Listen to this post

EcoFactor, Inc. is the holder of U.S. Patent No. 8,498,753, titled “System, Method and Apparatus for Just-In-Time Conditioning Using a Thermostat,” which focuses on optimizing climate control systems, particularly HVAC systems, in residential and commercial buildings. The patented technology utilizes a thermostat that considers external weather conditions and the thermal characteristics of individual structures to dynamically adjust temperature settings, aiming to improve energy efficiency and comfort. Within Claim 1, a method is outlined for decreasing the cycling time of a climate control system, comprising steps such as accessing historical temperature data, determining thermal performance values, and calculating the optimal activation time for the system based on various inputs.

In response to EcoFactor’s patent, Google LLC and ecobee, Inc. filed a petition for inter partes review (IPR) of claims 1–20 of the ‘753 patent, alleging their obviousness in view of prior art. Specifically, Google and ecobee asserted that the combination of U.S. Patent No. 5,197,666 (“Wedekind”) with U.S. Patent No. 6,216,956 (“Ehlers”) rendered the claims obvious. The initiation of the IPR proceedings by the Patent Trial and Appeal Board (PTAB) sparked a dispute over the interpretation of specific claim limitations, particularly the [1m] limitation in Claim 1, which involves determining activation time based on inputs such as internal and external temperatures.

After the initiation of the IPR proceedings, a disagreement ensued regarding whether Wedekind disclosed the element of claim limitation [1m] concerning the determination of activation time based on internal temperature. Google argued that Wedekind calculated activation time based on thermal performance values, which were derived from internal temperature values. EcoFactor disagreed, contending that each input in the [1m] limitation must be distinct, and that using the same value for multiple inputs would render parts of the claim limitation meaningless. Despite the absence of explicit claim construction arguments, the PTAB determined that each input in the [1m] limitation must be distinct and rejected Google’s obviousness theory, stating that it failed to use each of the five distinct inputs as required by the claim.

Appeal and Challenge:

Google appealed the PTAB’s decision, challenging the Board’s determination and alleging an erroneous claim construction of a limitation in Claim 1. Additionally, Google argued that the Board’s Final Written Decision violates the Administrative Procedure Act because Google had no notice or an opportunity to address the Board’s construction.

Issues Presented:

  1. Did the Board engage in implicit claim construction regarding the [1m] limitation in Claim 1?
  2. If so, was the Board’s claim construction erroneous?
  3. Did the Board violate the Administrative Procedure Act (APA) in its claim construction?


1. The Board did engage in implicit claim construction regarding the [1m] limitation in Claim 1.

The Federal Circuit laid out its legal framework regarding the determination of when the PTAB is engaging in claim construction by heavily drawing from established precedent set forth in HTC Corp. v. Cellular Comms. Equip., LLC, and Netword, LLC v. Centraal Corp. In HTC Corp. v. Cellular Comms. Equip., LLC, the court established that “[c]laim construction serves to define the scope of the patented invention and the patentee’s right to exclude.” Additionally, Netword, LLC v. Centraal Corp. determined that, “[c]laim construction is the judicial statement of what is and is not covered by the technical terms and other words of the claims.”

Using HTC and Netword as guiding lights, the Federal Circuit reasoned that when assessing whether a court or the Board has construed a claim, it is instructive to consider the outcome of the tribunal’s analysis. If the analysis delineates the scope and meaning of the patented subject matter, it is probable that the court or the Board has indeed construed the claim. Moreover, the Federal Circuit discussed the significance of the timing of the claim construction analysis by concluding that “[c]laim construction may be undertaken prior to or in tandem with the tribunal’s review of the allegedly infringing technology or prior art. The point in the proceeding at which the analysis occurs is not dispositive.”

Following the establishment of its legal framework, the Federal Circuit redirected its attention to the Board’s evaluation of the [1m] limitation in the ‘753 patent on appeal. While the Board stated that “[b]ecause no express construction is needed for our decision, we do not construe any of the claim limitations,” this comes in conflict with the fact that the Board later concluded that the [1m] limitation “recites five distinct inputs upon which the time is based at least in part” and thus it “requires that each of those inputs be a distinct component of the calculation of the first time prior to said target time.” Despite the Board’s assertion that it was not engaging in claim construction, the Federal Circuit found this assertion insufficient to dismiss the possibility that claim construction had taken place. In response, the Federal Circuit emphasized that implicit claim constructions have been identified even when the Board itself did not acknowledge such construction. Regardless of the absence of an explicit claim construction for a particular claim term, as seen in HTC where the court held that “[d]espite no express construction of [a claim term] below, [the] Board[’s] findings establish[ed] the scope of the patented subject matter,” the Federal Circuit emphasized the significance of the Board’s findings in delineating the scope of the patented subject matter.

The Federal Circuit concurred with Google’s contention that the Board’s evaluation of the [1m] limitation constituted claim construction because the outcome of the assessment delineated the scope of the [1m] limitation. The [1m] limitation outlines five inputs influencing the system’s activation timing, yet the claim provides no clear directives regarding their interplay or boundaries, leaving room for interpretation. By establishing that no input can be partially derived from another and insisting on the distinctness of each input, the Board effectively constrained the scope of the [1m] claim limitation. The impact of the Board’s limitation was apparent in its inability to refute Google’s argument that the measurement from the prior art reference could fulfill both the input requirements for [i] and [iii]. As a result, the Federal Circuit concluded that the Board’s assessment qualified as claim construction.

2. The claim construction by the Board was erroneous.

The Federal Circuit deemed the claim construction by the Board erroneous for several reasons. First, the Board interpreted the [1m] limitation in Claim 1 as requiring each input to be entirely distinct from the others, with no input based in part on another. However, this interpretation was not explicitly supported by the claim language or the specification of the patent. The claim language indicates that the activation time is based “at least in part on” each of the five inputs, suggesting flexibility in how the inputs can be utilized.

Additionally, the court found that the specification of the ‘753 patent does not impose any explicit requirement that the inputs must be entirely distinct. The Federal Circuit quotes Sunovion Pharms., Inc. v. Teva Pharms. USA, Inc. to support this rationale, which states that “[w]hen construing claim terms, we first look to, and primarily rely on, the intrinsic evidence, including the claims themselves, the specification, and the prosecution history of the patent, which is usually dispositive.” Contrary to the Board’s findings, the Federal Circuit determined that the specification contemplates scenarios where one input may be calculated based on another. “The claim language broadly recites that ‘a first time’ is determined ‘based at least in part on’ each of the five inputs. ’753 patent, 9:47– 61 (emphasis added). This language places no constraint on the manner in which the inputs are used.”

Finally, the Federal Circuit addressed the cases cited by the Board in support of its narrow construction. The Board relied on Becton Dickinson & Co. v. Tyco Healthcare Group, LP and Engel Indus., Inc. v. Lockformer Co. to support the notion that “separately listed claim elements are distinct components, regardless of the intrinsic record.” However, the Federal Circuit clarified that Becton and Engel do not mandate such an interpretation and instead merely establish a presumption that separately listed claim limitations may indicate distinct components but that the intrinsic record of the patent can rebut the presumption.

Overall, the Federal Circuit found that the Board’s construction of the [1m] limitation was overly restrictive and not adequately supported by the claim language, specification, or relevant case law. Due to the Board’s failure to recognize the flexibility inherent in the language of the claim and by imposing unnecessary limitations that were not justified by intrinsic evidence, the Federal Circuit vacated the Board’s Final Written Decision and remanded with instructions that the Board apply Google’s proposed construction.

3. The Board did not violate the Administrative Procedure Act (APA) in its claim construction.

In finding that the Board did not violate the Administrative Procedure Act (APA), the Federal Circuit relied on the legal reasoning of Qualcomm Inc. v. Intel Corp. and SAS Inst., Inc. v. ComplementSoft, LLC. Qualcomm states that “[T]he Board may adopt a claim construction of a disputed term that neither party proposes without running afoul of the APA.” SAS, however, limits Qualcomm, as the Federal Circuit points out, by making it so “the Board cannot, without notice and opportunity for the parties to respond, change theories midstream by adopting a claim construction in its final written decision that neither party requested nor anticipated.”

The Federal Circuit concluded that since the record indicated a dispute between the parties regarding the meaning and scope of the [1m] limitation during the IPR proceeding and the fact that neither party proposed an explicit claim construction, it is therefore implied that both parties acknowledged the central issue concerning the scope and boundaries of the five inputs listed in [1m] limitation. Consequently, both parties were provided notice and an opportunity to address this issue. Thus, the Federal Circuit concluded that “because Google ‘had notice of the contested claim construction issues and an opportunity to be heard,’ the Board’s claim construction of claim 1 did not violate the APA.”