`571-272-7822
`
`
`Paper 28
`Date: September 18, 2023
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`ECOBEE TECHNOLOGIES, ULC and GOGGLE LLC,
`Petitioner
`v.
`ECOFACTOR, INC.,
`Patent Owner.
`__________
`
`IPR2022-00969 and IPR2022-00983
`Patent 8,596,550
`__________
`
`Record of Oral Hearing
`Held: August 18, 2023
`__________
`
`
`
`Before SCOTT B. HOWARD, PAUL J. KORNICZKY, and
`BRENT M. DOUGAL, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`JUSTIN OLIVER, ESQ.
`of: Venable, LLP
`600 Massachusetts Avenue, NW
`Washington, D.C. 20001
`(202) 721-5423
`joliver@venable.com
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`JONATHAN LINK, ESQ.
`of: Russ August & Kabat
`12424 Wilshire Boulevard
`Los Angeles, California 90025
`(310) 826-7474
`jlink@raklaw.com
`
`
`
`The above-entitled matter came on for hearing August 18, 2023,
`
`commencing at 12:03 p.m. EDT, via Video-conference.
`
`
`
`
`
`2
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`P-R-O-C-E-E-D-I-N-G-S
`
`12:03 p.m.
`
`
`
`JUDGE KORNICZKY: Good day. This is Judge Paul Korniczky.
`
`With me on our panel is Judge Brent Dougal and Judge Scott Howard. I'd
`
`like to get appearances from everybody.
`
`
`
`
`
`Petitioner, would you please introduce yourself?
`
`MR. OLIVER: Thank you, Your Honor. Justin Oliver of Venable on
`
`behalf of the Petitioner, Ecobee Technologies.
`
`
`
`
`
`
`
`JUDGE KORNICZKY: Is anyone else on the line from Petitioner?
`
`MR. OLIVER: No, Your Honor.
`
`JUDGE KORNICZKY: Okay. Patent Owner? Patent Owner, we
`
`can't hear you. I think you're muted.
`
`
`
`
`
`
`
`MR. LINK: There you go. Thank you. Sorry about that.
`
`JUDGE KORNICZKY: Sorry. Yes, thanks, Counsel.
`
`MR. LINK: Yes. Good afternoon. My name is Jonathan Link of the
`
`law firm of Russ August & Kabat on behalf of the Patent Owner, Ecofactor.
`
`And I'm the only one on the line from my side.
`
`
`
`JUDGE KORNICZKY: All right. Thank you. Okay. So -- and our
`
`Court Reporter is Ms. Munoz.
`
`So, this hearing concerns, I guess, four cases. The first case is IPR
`
`2022-00969. The Petitioners are Ecobee Technologies, ULC and Google,
`
`LLC v. the Patent Owner, Ecofactor, Inc. The second case is IPR 2022-
`
`00983, which has the same parties. And the third case is IPR 2023-00355,
`
`which was Google, LLC v. Ecofactor, Inc. The '355 case has been joined
`
`with the '969 case. Then the fourth case is IPR 2023-00356, which is
`
`
`
`
`
`3
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`Google, LLC v. Ecofactor, Inc. And the '356 case has been joined with the
`
`'983 case.
`
`
`
`So, as we go forward today, please make sure that you identify the
`
`slides as you -- you know, we have the slides, we've reviewed all the
`
`materials. Please identify the slides for the Court Reporter, and the parties,
`
`and the Panel.
`
`
`
`And, Mr. Oliver, how much time would you -- so we have 60 minutes,
`
`how much time would you like to reserve for rebuttal?
`
`
`
`
`
`
`
`
`
`MR. OLIVER: Fifteen minutes, Your Honor.
`
`JUDGE KORNICZKY: I'm sorry, you said 15?
`
`MR. OLIVER: Yes.
`
`JUDGE KORNICZKY: Okay. I will try to remind you as we get
`
`closer to the 45 minutes. Let me get my timer going, one second.
`
`Do you gentlemen have any questions before we start?
`
`
`
`
`
`
`
`
`
`MR. OLIVER: No, Your Honor.
`
`MR. LINK: No, Your Honor.
`
`JUDGE KORNICZKY: All right. Mr. Oliver, why don't you start?
`
`MR. OLIVER: Thank you. May it please the Board. The claims of
`
`the '550 patent are directed to thermostatic controllers. The claims generally
`
`recite two separate learning functions that have been known in the field.
`
`First, predicting the rates of change for internal temperatures based on
`
`outside temperatures, and then using that prediction to automate setpoints in
`
`the future. This simply involves accounting for the effects of outside
`
`temperatures on the heating and cooling of internal spaces.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`4
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`And the second main point or learning feature of the claims is
`
`detecting changes to automated settings. And this simply involves that when
`
`a user make a manual adjustment to the thermostat, the thermostat
`
`recognizes and logs the change. This is typically done, as we'll see, to
`
`account for such changes in future programming.
`
`
`
`Both features have been long known in the prior art. For insistence,
`
`the Ehlers reference describes tracking how internal temperatures respond to
`
`various outside temperatures so as to predict that rate of change so that it can
`
`be used in the future in order to set new setpoints that conserve energy.
`
`With respect to tracking changes to setpoints, Ehlers describes doing so in
`
`the idea of learning from the user's preferences so that those preferences can
`
`be replicated in the future so that the user doesn't have to make such
`
`changes.
`
`
`
`In an attempt to avoid these clear teachings, the Patent Owner makes
`
`an off-base argument. Specifically with respect to Ehlers, Patent Owner
`
`argues that Ehlers' clear teachings of tracking internal temperatures with
`
`respect to outside temperatures should be ignored because Ehlers uses the
`
`term thermal gain to describe that rate of change of inside temperatures.
`
`
`
`As we'll discuss, Ehlers clearly discusses, and describes, and
`
`illustrates that the internal temperature change is tracked, and that rate of
`
`change is used. Notably, this Panel in a IPR of a child of the '550 patent,
`
`specifically IPR 20 --
`
`
`
`JUDGE KORNICZKY: Mr. Oliver, this is Judge Korniczky. Can
`
`you lean forward? When you lean back, we start to get an echo. When you
`
`lean forward, it seems to be a little bit better. Or move your --
`
`
`
`
`
`5
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`MR. OLIVER: Certainly, Your Honor.
`
`
`
`JUDGE KORNICZKY: I'm not sure. But we're getting a little echo it
`
`seems when you lean back.
`
`
`
`
`
`(Simultaneous speaking.)
`
`MR. OLIVER: Someone was moving on a floor above, so that might
`
`have been what you heard.
`
`
`
`
`
`JUDGE KORNICZKY: That might have been it. Okay.
`
`MR. OLIVER: In a related IPR of a child of the '550 patent, IPR
`
`2022-00538, this Panel already addressed this issue concerning Ehlers and
`
`the use of the term thermal gain, and correctly found that what Ehlers is
`
`really describing is a rate of change of internal temperatures, exactly what is
`
`required by the claims. For the reasons we'll discuss today, for the reasons
`
`that the Panel has already found with respect to the combination of Ehlers
`
`and Wruck in that prior IPR involving the same parties, and for the reasons
`
`set forth in the briefing, the prior art clearly teaches these features of the
`
`claims.
`
`
`
`Now with that, I'd like to turn to the demonstratives. Looking at Slide
`
`2 of the demonstratives, there are two main IPRs at issue here brought by
`
`Ecobee, not accounting for the joinder IPRs. Both of them have two
`
`grounds, and both of them are against the '550 patent. As the Panel can see,
`
`Ground 1 of each of those IPRs involves a combination the Ehlers and
`
`Wruck. One with respect to Claims 1 through 16, and the other with respect
`
`to Claims 17 through 23.
`
`
`
`One of the IPRs has a second ground that further combines harder
`
`reference, and we'll discuss that. And the other one for its second ground
`
`
`
`
`
`6
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`has the combination of Ols, Boait, and Wruck, which we'll also discuss. But
`
`before getting into the teachings, and I'll focus on Ehlers and Wruck as it
`
`seems to the primary dispute between the parties, I'd like to go over the
`
`claim language that's at issue with the specific limitations that are at issue so
`
`we have some context for our discussion.
`
`
`
`Slides 3 through 5 of Petitioner's demonstrative set forth the
`
`independent claims of the '550 patent, and highlighted for the convenience
`
`of the Panel are the claim limitations at issue. They include, with respect to
`
`Slide 3, Element 1C predicting a rate of change of temperatures inside the
`
`structure in response to at least changes in outside temperature. That is sort
`
`of the learning aspect of the prediction of rates of changes.
`
`
`
`And then that information is used in the next disputed claim term,
`
`which is shown with respect to 1D of Claim 1, calculating scheduled
`
`programming of the thermostat controller for one or more times based on
`
`that predicted rate of change, the schedule programming comprising at least
`
`a first automated setpoint. Simply automating a setpoint based on that use of
`
`the predicted rate of change.
`
`
`
`And then the final term that's in dispute is highlighted with respect to
`
`1E on Slide 3, and that is generating a difference value based on comparing
`
`an actual setpoint at a first time for said thermostatic controller to the first
`
`automated setpoint. Again, that's simply learning when a user makes a
`
`change, there is a difference value recited in Claim 1 that's not required by
`
`all of the claims, other claims just require the comparison of the actual
`
`setpoint to the automated setpoint.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`7
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`Now with that context, I'd like to give a quick overview of the '550
`
`patent looking at Slide 7 of the demonstratives. Slide 7 sets forth Figure 6B
`
`of the '550 patent, and what we see there are some light areas and some
`
`shaded areas that simply represent when the HVAC system is on versus
`
`when it is cycling off. And while thermostatic controllers tend -- or can be
`
`smart systems, the actual HVAC components are rather dumb components.
`
`They either blow cold air or they don't blow cold air. And a setpoint is
`
`maintained simply by adjusting the duty cycle, how much time the system is
`
`blowing cold air, for instance, versus how much time it's not. And that's
`
`shown on Figure 6B.
`
`
`
`And also shown on 6B is the learning function that will replicate
`
`exactly what we'll see in Ehlers, and that is from around noon to around 6:00
`
`or 7:00 p.m. the system has shut off and the system learns what happens to
`
`the internal temperature in response to the outside temperature. If it's warm
`
`outside, the inside temperature will also rise. And that rate of change is
`
`learned.
`
`
`
`And I'll note specifically the highlighted language from the '550
`
`patent at the bottom-left side of Slide 7. "Because the server will also log
`
`these inputs against other inputs including time of day, humidity, et cetera,
`
`the server will be able to predict at any given time on any given day the rate
`
`at which the inside temperature should change for given inside and outside
`
`temperatures." Thus for a given inside temperature and a given outside
`
`temperature, what's going to be the predictive rate of change? And as we'll
`
`see, and turning to Ehlers, that's actually what Ehlers describes.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`8
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`In that regard, I direct the Panel's attention to Slide 10 of the
`
`demonstratives. Again, the main argument that's been presented is that
`
`because Ehlers uses the term thermal gain, thermal gain should be
`
`understood to be energy absorption and not rate of change for internal
`
`temperatures. But that's simply not --
`
`
`
`JUDGE HOWARD: This is Judge Howard. I just have a question
`
`before we get into all this on thermal gain again. Since we've heard it
`
`multiple times, and as you've pointed out, we've made decisions on it, are
`
`those decisions binding on us? Should we just say we've already determined
`
`what thermal gain is in Ehlers and move on to the rest, find issue preclusion?
`
`Or do we need to go through this again and look at the record?
`
`
`
`MR. OLIVER: Your Honor, I don't believe you need to go through it
`
`again, because I believe issue preclusion applies. I think it might be helpful,
`
`and we were going to request it subsequent to this oral hearing, perhaps very
`
`short briefing from both sides on issue preclusion. But issue preclusion does
`
`pertain to IPRs and collateral -- whether it's called collateral estoppel or
`
`issue preclusion, in cases such as past licensing, 924 F.3rd 1243.
`
`
`
`And in issue preclusion and collateral estoppel, what is required is that
`
`the issue be the same. It doesn't have to be the same exact claims. It doesn’t
`
`have to be the overall same finding. What is required that the issues be the
`
`same, that they be relevant to the final decision, and that they be actually
`
`adjudicated such that the estopped party has had a full and fair opportunity.
`
`
`
`Here, we have a lot of the same issues that have been decided. That is
`
`what Ehlers teaches with respect to rates of change, and how its thermal gain
`
`is used as well as Wruck's use of the teaching of a difference value. Both of
`
`
`
`
`
`9
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`those are issues that were relevant to earlier cases, and because those earlier
`
`cases were adjudicated on those grounds, we would believe that collateral
`
`estoppel does apply.
`
`
`
`So I hope that answers your questions, and we're happy to provide
`
`additional briefing on that if the Panel believes that that would be helpful.
`
`
`
`JUDGE KORNICZKY: Counsel, this is Judge Korniczky. Have you
`
`discussed this with Mr. Link?
`
`
`
`MR. OLIVER: I have not discussed it with Mr. Link in advance. If
`
`we do make such a request, we'll do it through the normal practice of
`
`meeting and conferring and making a proposal in that regard.
`
`
`
`JUDGE KORNICZKY: If you do want a briefing, how much -- how
`
`many pages do you need?
`
`
`
`MR. OLIVER: I would think five or so pages would be more than
`
`enough, and probably even four pages if that would be helpful.
`
`
`
`
`
`JUDGE KORNICZKY: All right. Thank you.
`
`MR. OLIVER: As Your Honor pointed out, I know that thermal gain
`
`has been discussed many times before. I will give a, if it's okay with the
`
`Board, a synopsis of that just so the record in this particular case is clear on
`
`the issue. And I think the point I want to make mainly on this is there
`
`doesn't seem to be much of a dispute of what Ehlers actually teaches as far
`
`as changes in internal temperature relative to the changes in external
`
`temperature.
`
`
`
`And I'll note on Slide 10 of our demonstratives, we can sort of see that
`
`in a quote from Patent Owner's own Patent Owner response, which is
`
`provided at the bottom. And in discussing what we'll discuss, Figure 3D of
`
`
`
`
`
`10
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`Ehlers, Patent Owner had this to say as shown by the highlighting, "The
`
`lines appear to reflect a temperature rather than rates of energy increase." So
`
`there doesn't even seem to be a dispute from Patent Owner that what seems
`
`to be shown in Ehlers is a rate of change of temperatures and not a rate of
`
`change of energy absorption. Thus what is actually being shown is what is
`
`relevant to the claims at issue.
`
`
`
`JUDGE KORNICZKY: Counsel, this is Judge Korniczky. What
`
`figures was this quote referring to?
`
`
`
`
`
`
`
`MR. OLIVER: I believe this was referring to Figure 3D.
`
`JUDGE KORNICZKY: D? Okay. Thank you.
`
`MR. OLIVER: Yep. And we can see Figure 3D on the next slide,
`
`Slide 11 of the demonstratives. And as this Panel has heard for in related
`
`cases, and I'll be brief on this, what Ehlers shows is for different internal
`
`temperatures or internal setpoints, which are the starting temperature at
`
`which point the system was turned off, for instance, 72 or 76 degrees.
`
`
`
`The system learns from the rate of change of the indoor temperatures
`
`relative to different outside temperatures, 77, 90, 99. So if it's 90 degrees,
`
`we'll know it'll be one rate of change inside when the original setpoint was
`
`72. And if it heats up to 99 outside, we'll know that the rate of change will
`
`be different. So as the outside temperature changes, the predicted rate inside
`
`also changes.
`
`
`
`And with respect to that, I'd note on Slide 12, I won't go through this
`
`in detail, but there are quotes from Patent Owner's own expert concerning
`
`that idea that the indoor setpoint is that starting internal temperature, and that
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`11
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`is what is being shown there is the increase in internal temperature over time
`
`or time over temperature.
`
`
`
`And really what Figure 3D involves in Ehlers is the first part of the
`
`claim, the learning part, actually using information to learn what the
`
`predicted rate of change will be based on changes in outside temperature.
`
`And that learning is then used in what comes next. And what comes next
`
`can be seen with respect to for instance, Slide 15, and that is Figure 3E and
`
`3G show that once that information has been learned so that we know what
`
`the predicted rate of change will be, it will then be used to set automated
`
`setpoints in the future in order to conserve energy.
`
`
`
`I know the Patent Owner has made the argument that what are shown
`
`in Figures 3E and 3G are continuous changes in temperature over a 24-hour
`
`period. That doesn't make sense from what is described in Ehlers and what's
`
`been admitted to in cites that have been made with respect to admissions
`
`from the Patent Owner's own expert in our briefing. But we have one cite
`
`there on Slide 15, the third bullet point down where Ehlers explains that for
`
`3E and 3G, the setpoint, meaning the internal temperature, was fixed for the
`
`24-hour period.
`
`
`
`That's more true for 3E. As we'll see for 3G, there was ultimately a
`
`change. But the change or the inside temperature was maintained at a
`
`consistent point, and what is changed is the HVAC run time, how often the
`
`cold air has to come on and blow in order to maintain that setpoint.
`
`
`
`And we can see this from Slide 16 of the demonstratives, which
`
`shows Figure 3E. On the right, we have the HVAC run time, which is how
`
`often the system has to cycle on to blow cold air. That run time goes up in
`
`
`
`
`
`12
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`the middle of the day because it's the hottest part of the day. It goes down
`
`late in the evening because that's the coolest part of the day.
`
`
`
`And also, on the right-hand side with respect to the Y axis, we see the
`
`thermal gain rate. That is the information that was learned from the process
`
`or steps shown with respect to Figure 3D, meaning we've learned that for a
`
`given outside temperature, and a given internal temperature, the expected or
`
`predicted rate of increase of the internal temperature would be 1 degree per
`
`hour, or 2 degrees per hour, or 3 degrees per hour. And that is used to
`
`determine how to affect the setpoint to conserve energy. And that
`
`conservation of energy and the automated setpoint is shown with respect to
`
`Figure 3G, and I'll direct the Panel's attention to Slide 19 on that point.
`
`
`
`As explained in Ehlers, and I quote there from Slide 19, "System 3.08
`
`uses the learned thermal gain characteristics of the site 1.04 along with the
`
`customer selected allowable temperature variation range to maintain a flat
`
`level of demand and consumption." And that flat level of demand and
`
`consumption is what's shown in Figure 3G, which is shown on the next
`
`slide, Slide 20 of the demonstratives.
`
`
`
`And as this panel has probably heard before, what Ehlers does is it
`
`wants to conserve energy, and it conserves energy by not running the AC or
`
`blowing cold air that often. And it prefers, in this one example, to keep the
`
`run rate below 33 percent. As the outside temperature increases, there is a
`
`pressure put on the internal temperature, and it's harder to keep the inside
`
`space cool.
`
`
`
`Once that 33 percent run rate is reached and the system doesn't want
`
`to go above that, the way for it to account for that and keep the run rate at 33
`
`
`
`
`
`13
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`percent is to adjust the setpoint, and adjust the setpoint up in order to
`
`maintain that 33 percent. And it makes that calculation based on the
`
`predicted rate of change that is determined from the analysis associated with
`
`Figure 3D.
`
`
`
`If it knows that the inside temperature rises at a certain rate when the
`
`system is off, and just for context, HVAC systems use what's called a dead
`
`band, at least that's the terminology that the Patent Owner's expert uses, Dr.
`
`Palmer. Petitioner's expert Dr. Auslander calls that hysteresis. In any event,
`
`it's the same principle, and that is HVAC systems will maintain a setpoint of,
`
`for instance, 72 degrees by allowing the -- allowing the internal temperature
`
`to raise to perhaps 73, blow cold air until the temperature comes down to 71.
`
`Shut off. Allow the temperature to rise back to 73. And repeating that
`
`process. And that's part of that duty cycle.
`
`
`
`So knowing what the predicted rate of change is for a given inside and
`
`outside temperature, the system, in accordance with Figure 3G, can find out
`
`what offset how much higher the setpoint needs to be adjusted in order to
`
`maintain that conservation of the 33 percent run rate. And that's essentially
`
`what's shown. So Figure 3D we have the learning of the predicted rate of
`
`change, Figure 3G we have the application of the predicted rate of change.
`
`
`
`I'll note that one of the arguments made by Patent Owner is that
`
`because Ehlers' system allows the user to affect our boundaries of the
`
`automated setpoint. For instance, if the system can adjust temperatures
`
`upwards automatically from 72 to 73, and then to 74, and then to 75, the
`
`system may set boundaries of, for instance, 4 degrees such that the system is
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`14
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`allowed to offset the setpoint so long as it's not more than 4 degrees off set
`
`in one way or the other.
`
`
`
`Patent Owner's argument appears to be that because there are those
`
`sort of upper and lower guardrails, that anything that the system calculates in
`
`between those guardrails shouldn't be counted as automated. But that's
`
`simply not the case, and it's not the case based on statements in Ehlers, and
`
`it's not the case based on admissions by Patent Owner's own expert.
`
`
`
`In that regard, I would direct the Panel's attention to Slide 23 of the
`
`demonstratives, and we can see this automation from both quotes, but I'll
`
`focus on the left quote on Slide 23. And as is explained there, "The system
`
`3.08 using its computed thermal gain rate and the corresponding HVAC
`
`cycle run rate time projection computes the required effective setpoint offset
`
`needed to keep the HVAC cycle run time at the specified trigger level."
`
`There's a similar quote on the right, but the point is the same, and that is that
`
`this is a automatic computation where the system is determining what the
`
`offset temperature will be.
`
`
`
`And there doesn't seem to be much dispute from Patent Owner's own
`
`expert on this. We can see that from quotes on Slide 24 for instance. On the
`
`right-hand side, Patent Owner's own expert admitted, "Right, it's calculating
`
`an offset from the existing setpoint." He made a similar admission on the
`
`right-hand side. "The system may adjust the temperatures setpoint from that
`
`specified by the user." And there are additional quotes from the Patent
`
`Owner's expert on Slide 25 of the demonstratives.
`
`
`
`So with respect to that specific argument made by Patent Owner that
`
`the offsets are not automated because there are upper and lower guardrails
`
`
`
`
`
`15
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`beyond which the system can't go is simply not the case because within
`
`those boundaries, within those limits, it is clear and undisputed that Ehlers'
`
`system is automating or selecting those setpoints. The user had it at 72, the
`
`system might change it to 73. As it gets hotter outside, the system might
`
`have to change it to 74. And so on until eventually it hits those guardrails.
`
`
`
`I'll also note, looking at Slide 26, that this is not the only example in
`
`Ehlers of using these predicated rates of change to automate setpoints.
`
`Another example in Ehlers is what's called ramping which is cited to in the
`
`original petition and discussed a little bit further in the reply brief, in
`
`particular in the original expert declaration 1002 and in the reply expert
`
`declaration 1023 at Paragraph 28.
`
`
`
`For clarity, ramping is essentially selecting the rate of change between
`
`two different program modes. A user might have a occupied program mode
`
`and an unoccupied program mode such as that when the user goes to work
`
`for the day, they don't want to pay for cooling the house if no one's going to
`
`be home. So the system might have a 78 or some higher setpoint during that
`
`unoccupied mode, maybe from 8:30 a.m. to 6:00 p.m. But when 6:00 p.m.
`
`comes, the system is intended to go back down to the original setpoint for
`
`the user which would be potentially 72 degrees.
`
`
`
`Ramping refers to picking a sort of number of points along a group
`
`from one of those programs to another such that it's not just turning a system
`
`on full blast to get from 78 to 72, but picking a more even ramp such that
`
`there are intermediate setpoints from one to the other. And that is described
`
`in Ehlers. There's some examples of that discussion in the bottom left
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`16
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`quotes from Slide 26. But this is just another example of how Ehlers uses
`
`predicted rates of change in order to automate the setpoint program.
`
`
`
`With that said, Ehlers describes both the learning part, the predicting
`
`the rates of change, and it describes the selecting automated setpoints based
`
`on those predicted rates of change. The issues that were addressed by the
`
`Patent Owner or raised by the Patent Owner with respect to those arguments
`
`in the Patent Owner's response.
`
`
`
`The final issue raised by Patent Owner in response to a combination
`
`of Ehlers and Wruck has to do with the comparison of actual setpoints to
`
`automated setpoints. This typically happens in the context of manual
`
`adjustments by the user. So the actual setpoint is the manual adjustment that
`
`the user has made.
`
`
`
`Looking at Slide 28, as I mentioned, Claim 1 does this by selecting it
`
`or calculating a difference value. The difference value is the difference
`
`between the actual setpoint and the automated setpoint. Users, when they're
`
`uncomfortable, will adjust the thermostat to make it warmer or more colder
`
`to suit their needs, and that is the manual adjustment.
`
`
`
`All of the prior art we've been citing to, Ehlers, what we'll talk about
`
`later, Ols, they track this user behavior so as to see what changes the user is
`
`making with the intent of replicating those user changes in the future freeing
`
`the user from having to make those changes on their own. To do that, they
`
`compare the changed value, the actual setpoint to the automated setpoint
`
`what was there --
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`
`
`
`
`17
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`JUDGE KORNICZKY: Counsel, this is Judge Korniczky. So I
`
`understand that your position is that Ehlers compares the actual setpoint and
`
`the automated setpoint, right?
`
`
`
`
`
`MR. OLIVER: Correct.
`
`JUDGE KORNICZKY: And you say that if Ehlers doesn't do that,
`
`then you rely on Wruck?
`
`
`
`
`
`MR. OLIVER: Yes. Yes, that's correct.
`
`JUDGE KORNICZKY: So where does Ehlers calculate or generate
`
`the difference value?
`
`
`
`MR. OLIVER: So that's the -- that's what Wruck is relied upon for.
`
`What Ehlers describes is learning from users' adjustments. What a
`
`difference value is is one mathematical operation for determining whether a
`
`change has been made.
`
`
`
`JUDGE KORNICZKY: No, I understand. But -- so in Claim, I
`
`believe it was 17, it just requires that we have to compare the actual setpoint
`
`and the automated setpoint. Is that right?
`
`
`
`
`
`MR. OLIVER: Correct.
`
`JUDGE KORNICZKY: It doesn't require that you generate the
`
`difference value?
`
`
`
`
`
`MR. OLIVER: Correct.
`
`JUDGE KORNICZKY: So -- and for Claim 17 at least, you're only
`
`relying on Ehlers. You don't really need Wruck.
`
`
`
`MR. OLIVER: Wruck's not needed for Claim 17.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`
`
`
`
`18
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`JUDGE KORNICZKY: So -- I'm sorry. So let's go back then to this
`
`Claim 1. Where does Ehlers actually calculate or generate the difference
`
`value?
`
`
`
`MR. OLIVER: Ehlers does not specifically describe calculating a
`
`difference value. Ehlers describes what is the underlying principle and that
`
`is detecting and determining a change. It does not --
`
`
`
`
`
`
`
`
`
`(Simultaneous speaking.)
`
`JUDGE KORNICZKY: Right. Comparing the two values.
`
`MR. OLIVER: What's that?
`
`JUDGE KORNICZKY: It compares the two values, the automated
`
`setpoint and the actual setpoint.
`
`
`
`
`
`MR. OLIVER: Correct, Your Honor. It does not -- sorry. Go ahead.
`
`JUDGE KORNICZKY: Okay. Go ahead. So then now where does
`
`Wruck calculate or generate this difference value?
`
`
`
`MR. OLIVER: So with respect to Wruck, we can see that on Slide 30
`
`of the demonstratives -- and before getting to that, I'll give some context.
`
`Cited in the Petitioner are various points to Wruck that give a little bit of
`
`background of what Wruck is doing. And Wruck is about overrides to the
`
`thermostats and how to track those overrides, how to account those
`
`overrides, and how to make sure the system knows that the overrides took
`
`place. Some of the sections of Wruck cited in the petition on this point
`
`include Paragraph 14 of Wruck which discusses override setpoints.
`
`
`
`JUDGE KORNICZKY: Hold on, one second. Let me get there
`
`because I have reviewed these paragraphs, and -- hold on, let me get there.
`
`So you said Paragraph 14?
`
`
`
`
`
`19
`
`
`
`
`
`IPR 2022-00969 and IPR 2022-00983
`Patent 8,596,550
`
`
`
`
`MR. OLIVER: Yes.
`
`
`
`
`
`JUDGE KORNICZKY: Okay. I'm there.
`
`MR. OLIVER: So what are discussed generally throughout these
`
`paragraphs that have been cited, 14 is one example of them, is that Wruck is
`
`about determining when overrides have been made and how to account for
`
`those. So the -- Paragraph 14, Paragraph 104 talk about these different ways
`
`that overri