`By: Jeffrey L. Johnson (Jeffrey.johnson@dlapiper.com )
`DLA PIPER LLP (US)
`1000 Louisiana, Suite 2800
`Houston, TX 77002
`Telephone: 713.425.8400
`Facsimile: 713.425.8401
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_______________
`
`MICRO MOTION, INC.
`
`Petitioner
`
`v.
`
`INVENSYS SYSTEMS, INC.
`
`Patent Owner
`
`_______________
`
`Case IPR 2014-00393
`
`U.S. Patent No. 7,571,062
`
`____________________________________________________________
`
`PATENT OWNER RESPONSE PURSUANT TO 37 C.F.R. § 42.120
`
`
`
`
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`Table of Contents
`
`Page
`INTRODUCTION .......................................................................................... 1
`I.
`THE ’062 PATENT ........................................................................................ 2
`II.
`III. CLAIM CONSTRUCTION ........................................................................... 6
`IV. CLAIMS 1 AND 29 ARE NOT ANTICIPATED BY ROMANO ................ 7
`A.
`Romano Does Not Disclose Using Both Sensor Channel
`Signals to Generate the Drive Signal ................................................... 9
`Romano’s Phase Adjustment Does Not Compensate for Time
`Delays Associated With Multiple Components ................................. 21
`CONCLUSION ............................................................................................. 26
`
`B.
`
`V.
`
`
`
`
`
`i
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`Exhibit
`ProQuest Dialog Search Strategy, “Cash in on Flowmeter Innovation:
`there is an abundance of new technology, not only for sophisticated
`uses, but also for fundamental ones,” Marshall, 03/2003
`ISA Show products, 12/01/02
`MICRO MOTION WHITE PAPER, Explaining how two-phase flow
`affects mass flowmeters, 2004
`Press Information, Invensys Digital Coriolis Mass Flowmeter
`Receives Control Engineering Editors’ Choice Award, 01/30/03
`Chemical Engineering; Capitalizing on Cold Chemistry, Cash in On
`Flowmeter Innovation; 03/2003, INVENSYS0129339
`Model RFT9739 Rack-Mount Transmitter Instruction Manual, Version
`3 Transmitters, February 2000,” INVENSYS0111554
`MICRO MOTION WHITE PAPER, The Micro Motion® ELITE®
`Promise, Patten, 2005
`Micro Motion Press Releases, Emerson Announces Next Generation
`Enhancements to Micro Motion® Coriolis Flowmeters, 06/29/06
`IPR2013-00223, 08/15/13 Decision, Paper 9
`CBM2012-00003, 10/25/12 Order, Paper 7
`IPR2012-00006, 5/10/13, Paper 43
`IPR2013-00054, 4/8/13, Paper 12
`(Exhibit served on Petitioner and not filed with PTAB)
`(Exhibit served on Petitioner and not filed with PTAB)
`Declaration of Dr. Jeffrey S. Vipperman
`Datasheet for Philips MUX
`Datasheet for Maxim MUX
`Claim Construction “Memorandum Opinion and Order” (Dkt. No.
`203) from Invensys Systems, Inc. v. Emerson Electric Co., et al., CA.
`No. 6:12-cv-00799-LED (E.D. Tex.)
`Excerpt from Horowitz & Hill, “The Art of Electronics” (2d ed. 1989)
`Excerpt from “Harris’ Shock And Vibration Handbook” (5th ed. 2002)
`Sidman 8/6/2014 Transcript from IPR2014-00170
`
`i
`
`
`
`Ex. No.
`2001
`
`2002
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`2008
`
`2009
`2010
`2011
`2012
`2013
`2014
`2015
`2016
`2017
`2018
`
`2019
`2020
`2021
`
`
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`Sidman 8/7/2014 Transcript from IPR2014-00170
`U.S. Patent No. 6,754,594 (Ex. 1001 in IPR2014-00390)
`Declaration of Dr. Michael D. Sidman (Ex. 1002 in IPR2014-00390)
`
`
`2022
`2023
`2024
`
`
`
`
`
`ii
`
`
`
`
`I.
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`INTRODUCTION
`The Petition in this inter partes review sought cancellation of twelve claims
`
`of U.S. Patent No. 7,571,062 (the “’062 patent”, Ex. 1001). The Board instituted
`
`trial in this on only four claims: claims 1, 19, 40 and 45. In a separate motion,
`
`Patent Owner has canceled claims 40 and 45, leaving only claims 1 and 29 at issue.
`
`The only ground for which trial of claims 1 and 29 was instituted is anticipation
`
`under 35 U.S.C. § 102 by U.S. Patent No. 4,934,196 (“Romano”, Ex. 1006) and
`
`thus only that ground remains at issue.
`
`Claim 1, from which claim 29 depends, includes a requirement to “adjust a
`
`phase of the drive signal to compensate for a time delay associated with
`
`components connected between the sensor and the driver.” The linchpin of
`
`Petitioner’s anticipation challenge to claim 1 over Romano is its expert’s assertion
`
`that signals from both velocity sensor signals are used to generate the drive signal
`
`in a digital drive embodiment of Romano, and thus a certain phase shift applied to
`
`digitized samples of the right velocity sensor signal for measurement purposes– the
`
`only phase shift identified in the Petition or the accompanying declaration -
`
`“propagates through” as a phase shift of the drive signal output in that
`
`embodiment. This assertion is demonstrably false. As discussed in detail below,
`
`and as distinct from measurement operations, Romano actually discloses that only
`
`one sensor signal – the left sensor signal – is used to generate the drive signal
`
`
`
`1
`
`
`
`
`output in the digital drive embodiment. This means that the phase shift applied to
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`the right sensor signal relied on by Petitioner has no effect on the phase of the
`
`drive signal. During cross examination in an earlier inter partes review for a
`
`nearly identical limitation, Petitioner’s expert could not identify any actual
`
`disclosure in Romano of using right sensor signals to generate the drive signal in
`
`this embodiment, and further admitted that it was not true that both velocity sensor
`
`signals are necessarily used to generate the drive signal in the embodiment in
`
`which the phase shift is applied to the right channel sensor. This admission, which
`
`is confirmed by Patent Owner’s expert, is fatal to Petitioner’s anticipation
`
`challenge. Furthermore, even if the right channel sensor signal were to be used to
`
`generate the drive signal in the digital drive embodiment, the phase shift applied to
`
`that signal would have no effect on the drive signal. Accordingly, because the only
`
`assertion regarding any phase adjustment of Romano’s drive signal identified in
`
`either the Petition or the accompanying expert declaration is false, Petitioner has
`
`not demonstrated that Romano anticipates claims 1 or 29.
`
`II. THE ’062 PATENT
`The inventions claimed in the ’062 patent and its family represent nothing
`
`less than landmark developments in the field of Coriolis flowmeters. The ’062
`
`patent family solved problems associated with disturbed flow conditions, including
`
`two-phase and batch flow, that had confounded the industry for decades. Two-
`
`
`
`2
`
`
`
`
`phase flow—which goes by many specific names, including aeration, entrained
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`gas, high gas void fraction, slug flow, stratified flow, bubbles in the flow, etc.—
`
`describes the presence of two phases of material (e.g., liquid and gas) flowing
`
`through the flowtube of a Coriolis flowmeter at the same time. Traditional
`
`Coriolis flowmeters simply failed in the presence of two-phase flow. For this
`
`reason, Petitioner and third-parties described the elusive solution to the two-phase
`
`flow problem as “one of the Holy Grails” of Coriolis flowmeter technology. (Ex.
`
`2001, Rebekkah Marshall, Cash in on Flowmeter Innovation, CHEM. ENG’G 25,
`
`27 (Mar. 2003, p.27).) Petitioner even derided the Patent Owner’s claims that it
`
`had solved the two-phase flow problem—this occurred about one year before
`
`Petitioner unveiled Coriolis flowmeters using the patented technology. (Ex. 2002,
`
`ISA | Show products (December 1, 2002); Ex. 2003 “MICRO MOTION WHITE
`
`PAPER Explaining how two-phase flow affects mass flowmeters,” (2004).)
`
`The ’062 patent teaches a control and measurement system for a digital
`
`flowmeter. A digital flowmeter is one that not only makes measurements
`
`digitally—that is generally a given—but also digitally generates a drive signal to
`
`control conduit oscillation. (Ex. 1001, 1:58-67, 3:4-14.) The ’062 patent discloses
`
`multiple embodiments of the mechanical components of a Coriolis flowmeter and
`
`digital control and measurement system. (Id. 3:15-7:17.) These embodiments
`
`each include (1) a vibratable conduit, (2) a digital controller to measure conduit
`
`
`
`3
`
`
`
`
`vibration and to generate a drive signal to control conduit vibration, (3) a sensor
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`between the conduit and digital circuitry to sense conduit vibration, and (4) a
`
`driver between the digital circuitry and the conduit to drive conduit vibration based
`
`on the drive signal from the digital controller. (Id. 3: 4-14.)
`
`An exemplary embodiment of the primary components of the digital
`
`flowmeter system is illustrated in Figure 1 of the ’062 patent (reproduced below).
`
`(Id. Fig. 1.)
`
`The digital controller above may be comprised of “a processor, a field-
`
`programmable gate array, an ASIC, other programmable logic or gate arrays, or
`
`programmable logic with a processor core.” (Id. 8: 29-33.)
`
`
`
`
`
`4
`
`
`
`The ’062 patent discloses multiple features and capabilities created by the
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`
`
`digital circuitry that allow for precise measurement and control that facilitate
`
`accurate measurement in the presence of two phase flow. For example, the patent
`
`discloses the ability to digitally generate drive signals using multiple different
`
`drive modes. (Ex. 1001, 4:38-58.) The ’062 patent also discloses the ability to use
`
`a proportional-plus-integral (PI) control algorithm to control the motion of the
`
`conduit. (Id. 5:10-12.) The ’062 patent also discloses a control system capable of
`
`“apply[ing] a negative gain to the sensor signal to reduce motion of the conduit.”
`
`(Id. 5:43-45.) And, as recited in claim 1, the digital circuitry of ’062 patent also
`
`discloses the ability to compensate for time delays “associated with components
`
`connected between the sensor and the driver” in the digital flowmeter. (Id. 7:14-
`
`18.) These are just some of the features and capabilities disclosed in the ’062
`
`patent that can be used to achieve precise measurement and control.
`
`The ’062 patent acknowledges that prior flowmeter control mechanisms
`
`lacked sufficient control capability, precision and adaptability to adjust the conduit
`
`drive signal to overcome problems induced by variations in material flow within
`
`the conduit associated with two phase flow. (Ex. 1001, 2:1-6, 47:40-53.)
`
`Processing separate batches of fluid through the flowtube is another instance in
`
`which the digital flowmeter is vastly superior to prior analog flowmeters. (Ex.
`
`1001, 51:5-26.) Thus, the digital flowmeters disclosed in the ’062 patent are
`
`
`
`5
`
`
`
`
`substantially superior to previous analog drive flowmeters for multiple
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`applications.
`
`III. CLAIM CONSTRUCTION
`A claim should be given its broadest reasonable construction in light of the
`
`specification during an inter partes review. See 37 C.F.R. §42.100(b). However,
`
`the broadest reasonable interpretation of the claims must also be consistent with
`
`the interpretation that those skilled in the art would reach. In re Cortright, 165
`
`F.3d 1353, 1358, 49 USPQ2d 1464, 1467 (Fed. Cir. 1999); MPEP 2111.
`
`In the Related Matter identified in the parties’ respective filings to this
`
`Board (Paper No. 2, p.1; Paper No. 8, p.2), Invensys Systems, Inc. v. Emerson
`
`Electric Co., et al., CA. No. 6:12-cv-00799-LED (E.D. Tex.), the court issued a
`
`Memorandum Opinion and Order (Dkt. No. 203) which construed certain claim
`
`terms in U.S. Patent Nos. 7,124,646 (“the ’646 Patent”); 7,136,761 (“the ’761
`
`Patent”); 6,311,136 (“the ’136 Patent”); 7,505,854 (“the ’854 Patent”); 6,754,594
`
`(“the ’594 Patent”); 7,571,062 (“the ’062 Patent”); and 8,000,906 (“the ’906
`
`Patent”) (Ex. 2018). However, none of the terms in that order are relevant to the
`
`terms of the claims at issue herein. With respect to claims 1 and 29 of the ’062
`
`patent, Patent Owner respectfully submits that all terms should be interpreted in
`
`accordance with their plain and ordinary meaning for the purposes of this
`
`proceeding. Ex. 2015 (Vipperman Decl.) ¶ 34.
`
`
`
`6
`
`
`
`
`IV. CLAIMS 1 AND 29 ARE NOT ANTICIPATED BY ROMANO
`Claim 1 includes a requirement to “adjust a phase of the drive signal to
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`compensate for a time delay associated with components connected between the
`
`sensor and the driver.” Ex. 1001 at 55:37-40. In order to show that Romano
`
`disclosed this limitation of claim 1, the Petition first identifies a 2π/128 radian
`
`phase shift applied to the digitized right channel sensor signal samples for
`
`measurement purposes. Petition at 28 (citing Ex. 1002 (Sidman Decl.) at ¶ 196).1
`
`Applying a phase shift to the right channel sensor signal is not, of course, adjusting
`
`a phase of the drive signal. In order to bridge this gap between the disclosure of
`
`Romano and the requirements of claim 1 of the ’062 patent, the Petition asserts that
`
`both the right and the left velocity sensor signals are “used to generate the drive
`
`signal” in the digital drive embodiment and “thus the correction of the phase shift
`
`in the right channel propagates through as a phase shift of the drive signal.”
`
`Petition at 28 (citing Ex. 1002 (Sidman Decl.) at ¶ 198). Neither the Petition nor
`
`Dr. Sidman’s declaration identify anything other than this 2π/128 radian phase
`
`
`1 The portion of Romano relied on by Petitioner and Dr. Sidman refers to a phase
`
`shift of 2P/128 radians. The “P” in this phase shift would be understood by one of
`
`ordinary skill in the art to refer to π. Ex. 2022 (8/7/14 Sidman Tr.) at 68:12-21.
`
`
`
`7
`
`
`
`
`shift applied to the right channel signal as satisfying the “adjust a phase of the
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`drive signal” limitation.
`
`However, the assertions of Petitioner and Dr. Sidman are objectively false.
`
`Ex. 2015 ¶¶ 48-100. Unlike for measurement purposes, Romano does not disclose
`
`that both the right and the left channel signals are used to generate the drive signal
`
`in the digital drive embodiment. Rather, Romano discloses that only one sensor
`
`signal, the left sensor signal, is used to generate the drive signal in this digital drive
`
`embodiment. Thus, the phase shift Romano teaches applying to the right sensor
`
`signal for measurement purposes cannot “adjust the phase of the drive signal,”
`
`since the right channel signal is not used to generate the drive signal.
`
`And, even if the right channel signal were to be used to generate the drive
`
`signal, which is not disclosed in Romano, the 2π/128 radian phase shift applied to
`
`the right channel signal would have no effect on the phase of drive signal. For the
`
`reasons discussed below, the discrete Fourier transform (DFT) Romano discloses
`
`for generating the drive signal is not affected by the phase of the sensor signal;
`
`therefore, the phase shift applied to the right channel signal would essentially be
`
`lost and would not “propagate through” to the drive signal as asserted by Petitioner
`
`and Dr. Sidman.
`
`Further, to the extent that the 2π/128 radian phase shift applied to the right
`
`channel signal compensates for any delay, it is a delay associated with a single
`
`
`
`8
`
`
`
`
`component, the multiplexer, rather than a delay associated with multiple
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`“components” connected between the sensor and the driver as required by claim 1.
`
`Accordingly, Petitioner has failed to establish anticipation of claim 1 for at least
`
`these reasons.
`
`A. Romano Does Not Disclose Using Both Sensor Channel Signals to
`Generate the Drive Signal
`
`In support of its assertion that Romano discloses using both sensor channels
`
`to generate the drive signal in the digital drive embodiment of Fig. 3 of Romano,
`
`the Petition and Dr. Sidman cite three portions of Romano. The Petition and Dr.
`
`Sidman first cite Romano at 24:32-60 (Petition at 28; Ex. 1002 ¶ 198). This
`
`portion of Romano discloses that “a digitally based drive circuit, shown in dotted
`
`lines and formed of latch 388, digital-analog (D/A) converter 390, filter 392 and
`
`amplifier 394 could be used in lieu of analog drive circuit 40 shown in Fig. 4.” Ex.
`
`1006 at 24:32-36; Ex. 2015 ¶ 67. The components referred to in this passage are
`
`shown in the lower right hand corner of Fig. 3, reproduced below (annotations in
`
`red added):
`
`
`
`9
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`microprocessor
`
`
`
`digitally based drive circuit
`
`
`
`The remainder of the cited passage of Romano is set forth below:
`
`As discussed in detail below, microprocessor 330, as shown in Fig. 3,
`calculates the magnitude of a succession of frequency components,
`using the DFT – specifically using equation 4 above, to locate the
`frequency at which the flow tubes resonantly vibrate, i.e., that
`frequency component at which the magnitude of the DFT reaches a
`peak value. Therefore, once this frequency component is known,
`microprocessor 330 can readily generate a quantized sinusoidal
`waveform at exactly this frequency. Specifically, once the frequency
`component is found, microprocessor 330 could easily set the period at
`
`10
`
`
`
`
`
`
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`which a sine look up table (not shown and which can either be
`situated internal to or more likely external to the microprocessor) is
`successively and consecutively indexed, through well known circuitry
`not shown, to produce a continuous series of multi-bit digital values
`that represent this waveform. Each of these values would be applied
`to latch 388 which, in turn, would apply the value to D/A converter
`390. This converter would produce an equivalent analog voltage. This
`analog voltage would then be applied to low pass filter 392 to remove
`unwanted high frequency noise. The resulting filtered value would
`then be amplified by amplifier 394 to an appropriate drive level and
`thereafter routed, via lead 396, to drive coil 180.
`
`Ex. 1006 at 24:36-60; Ex. 2015 ¶ 68. Nothing in this description of the digitally
`
`based drive discloses the use of both sensor channels to generate the drive signal.
`
`In fact, Romano makes clear that only a single sensor signal—the left channel
`
`sensor signal—is used by the microprocessor 330 to generate the drive signal.2 Ex.
`
`Ex. 2015 ¶ 69.
`
`
`
`Romano discusses the software executed by the microprocessor 330 of Fig.
`
`3 above starting at col. 26, line 60. This passage discloses that the microprocessor
`
`2 The reference to Equation 4 in this passage further confirms that only a single
`
`sensor signal is used to calculate the DFT. Ex. 2015 (Vipperman Decl.) ¶ 69.
`
`Equation 4, which appears in Ex. 1006 at 12:7-13 and which gives the magnitude
`
`at a particular frequency, references only a single signal, s(KT). Ex. 2015 ¶ 69.
`
`
`
`11
`
`
`
`
`330 first executes an initialization that involves, among other things, determining
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`the natural frequency at which the flow tubes are vibrating. Ex. 1006 at 26:67-
`
`27:3. This natural frequency is the resonance frequency discussed in the passage
`
`of Romano at 24:36-60 reproduced above. Ex. 2015 (Vipperman Decl.) ¶¶ 71-72.
`
`
`
`The initialization routine 600 is discussed in detail at 28:26-30:45. After
`
`performing a number of diagnostics, the initialization routine first determines the
`
`“fundamental frequency” at which both flow tubes vibrate by calling the DFT
`
`Routine 700 at step 625. Ex. 1006 at 29:13-30; Ex. 2015 ¶ 73. This “fundamental
`
`frequency” is yet another name for the resonance frequency discussed in the
`
`passage of Romano at 24:36-60 reproduced above. Ex. 2015 (Vipperman Decl.) ¶
`
`73. In particular, Romano discloses that only the left sensor channel is used by
`
`the microprocessor 3330 to determine this resonance frequency:
`
`To save processing time, a power spectrum is computed at a fairly
`"coarse" resolution, using the discrete fourier transform, for one of the
`velocity waveforms, illustratively that produced by the left velocity
`sensor. This operation occurs within block 625 which invokes DFT
`Routine 700 which, when executed and as discussed in detail shortly
`in conjunction with FIGS. 7A-7B, samples one of the velocity
`waveforms at a fixed sampling rate and calculates the magnitude
`(squared--for reasons that will become clearer later) of all the
`frequency components, from n=1, . . . , N-1 (here N-1 equals the value
`"63") that comprise the discrete fourier transform of the waveform
`
`
`
`12
`
`
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`produced by the left velocity sensor. The coarse search is preferably
`undertaken with a sampling frequency of 640 Hz which, with "64"
`samples per measurement interval, provides 32 different frequencies
`at a resolution (1/NT) of approximately 5 Hz. Once the coarse power
`spectrum has been computed, execution proceeds, as shown in FIGS.
`6A-6B, to block 630 which determines the maximum value within
`that spectrum and selects the frequency corresponding to that
`maximum (nmax) as being the initial fundamental frequency at which
`both flow tubes resonantly vibrate.
`
`Ex. 1006 at 29:17-40 (emphasis added); Ex. 2015 ¶ 74. This description the
`
`Initialization Routine 600 and the DFT Routine 700 corresponds to Romano’s
`
`disclosure at col. 24, lines 36-42, that “as discussed in detail below,
`
`microprocessor 330 . . . calculates the magnitude of a succession of frequency
`
`components, using the DFT . . . to locate the frequency at which the flow tubes
`
`resonantly vibrate, i.e., that frequency component at which the magnitude of the
`
`DFT reaches a peak value.” Ex. 2015 (Vipperman Decl.) ¶¶ 74-80.
`
`
`
`This disclosure that only the left sensor channel is used to determine the
`
`resonance frequency is confirmed in the detailed discussion of the DFT Routine
`
`700 discussed in Ex. 1006 at 30:46-32:13. Ex. 2015 ¶ 81. The passage discloses
`
`that “this routine samples one of the velocity waveforms at a fixed sample rate and
`
`calculates the magnitude of all frequency components . . . that comprise the
`
`
`
`13
`
`
`
`
`discrete Fourier transform of the waveform produced by the left velocity sensor.”
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`Ex. 1006 at 30:49-54 (emphasis added).
`
`
`
`Thus, the passage of Romano at 24:32-60 relied on by the Petition and Dr.
`
`Sidman as support for the assertion that both sensor channels are used by the
`
`microprocessor 330 to generate the drive signal is clearly incorrect. The
`
`calculation of the DFT to determine the resonance frequency in the DFT Routine
`
`700 of Romano is performed using only a single sensor channel, the left sensor
`
`channel. Ex. 2015 (Vipperman Decl.) ¶ 82. Nothing else in this cited passage
`
`discloses using both sensor channels for generating the drive signal. Accordingly,
`
`the first section of Romano cited by Petitioner and Dr. Sidman does not support the
`
`assertion that Romano discloses that the microprocessor 330 uses both sensor
`
`channels to generate the drive signal; in fact, it proves the opposite: only a single
`
`sensor channel is used to generate the drive signal.
`
`
`
`Although not cited in either the Petition or Dr. Sidman’s declaration, it
`
`appears from questions asked of Patent Owner’s expert, Dr. Vipperman, during
`
`cross examination in IPR2014-00170 that Petitioner may attempt to rely on the
`
`Frequency Tracking Routine 1300 described in Romano at 40:11-41:45, and in
`
`particular the passage of Romano at 40:25-29 (“[f]requency changes are
`
`proportional to the phase difference between the real and imaginary components of
`
`either one of the velocity sensor waveforms measured with respect to the zero
`
`
`
`14
`
`
`
`
`crossing of that waveform.”) as support for the proposition that the right channel
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`sensor signal is used to generate the drive signal and/or that the phase shift applied
`
`to the right channel sensor signal propagates through to the drive signal. This
`
`assertion is incorrect. As explained by Dr. Vipperman, the Frequency Tracking
`
`Routine 1300 has no effect on the generation of the drive signal because (1) the
`
`Frequency Tracking Routine 1300 is not used in the generation of the drive signal,
`
`and (2) even if Frequency Tracking Routine 1300 were to be used in the generation
`
`of the drive signal, and even if the right channel sensor signal were to be used in
`
`the Frequency Tracking Routine 1300, the drive signal would be unaffected by the
`
`2π/128 radian phase shift applied to the right channel sensor signal. Ex. 2015 ¶¶
`
`83-86.
`
`
`
`The second portion of Romano cited by the Petition and Dr. Sidman as
`
`support for the proposition that the “right and left channel signals are both used by
`
`the microprocessor 330 to generate the drive signal” is the passage at 25:26-30.
`
`Petition at 28, citing Ex. 1002 (Sidman Decl.) ¶ 197. This passage of Romano
`
`appears in the description of an analog drive circuit 40 shown in Fig. 4 of Romano.
`
`Ex. 1006 at 25:21-23. This analog drive circuit 40 is what is replaced by the
`
`digitally based drive circuit discussed above. Id. at 24:32-36. The analog drive
`
`circuit 40 does use both the left and the right channel sensors to generate the drive
`
`signal. See ex. 1006 at Fig. 4 (showing left and right channels signals being input
`
`
`
`15
`
`
`
`
`to summer 405). However, as admitted by Dr. Sidman during cross examination in
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`IPR2014-00170, the 2π/128 radian phase shift applied to the digitized right channel
`
`sensor signals in Fig. 3 is not applied to the analog right channel sensor signals in
`
`Fig. 4. Ex. 2022 (8/7/14 Sidman Tr.) at 81:22-82:5; see also Ex. 2015 (Vipperman
`
`Decl.) ¶ 87. Thus, Fig. 4 of Romano does not disclose that the 2π/128 radian phase
`
`shift identified by the Petition and Dr. Sidman propagates through to the drive
`
`signal in the analog drive embodiment. Ex. 2015 ¶ 86.
`
`
`
`Although not identified or explained in either the Petition or Dr. Sidman’s
`
`declaration, during cross examination in IPR2014-00170 Dr. Sidman asserted a
`
`new argument that he believed that the statement in Romano that the digitally
`
`based drive circuit disclosed in Figure 3 could be used in lieu of the analog drive
`
`circuit 40 disclosed in Figure 4, coupled with Dr. Sidman’s belief that Romano
`
`“teaches no alternative” to using both sensor signals to generate the drive signal,
`
`was a disclosure that both sensor signals are used to generate the drive signal. Ex.
`
`2022 (8/7/14 Sidman Tr.) at 82:10-84:4.3
`
`
`3 This testimony that Romano “teaches no alternative” to using both sensor signals
`
`occurred before Dr. Sidman was directed to the description of the DFT Routine
`
`700 discussed above.
`
`
`
`16
`
`
`
`However, Dr. Sidman’s belief that Romano discloses no alternative to using
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`
`
`
`both sensor signals is not sufficient to demonstrate anticipation of claim 1 for at
`
`least two reasons. First, prior art systems (both analog and digital) that used only
`
`one sensor signal to generate a drive signal were known in the art. For example,
`
`the Kalotay patent discloses both an analog circuit (Fig. 3) and a digital circuit
`
`(Fig. 4) that generate a drive signal using only the left sensor signal. Ex. 1008 at
`
`Fig. 3 and 4; see also Ex. 2021 (8/6/14 Sidman Tr.) at 136:5-17 , Ex. 2022 (8/7/14
`
`Sidman Tr.) at 44:11-20, and Ex. 1013 (Vipperman Decl.) ¶¶ 88-89. Thus, even if
`
`it were true (which it is not) that Romano did not disclose any alternative to using
`
`both sensor signals to generate the drive signal, there can be no argument that
`
`generating a drive signal inherently requires using both the left and the right
`
`velocity sensor signals. Second, and more importantly, Dr. Sidman’s assertion is
`
`simply wrong because Romano itself discloses that the DFT Routine 700 executed
`
`by the microprocessor 330 uses only a single sensor channel to generate the drive
`
`signal. Ex. 2015 ¶ 89. Accordingly, this second passage of Romano cited by the
`
`Petition also does not support the assertion that Romano discloses that both sensor
`
`signals are used by the microprocessor 330 to generate the drive signal. In reality,
`
`Dr. Sidman’s testimony is an obviousness argument, and is therefore not sufficient
`
`to support the anticipation ground that is the subject of this inter partes review.
`
`See Net Moneyin, Inc. v. Verisign, Inc., 545 F.3d 1359, 1371 (Fed. Cir. 2008)
`
`
`
`17
`
`
`
`
`([a]nticipation requires the presence in a single prior art disclosure of all elements
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`of a claimed invention arranged as in the claim).
`
`
`
`It is also noteworthy that Dr. Sidman did not allege that Romano disclosed
`
`combining the right and left sensor signals and using them to generate the drive
`
`signal in a digital drive embodiment when he had the chance to do in IPR2014-
`
`00390, as he did during the aforementioned cross examination when he had no
`
`other choice to support his theory that the phase shift applied to the right channel
`
`“propagate through” to the drive signal for claim 1. In IPR2014-00390, Dr.
`
`Sidman alleged that one claim that required only using a single sensor channel in a
`
`digital drive was obvious over Romano alone, but did not allege that a second
`
`claim with the only additional requirement that left and right sensor signals be
`
`combined and used to generate the drive signal in a digital drive was obvious over
`
`Romano alone. Ex. 2015 ¶ 90. The combination of the left and right sensor
`
`signals in the analog embodiment of Fig. 4 is the very disclosure that Dr. Sidman
`
`relies on here in order to support his theory that Romano’s digital drive
`
`embodiment in Fig. 3 also combines the left and right sensor signals to generate the
`
`drive signal. If Dr. Sidman really believed that Romano also disclosed combining
`
`the left and right channel sensor signals to generate the drive signal in the
`
`embodiment of Fig. 3 as he did during cross examination, he would have alleged
`
`that Romano alone made claim 3 of the ‘594 patent obvious because this is the
`
`
`
`18
`
`
`
`
`only limitation in claim 3 not also found in claim 1. That he did not even make an
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`obviousness argument highlights that fact that Romano does not disclose that his
`
`digital drive embodiment of Fig. 3 uses both sensor signals to generate the drive
`
`signal. Id.
`
`
`
`The third passage cited in the Petition and Dr. Sidman’s declaration to
`
`support the assertion that Romano discloses that both sensor signals are used to
`
`generate the drive signal is Fig. 3. Petition at 28, citing Ex. 1002 (Sidman Decl.)
`
`¶ 197. However, during cross examination in IPR2014-00170, Dr. Sidman
`
`admitted that Romano did not “restrict his digital implementation [as shown in Fig.
`
`3] in that way.” Ex. 2022 (8/7/14 Sidman Tr.) at 89:7-21. That is, there is no
`
`disclosure in Romano that the digital implementation in Figure 3 uses both sensor
`
`signals to generate the drive signal. Accordingly, the third passage of Romano
`
`cited by the Petition and Dr. Sidman also fails to support the assertion that Romano
`
`discloses that both sensor signals are used to generate the drive signal. Ex. 2015 ¶
`
`91.
`
`
`
`For all of the foregoing reasons, the assertion in the Petition and Dr.
`
`Sidman’s declaration that both sensor signals are used to generate the drive signal
`
`is wrong. Indeed, Dr. Sidman admitted during cross examination that Romano
`
`does not disclose that the digitally based drive circuit discussed at 24:32-60 uses
`
`both sensor signals:
`
`
`
`19
`
`
`
`
`
`Case IPR 2014-00393
`U.S. Patent No. 7,571,062
`
`
`
`Ex. 2022 (8/7/14 Sidman Tr.) at 84:19-85:1 (emphasis added). And, after he was
`
`pointed to the description of the DFT Routine 700 discussed above, Dr. Sidman
`
`again admitted that it was possible to use only one sensor signal to generate the
`
`drive signal. Ex. 2022 (8/7/14 Sidman Tr.) at 107:8-108:7.
`
`
`
`These admissions are fatal to Petitioner’s assertion that Romano