For example, Dr. Grady testified under oath that:
`
`• In the “ZVRT” state of the Challenged Patent’s preferred embodiment,
`
`“your rotor is turning, so you tell it to pretend like [the grid]’s there,
`
`10659019
`
`
`
`- 4 -
`
`
`
`
`
`
`
`okay, start where you left off and just keep it lined up. And when it
`
`comes back on, you hope it’s synchronized in.” Ex. 1043, 57:7-58:2.
`
`• “During a zero voltage event (i.e., when the grid voltage cannot be
`
`observed), the PLL state machine facilitates driving the PLL phase angle
`
`to a value that would be in effect if there was no grid disturbance and
`
`further facilitates operating the PLL in a fixed-frequency mode. This
`
`control strategy prevents the PLL from ‘wandering’ during a zero
`
`voltage event, and, therefore, facilitates the wind turbine remaining
`
`connected during and subsequent to the zero voltage event.” Ex. 1041,
`
`¶ 41 (citation omitted).2
`
`• The fixed-frequency mode referenced above—which Dr. Grady
`
`explained “you have to switch over to . . . while the grid is not visible so
`
`that you improve your chances of being synchronized when the grid does
`
`recover,” Ex. 1043, 108:13-109:3—can be implemented without any
`
`special accommodation of the anticipated return of the grid waveform at
`
`all: “[Y]ou could just have a switch that just switched it over and say,
`
`don’t try to estimate [when] the grid frequency is gone, let’s make it
`
`play through as though nothing happened, and when the grid comes
`
`
`2 Patentee echoed this position in a legal brief. See Ex. 1042, 11-12.
`
`10659019
`
`
`
`- 5 -
`
`
`
`
`
`
`
`back you hope it’s still connected; it does not require a state machine.”
`
`Id., 58:20-59:10.
`
`These prior statements directly contradict Dr. Grady’s claims in this
`
`proceeding that the grid phase or frequency can change after a fault and that
`
`ZVRT requires special accommodation of such changes. E.g., Ex. 2012,
`
`¶¶ 14, 33, 44, 45, 46, 47, 48, 89; see Ex. 1040, ¶¶ 10-14. And even absent
`
`these contradictions, Dr. Grady’s declaration in its entirety would have
`
`warranted at most minimal weight, because it almost exclusively parrots the
`
`attorney argument in the Patent Owner Response. See, e.g., Masabi Ltd. v.
`
`Bytemark, Inc., IPR2017-01449 (PTAB), Paper 38 at 50 (Dec. 3, 2017)
`
`(“Merely repeating an argument from the P.O. Response in the declaration
`
`of an expert does not give that argument enhanced probative value.”).
`
`2.
`
`Patentee’s New Position About Grid-Fault
`
`Characteristics Is Not Supported By Relevant Evidence.
`
`Even if Patentee’s new position about alleged “unpredictable” post-fault
`
`grid waveforms could be reconciled with the contradictory testimonies of its
`
`expert, Patentee fails to provide any relevant evidence in support thereof. The
`
`scant materials Patentee cites for that proposition generally discuss only the
`
`tendency of a near-fault generator’s rotor to change its rotational frequency
`
`and phase, not the electrical waveform of the grid itself—and these mechanical
`
`10659019
`
`
`
`- 6 -
`
`
`
`
`
`
`
`generator-side issues had already been solved by prior-art LVRT wind-turbine
`
`systems that Patentee emphatically distinguishes from ZVRT.
`
`For example, Patentee cites to EON’s statement that “[o]nce a fault in
`
`the ENE network is cleared or after a three-phase automatic reclosure, the
`
`operator of a generating unit must expect that the voltages in the networks of
`
`ENE and at the network connection of the connectee could be asynchronous.”
`
`Ex. 1007, 17 (quoted in POR, 8). This describes the behavior of a generator
`
`near a zero-voltage fault, as Patentee’s own expert explains:
`
`If a generator is close to the faulted point, its voltage
`
`drops during the fault, power flow out of the
`
`generator drops, and the rotor speeds up due to power
`
`imbalance. The worst case is when the voltage
`
`output is zero, because all mechanical output power
`
`goes towards increasing [rotor] inertia via speed.
`
`This condition is referred to as asynchronous voltage
`
`and EON expressly warns that this condition can
`
`exist once the fault is cleared.
`
`Ex. 2012, ¶ 14 (citing Ex. 1007, 17). Patentee’s similar assertion elsewhere
`
`that an “aspect of the post-fault scenario is that both generator speeds and
`
`voltage phase angles will differ—often dramatically—from their pre-fault
`
`10659019
`
`
`
`- 7 -
`
`
`
`
`
`
`
`values” similarly relates on its face only to generator, not grid, behavior. POR,
`
`7-8; Ex. 2012, ¶ 14. See Ex. 1040, ¶¶ 15-18.
`
`Undermining the import of these general assertions about generators,
`
`Patentee’s expert conceded in cross-examination that during the 0.15-second
`
`ride-through period required by EON, “there’s not going to be much
`
`speedup . . . . So I don’t think speedup is going to be a very big problem in
`
`the time frame we’re talking about.” Ex. 1046, 94:22-95:10; Ex. 1040, ¶ 20.
`
`And, as discussed in the Petition and supporting materials, potential
`
`asynchronicity due to generator-side rotor speed-up of this sort had already
`
`been solved in wind turbines by LVRT-capable prior art such as Janssen and
`
`Erdman, using wind-turbine-specific techniques such as blade pitch control that
`
`counteract speed-up and others. See, e.g., Pet., 6; Ex. 1003, ¶¶ 66-67, 79, 83;
`
`Ex. 1040, ¶¶ 19, 21.
`
`Patentee cites to its expert’s ambiguous, conclusory, and unexplained
`
`discussion of “[r]eal synchrophasor measurements” (which does not cite or
`
`provide even a single such measurement). Ex. 2012 ¶ 47. Such
`
`unsubstantiated testimony should not be credited. Ex. 1040, ¶ 22; see Johns
`
`Manville Corp. v. Knauf Insulation, Inc., IPR2015-01402, Paper 45 at 16
`
`(PTAB Oct. 9, 2016) (“Dr. Smith essentially states what Patent Owner argues
`
`in the Response . . . . We give such conclusory, unsupported assertions by
`
`10659019
`
`
`
`- 8 -
`
`
`
`
`
`
`
`Patent Owner’s expert little weight.”) (citing In re Am. Acad. of Sci. Tech Ctr.,
`
`367 F.3d 1359, 1368 (Fed. Cir. 2004)).
`
`3.
`
`The Deng Systems Teach Zero-Voltage Faults And Can
`
`Handle “Unpredictable” Post-Fault Grid Waveforms.
`
`Premised on its unproven allegations about the behavior of the grid
`
`during and after three-phase zero-voltage faults, Patentee suggests that the art
`
`cannot handle such alleged behavior because it does not teach grid faults and
`
`because “Deng’s system simply follows the same stable, consistent voltage
`
`phase and frequency waveform during an open circuit condition.” POR, 22-23.
`
`Patentee’s assertion about both Deng references is incorrect. Those
`
`references specifically discuss zero-voltage faults, and “ride through” those
`
`faults with digital controllers that sample the grid thousands of times per
`
`second. Once these systems detect clearance of fault conditions, their output
`
`waveforms conform with those of the restored grid within one cycle of the grid
`
`waveform—even if that waveform has changed—as explained below.
`
`(a) Zero-Voltage Faults
`
`Patentee incorrectly argues that Deng ‘988 addresses only “open circuit”
`
`conditions and “only a solution for scenarios when its system loses sight of the
`
`normally operating grid voltage” rather than scenarios with the grid voltage
`
`itself decreasing to approximately zero volts. POR, 28, 30 (emphasis omitted).
`
`This argument disregards Deng ‘988’s express teaching about “grid voltage
`- 9 -
`
`10659019
`
`
`
`
`
`
`
`
`
`fault[s]” and “abnormal operating conditions” wherein “grid voltage is lost.”
`
`E.g., Ex. 1009, 6:20, 1:31-35; see also Ex. 1046, 135:3-14 (Patentee’s expert
`
`referring to a grid fault as “abnormal conditions”). Ex. 1040, ¶¶ 23-24.
`
`Indeed, Deng ‘988 expressly distinguishes actual grid voltage (“grid
`
`voltages Va_g, Vb_g, and Vc_g”)—which is what goes to zero in Deng ‘988—
`
`from its own measured samples of that voltage (“Va_g_k, Vb_g_k, and
`
`Vc_g_k are the sampled three-phase grid voltages of Va_g, Vb_g, Vc_g at the
`
`time step ‘k’“). Ex. 1009, 3:58-59, 4:7-9; see also id., 3:9-18. For example,
`
`Figure 5 shows a scenario with grid voltage Va_g—not the system’s
`
`measurement Va_g_k thereof—decreasing to 0, i.e., actual zero-voltage-fault
`
`conditions on the grid:
`
`Ex. 1040, ¶¶ 26-29. Thus, where Patentee incorrectly argues that “the grid
`
`voltages described in Deng ‘988 are measured from the perspective of Deng’s
`
`
`
`10659019
`
`
`
`- 10 -
`
`
`
`
`
`
`
`system,” POR 31, Patentee overlooks Deng ‘988’s express distinction between
`
`voltage measurements and actual grid behavior.
`
`Further, as can be seen in Figure 5’s depiction of grid voltage Va_g, the
`
`voltage amplitude continuously decreases before the fault voltage period and
`
`continuously increases thereafter. This waveform depicts a zero-volt grid fault
`
`condition; it is inconsistent with the abrupt “open circuit” that Patentee ascribes
`
`to Deng ‘988. Ex. 1040, ¶¶ 30-32.
`
`Finally, Patentee dwells on Deng ‘988’s example of “switch ‘bounce,’“
`
`POR, 30, but that single example does not limit the reference’s broader
`
`discussion of zero-voltage grid faults. And Deng ‘988 does not define “switch
`
`bounce” in relation to circuit breaker 16, a relationship Patentee appears to
`
`assume (but never proves). Ex. 1040, ¶¶ 23-25.
`
`In fact, Patentee interprets “switch bounce” using an inapposite
`
`definition taken from a chapter about sensors in a book on embedded
`
`microprocessors (Ex. 2019). POSAs would not understand that book to
`
`correctly define “switch bounce” in this context; for example, the “switch”
`
`referenced in Ball (Ex. 2019) is not a high-power switch used to control
`
`electrical current, but a mechanical switch used to drive a low-power input to a
`
`microprocessor. See id., 88-89; Ex. 1040, ¶ 25. Patentee’s reproduction of
`
`Ball’s Figure 3.30 on page 30 of the Response removes most of the figure,
`
`concealing this context:
`
`10659019
`
`
`
`- 11 -
`
`
`
`
`
`
`
`
`Ball does not discuss electric conversion systems or grids, which have different
`
`electrical behavior than microprocessor inputs. This context confirms Ball’s
`
`irrelevance here.
`
`(b)
`
`“Unpredictable” Post-Fault Waveforms
`
`Because both Deng references generate output waveforms digitally, they
`
`accommodate any changes to the grid waveform within one cycle thereof:
`
`“synchronization is guaranteed between an AC power source 100, such as a
`
`SPC, and a power grid 190 . . . . This is because the initial angle θ0 is equal to
`
`the actual grid voltage phase angle γ at the initiation of every cycle.” Ex. 1012,
`
`6:33-36; see also Ex. 1009, 6:4-9; Ex. 1040, ¶¶ 33-37. In other words, once the
`
`Deng systems detect recovery from a fault, their output waveform promptly
`
`10659019
`
`
`
`- 12 -
`
`
`
`
`
`
`
`conforms to the detected frequency and phase of the grid, even if these
`
`characteristics have changed. Ex. 1040, ¶¶ 38-41; see also Ex. 1003, ¶ 95.
`
`Patentee claims that Figure 5 of Deng ‘988 shows no change in the
`
`frequency or phase of the grid and therefore would not be understood to show a
`
`zero-voltage fault. But Figure 5 is an idealized example; it shows “a fault
`
`condition,” not every possible fault condition. Ex. 1009, 2:52-55. Patentee’s
`
`expert testified that an analogous figure in Janssen (Figure 1) describing
`
`exemplary fault voltage levels is “an idealized example, because they have to
`
`give – every fault is like clouds, they’re all different.” Ex. 1046, 62:6-62:14.
`
`POSAs would understand that the “abnormal operating conditions” of the Deng
`
`references could yield a number of actual waveforms and that, as the Deng
`
`references expressly state, their digital controllers can handle the concomitant
`
`grid conditions. Ex. 1040, ¶ 31.
`
`B.
`
`POSAs Would Have Combined the Static Power Converter of
`
`Deng ‘988 With the Wind Turbine of Janssen, Which Itself
`
`Uses a Static Power Converter.
`
`Patentee argues that POSAs would not have applied Deng ‘988 to wind
`
`turbines such as Janssen because “static power converters are mechanically and
`
`operationally very different than large scale electrical machines like wind
`
`turbines or power plants like wind farms.” POR, 24. But Patentee fails to
`
`respond to (let alone rebut) Petitioner’s evidence of the many reasons why
`- 13 -
`
`10659019
`
`
`
`
`
`
`
`
`
`POSAs would in fact have applied Deng ‘988’s teachings about static power
`
`converters to Janssen’s static power converter. E.g., Ex. 1003, ¶¶ 81-103. And
`
`aside from its expert’s conclusory testimony, Patentee does not provide any
`
`evidence to support its own assertion about the alleged differences.
`
`As an initial matter, wind turbines use both a generator (which generates
`
`electricity) and a static power converter (which converts that electricity to
`
`make it suitable for the grid), as Professor Blaabjerg’s annotated Figure 2 of
`
`Janssen shows below. See Ex. 1046, 100:15-18. The term “static” identifies
`
`power converters that do not use rotating elements. See Ex. 1046, 112:13-
`
`113:7, 113:9-114:12; Ex. 2020, 132:12-15; Ex. 1045, 1:36-38. POSAs “would
`
`understand that in order to work with a wind turbine, a static power converter
`
`is necessary.” Ex. 2020, 131:20-132:3, 129:3-13; Ex. 1040, ¶¶ 90-92. Indeed,
`
`Patentee’s own expert admits that Janssen uses a static power converter, Ex.
`
`1046, 111:7-113, and concedes in his declaration more broadly that “wind
`
`turbine generators do include static power conversion components.” Ex. 2012,
`
`¶ 39. These facts belie any claim that static converters or the techniques of
`
`Deng are incompatible with Janssen, let alone all wind turbines subject to
`
`ZVRT requirements as Patentee appears to claim. Ex. 1040, ¶¶ 43-48.
`
`10659019
`
`
`
`- 14 -
`
`
`
`
`
`
`
`
`Patentee’s alleged support for confusing and contrasting these distinct
`
`elements is a single statement in Deng ‘650 (which is not even part of Ground
`
`1 and thus does not support Patentee’s argument here) about generators. POR
`
`24 (citing Ex. 1012, 1:38-42). But Deng ‘650 expressly teaches that it is
`
`tackling a power conversion in power converters used with generators large
`
`and small: “As with the connection of a large power plant to an existing
`
`power grid, when connecting an additional AC power source, such as a static
`
`power converter, to a three-phase power grid, synchronization of the
`
`amplitudes, frequencies, and phase angles . . . is required.” Id., 1:48-54; see
`
`also id., Abstract; Ex. 1040, ¶¶ 49-51. This is precisely the same problem that
`
`the Challenged Patent purports to solve. See Pet., 6-7.
`
`10659019
`
`
`
`- 15 -
`
`
`
`
`
`
`
`Patentee disregards this voluminous evidence showing that POSAs
`
`would have combined Deng ‘988 with Janssen (and Deng ‘650 with Erdman),
`
`and neither Patentee nor its expert identifies even a single feature of either
`
`Deng reference that would make them unsuitable for a wind turbine subject to
`
`grid-code ZVRT requirements. See also Section V. Indeed, the Deng
`
`solutions are applicable to all AC power sources. Pet., 37; Ex. 1003, ¶¶ 148-
`
`150.
`
`III. Ground 2 Anticipates Claims 1 and 7.
`
`Petitioner has shown with evidence that Teodorescu teaches every
`
`element of the Challenged Claims. See, e.g., Pet., 27-36.
`
`Patentee responds that Teodorescu disconnects from the grid during a
`
`fault and that it does not discuss zero-voltage faults, but Patentee does not
`
`provide any evidence to support these allegations absent its expert’s conclusory
`
`testimony. Professor Blaabjerg, the senior author of Teodorescu, confirmed
`
`under cross-examination that the wind turbine of Teodorescu does not
`
`disconnect from the grid in any of its three disclosed modes: “In grid-
`
`connected mode, we are grid connected. In standalone mode, we are grid
`
`connected. In idle mode, we are grid connected, . . . .” Ex. 2020, 217:15-17;
`
`see also id., 218:8-9; 203:13-14; 207:4-5. Patentee’s expert, on the other hand,
`
`directly contradicts his own prior testimony about the crux of Patentee’s
`
`argument.
`
`10659019
`
`
`
`- 16 -
`
`
`
`
`
`
`
`A. Teodorescu Does Not Disconnect During Stand-Alone Control
`
`Mode.
`
`The Board has already rejected Patentee’s lead argument against Ground
`
`2—that the Board should simply discard the Teodorescu reference actually
`
`asserted and analyze instead a different reference, Tirumala. Institution
`
`Decision, Paper 9 [hereinafter “ID”], 59-60. Although Patentee copies nearly
`
`verbatim its previously rejected Tirumala discussion, it offers no new relevant
`
`evidence in support of that analysis. Compare Paper 6, 35-37, with POR, 38-
`
`40. It relies on a single snippet of Professor Blaabjerg’s testimony, which
`
`speaks specifically to Tirumala and offers no insight into Teodorescu. Ex.
`
`2020, 221:17-222:1; Ex. 1040, ¶¶ 53-54.
`
`The mere fact that Teodorescu cites to another paper about a system that
`
`might disconnect does not mean that Teodorescu similarly disconnects.
`
`Patentee never acknowledges the most obvious explanation for the “advantage”
`
`over Tirumala referenced in Teodorescu: Teodorescu does not need the extra
`
`switching hardware because it does not disconnect from the grid when
`
`switching modes. Ex. 1040, ¶ 55.
`
`Patentee also argues that Figure 17(b) shows that Teodorescu’s power
`
`converter “is not yet connected to the grid” during the Stand-Alone Control
`
`Mode (“SACM”) time period -0.02s to 0.00s. POR, 44 (emphasis omitted).
`
`But this assertion is belied by the figure, which shows non-zero currents ia, ib,
`- 17 -
`
`10659019
`
`
`
`
`
`
`
`
`
`and ic during that time period between the converter and the grid, as observed
`
`at nodes iA, iB, and iC in the schematic representation of Teodorescu’s system in
`
`Figure 3. Ex. 2020, 200:4-201:9. Current could not flow, of course, if the
`
`converter was disconnected from the grid, and Patentee never addresses this
`
`fundamental inconsistency between its argument and Figure 17(b). Ex. 1040,
`
`¶¶ 56-57.
`
`B.
`
`Teodorescu Does Not Disconnect During Idle Mode.
`
`Figure 17(b) shows that current continues to flow through Figure 3’s
`
`nodes iA, iB, and iC during the Idle-Mode period of 0.00s to 0.03s, disproving
`
`Patentee’s contention that “when Teodorescu’s system transitions into idle
`
`mode . . . it also controls the grid converter in such a way that current cannot
`
`flow, thus electrically disconnecting the control system from the grid.” POR,
`
`50. Professor Blaabjerg confirmed that the current is non-zero during the Idle-
`
`Mode period as a result of the system’s connection to the grid. Ex. 2020,
`
`204:2-21; Ex. 1040, ¶¶ 58-61. Patentee dismisses this evidence in a footnote of
`
`attorney argument, but offers no evidence of its own beyond an
`
`unsubstantiated, parroting expert assertion. See POR, 50 n.5; Ex. 2012, ¶ 79.
`
`The crux of Patentee’s and its expert’s argument about Idle Mode—that
`
`Teodorescu’s suspension of transistor firing in its converter somehow results in
`
`a disconnection from the grid, see POR, 49-50, Ex. 2012, ¶¶ 78-79—
`
`contradicts previous sworn testimony of that very same expert, who explained
`- 18 -
`
`10659019
`
`
`
`
`
`
`
`
`
`that such suspension is not only within the scope of the Challenged Claims but
`
`a fundamental aspect of the invention and its preferred embodiments.
`
`For example:
`
`Q. . . . Does suspending the firing of the converter
`
`mean that the converter is disconnected from the
`
`grid?
`
`A. No, it’s not disconnected at all. Like I told you,
`
`the way power electronic circuits works is, you open
`
`and close these switches, power electronic switches.
`
`They are not physically opened and closed
`
`electrically, and you are and doing it at a very high
`
`rate. Instead of doing this all the time, you simply
`
`halt for the moment, you know, and you see this big
`
`zero voltage event, it’s only going to last a fraction,
`
`you know, this long, and then when it comes back,
`
`you restart. And that is not disconnecting, it is
`
`simply suspending it.
`
`Ex. 1043, 59:11-25. Dr. Grady also confirmed that the preferred embodiments
`
`of the Challenged Patent “suspend the firing of the converter in just the right
`
`way at the right time to get you through the zero voltage ride through problem”
`
`10659019
`
`
`
`- 19 -
`
`
`
`
`
`
`
`as described in 6:14-19 of the Challenged Patent. Id., 108:13-109:1, 112:3-24;
`
`Ex. 1040, ¶¶ 62-64.
`
`Current can and does flow through the diodes in a static power converter
`
`such as that of Teodorescu. Indeed, such diodes cannot be switched off the
`
`way transistors can. Ex. 2020, 178:25-181:8; Ex. 1040, ¶¶ 65-66. Patentee
`
`attempts to discount this conductivity by arguing the diodes will not conduct
`
`current in a particular scenario, but provides no evidence to support its claim
`
`(e.g., that “the DC voltage labelled VDC in Figure [of Teodorescu] will always
`
`be controlled to be higher than the line-to-line voltage of grid connection points
`
`‘A,’ ‘B’,’ and ‘C,’“ POR, 49)—other than unsubstantiated expert testimony,
`
`which has no probative weight.
`
`C. Teodorescu Concerns Zero-Voltage Grid Faults.
`
`Patentee incorrectly argues that Teodorescu does not contemplate zero-
`
`voltage grid faults. See, e.g., POR, 37. This argument ignores Teodorescu’s
`
`identification of grid failure as its motivation. See, e.g., Ex. 1010, 1323 (“A
`
`novel automatic mode switch method . . . is developed in order to detect grid
`
`failure or recovery and switch the operation mode accordingly.”); id. (“[A]
`
`novel PLL-based method for grid failure detection and automatic mode
`
`switching is proposed where the phase difference between the grid and the
`
`inverter is used to determine grid failure and restoring.”); Ex. 1040, ¶¶ 67-70.
`
`10659019
`
`
`
`- 20 -
`
`
`
`
`
`
`
`Patentee misquotes Professor Blaabjerg’s statement that “the grid is
`
`always there.” POR, 36. Professor Blaabjerg was simply responding to
`
`insinuations that Teodorescu’s wind turbine disconnects physically from the
`
`grid when it experiences a zero-voltage fault. Ex. 1040, ¶ 71.
`
`D. Teodorescu Does Not Require Division by Zero During Zero-
`
`Voltage Faults.
`
`Finally, Patentee argues that “if the grid voltage decreased to zero due to
`
`a zero-voltage event on the grid, then grid voltage components Vx and Vy [of
`
`Equation 5] would each go down to zero,” allegedly resulting in division by
`
`zero. POR, 51. But Equation 5 of Teodorescu is inapposite during fault
`
`conditions or during SACM; it is presented only in connection with Grid-
`
`Connected Control Mode (“GCCM”), when the PLL is used to track the
`
`unfaulted grid—as made clear by the italicized heading on page 1325. When
`
`Teodorescu’s system detects impending fault conditions, it switches away from
`
`GCCM and “the PLL is disabled,” Ex. 1010, 1329. Equation 5 then no longer
`
`models the system for the duration of the fault. There is thus never any risk of
`
`division by zero. Ex. 1040, ¶¶ 72-74.
`
`IV. Ground 3A Teaches and Renders Obvious Claims 1 and 7.
`
`Patentee’s arguments against Ground 3A are the same as those it makes
`
`in Ground 1. For example, Patentee argues that “Deng ‘650’s grid never
`
`decreases to ‘approximately zero volts,’“ POR, 58-64, but, as discussed in the
`- 21 -
`
`10659019
`
`
`
`
`
`
`
`
`
`Petition, the Deng systems do tackle such grid faults. Deng ‘650, for example,
`
`discusses “fault conditions, when the PLL loses synchronization with the grid”
`
`and when “grid voltage is lost.” Ex. 1012, 1:62-2:2. Deng explains
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
