DCT

4:23-cv-00128

Bell Semiconductor LLC v. Texas Instruments Inc

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: Bell Semiconductor, LLC v. Texas Instruments, Incorporated, 4:23-cv-00128, E.D. Tex., 02/17/2023
  • Venue Allegations: Venue is alleged to be proper based on Defendant operating a "regular and established place of business" in the district, specifically an 80,000 square foot fabrication facility in Sherman, Texas, and committing acts of infringement therein.
  • Core Dispute: Plaintiff alleges that Defendant’s design and manufacturing processes for certain semiconductor devices infringe six U.S. patents directed to methodologies for integrated circuit design, including dummy metal insertion and handling engineering changes.
  • Technical Context: The technology concerns automated methods within the semiconductor design flow used to ensure chip manufacturability and performance, primarily by intelligently placing non-functional "dummy fill" material to achieve physical planarity and managing design revisions efficiently.
  • Key Procedural History: The complaint asserts that the patents originate from a portfolio developed by Bell Labs, Lucent Technologies, Agere Systems, and LSI Corporation. U.S. Patent 7,007,259, one of the asserted patents, was the subject of an ex parte reexamination where its patentability was confirmed.

Case Timeline

Date Event
2000-01-18 U.S. Patent 6,436,807 Priority Date
2002-08-20 U.S. Patent 6,436,807 Issue Date
2003-07-31 U.S. Patent 7,007,259 Priority Date
2003-10-10 U.S. Patent 7,260,803 Priority Date
2004-09-22 U.S. Patent 7,149,989 Priority Date
2004-11-17 U.S. Patent 7,396,760 Application Filing Date
2004-12-17 U.S. Patent 7,231,626 Application Filing Date
2006-02-28 U.S. Patent 7,007,259 Issue Date
2006-12-12 U.S. Patent 7,149,989 Issue Date
2007-06-12 U.S. Patent 7,231,626 Issue Date
2007-08-21 U.S. Patent 7,260,803 Issue Date
2008-07-08 U.S. Patent 7,396,760 Issue Date
2023-02-17 Complaint Filing Date
2023-07-05 U.S. Patent 7,007,259 Reexamination Certificate Issue Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 7,007,259 - "Method for Providing Clock-Net Aware Dummy Metal Using Dummy Regions"

Issued February 28, 2006

The Invention Explained

  • Problem Addressed: The patent addresses the problem that prior art methods for inserting "dummy metal" into integrated circuit designs used a large, hardcoded "stay-away" distance from sensitive clock nets. This often made it "impossible to insert enough dummy metal into a tile to meet the required minimum density," necessitating an "involved, iterative process" of rerunning the design tools, which could "significantly impact the design schedule" (’259 Patent, col. 2:3-18; Compl. ¶¶26-27).
  • The Patented Solution: The invention proposes a method, implemented in a software tool, to "minimize[] the negative timing impact of dummy metal on clock nets, while at the same time achieving minimum density in a single run" (’259 Patent, col. 2:19-23). It does this by first identifying all free spaces as "dummy regions" and then prioritizing them so that the regions located adjacent to clock nets are filled with dummy metal last (Compl. ¶28). The flowchart in the patent's Figure 2 depicts this process of iterating through design objects to define and tag dummy regions based on their proximity to clock nets (’259 Patent, Fig. 2).
  • Technical Importance: This approach allows for meeting manufacturing density requirements in a single, efficient pass, reducing design cycle time and costs associated with iterative design modifications (Compl. ¶30).

Key Claims at a Glance

  • The complaint asserts independent claim 1 (Compl. ¶31). The patent also contains independent claims 18 and 35.
  • Independent Claim 1 requires:
    • A method for inserting dummy metal into a circuit design that includes objects and clock nets.
    • Identifying free spaces on each layer of the circuit design as "dummy regions."
    • Prioritizing the dummy regions such that those located adjacent to clock nets are filled with dummy metal last, to minimize timing impact.
  • The complaint does not explicitly reserve the right to assert dependent claims.

U.S. Patent No. 6,436,807 - "Method for Making an Interconnect Layer and a Semiconductor Device Including the Same"

Issued August 20, 2022

The Invention Explained

  • Problem Addressed: The patent identifies that conventional layout algorithms place dummy fill based on a "predetermined set density," which is independent of the density of adjacent active features. This can lead to "unnecessarily placing dummy fill features," which increases parasitic capacitance, and can also cause "variations in the density of the interconnect layer" that interfere with the planarization process (’807 Patent, col. 2:17-37; Compl. ¶38).
  • The Patented Solution: The invention describes a method for making a layout that achieves uniform density to facilitate planarization (’807 Patent, col. 2:37-46). The process, illustrated in the flowchart of Figure 3, involves first determining the actual "active interconnect feature density" for each region of the layout, and then "adding dummy fill features to each layout region to obtain a desired density" (’807 Patent, Fig. 3, steps 52, 54; Compl. ¶41). The method also involves defining a minimum lateral dimension for the dummy fill features based on the deposition bias of the dielectric layer (’807 Patent, col. 6:1-10).
  • Technical Importance: This method facilitates uniform planarization during manufacturing while minimizing the addition of unnecessary dummy fill, which in turn "deceases the parasitic capacitance of the interconnect layer" (Compl. ¶42).

Key Claims at a Glance

  • The complaint asserts independent claim 1 (Compl. ¶43). The patent also contains independent claim 9.
  • Independent Claim 1 requires:
    • A method for making a layout for an interconnect layer to facilitate uniformity of planarization.
    • Determining an active interconnect feature density for each of a plurality of layout regions.
    • Adding dummy fill features to each layout region to obtain a desired density, where the adding step comprises defining a minimum dummy fill feature lateral dimension based upon a dielectric layer deposition bias.
  • The complaint does not explicitly reserve the right to assert dependent claims.

U.S. Patent No. 7,260,803 - "Incremental Dummy Metal Insertions"

Issued August 21, 2007

  • Technology Synopsis: The patent addresses the inefficiency of rerunning the entire dummy fill tool, a process that can take up to 30 hours, after a minor design change or Engineering Change Order (ECO) (Compl. ¶¶50-51). The patented solution is a method that, after a design change, performs a check to see if any existing dummy metal objects intersect with new or changed design objects and, if so, deletes only the intersecting dummy metal objects, thereby "avoiding having to rerun the dummy fill tool" (Compl. ¶¶53, 56).
  • Asserted Claims: Independent claim 1 is asserted (Compl. ¶55).
  • Accused Features: TI's accused processes allegedly use design tools to perform a Design Rule Check (DRC) after an ECO to identify rule violations, and then "trim metal fill geometries that cause the short or DRC violation" (Compl. ¶¶127-128).

U.S. Patent No. 7,149,989 - "Method of Early Physical Design Validation and Identification of Texted Metal Short Circuits in an Integrated Circuit Design"

Issued December 12, 2006

  • Technology Synopsis: The patent addresses the dilemma that running design validation checks late in the process is costly if faults are found, while running them early produces a large number of false errors due to the incomplete design (Compl. ¶¶62-63). The solution is a method that uses a reduced, specific rule deck that includes "only physical design rules that are specific to texted metal short circuits" to perform early validation, allowing for early defect detection without excessive false positives (Compl. ¶¶64, 67).
  • Asserted Claims: Independent claim 1 is asserted (Compl. ¶67).
  • Accused Features: TI's accused processes allegedly employ a design tool with a "short finder" or "short locator" functionality that generates and uses a specific rule deck to identify texted metal short circuits between different signal sources (Compl. ¶139).

U.S. Patent No. 7,231,626 - "Method Of Implementing An Engineering Change Order In An Integrated Circuit Design By Windows"

Issued June 12, 2007

  • Technology Synopsis: The patent addresses the inefficiency of running resource-intensive processes like routing and validation on an entire multi-million-cell circuit just to implement a small ECO affecting a few cells (Compl. ¶¶76-77). The invention provides a method that creates a "window" that encloses the design change, performs "incremental routing" only for nets within that window, and then replaces the corresponding area in a copy of the design with the results (Compl. ¶¶78, 83).
  • Asserted Claims: Independent claim 1 is asserted (Compl. ¶83).
  • Accused Features: TI's accused processes allegedly perform incremental routing, parasitic extraction, and design rule checks only for nets enclosed within a "window defining the ECO" (Compl. ¶¶148-150).

U.S. Patent No. 7,396,760 - "Method and System for Reducing Inter-Layer Capacitance in Integrated Circuits"

Issued July 8, 2008

  • Technology Synopsis: The patent identifies that prior dummy fill methods focused on achieving density and considered capacitive effects only within a single layer, failing to address "interlayer bulk capacitive effect" caused by the overlap of dummy fill on successive layers (Compl. ¶¶91-92). The patented solution is a method that treats each consecutive pair of layers together, determines the overlap between their respective dummy fill spaces, and minimizes that overlap by rearranging the dummy fill features (Compl. ¶¶94, 97).
  • Asserted Claims: Independent claim 1 is asserted (Compl. ¶97).
  • Accused Features: TI's accused processes allegedly use timing-aware tools to "rearrange dummy fill to minimize its overlap in successive layers," including the ability to stagger the fill (Compl. ¶¶160-161).

III. The Accused Instrumentality

Product Identification

  • The complaint identifies the "Exemplary Accused Products" as specific Texas Instruments devices, including the TI 66AH2H12 Multicore DSP+ARM KeyStone II SoC, ADS1261, DRV2614, RM48L952, and TM4C123GH6PGE (Compl. ¶1). The infringing acts are the "patented processes/methodologies" used to design and manufacture these products, which the complaint terms the "Accused Processes" (Compl. ¶13, ¶104).

Functionality and Market Context

  • The accused functionality resides in the electronic design automation (EDA) processes used by TI. The complaint alleges that TI employs a variety of design tools from vendors like Cadence, Synopsys, and/or Siemens to perform the patented methods for inserting dummy metal and managing design changes (Compl. ¶104, ¶115). The complaint alleges that TI manufactures products using these processes at a substantial fabrication facility in Sherman, Texas, which produces over 4,500 device types for automotive, commercial, military, and space applications (Compl. ¶18).

IV. Analysis of Infringement Allegations

The complaint references but does not provide claim chart exhibits (Exhibits G-L). The following analysis is based on the narrative allegations of infringement in the complaint's body.

’259 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
(a) identifying free spaces on each layer of the circuit design suitable for dummy metal insertion as dummy regions TI employs a design tool to identify free spaces on each layer of the circuit designs for its Accused Products suitable for dummy metal insertion. ¶105 col. 2:29-31
(b) prioritizing the dummy regions such that the dummy regions located adjacent to clock nets are filled with dummy metal last, thereby minimizing any timing impact on the clock nets TI's design tools prioritize dummy regions by assigning a "high cost" to adding metal fill near clock nets. This alleged cost assignment results in dummy regions adjacent to clock nets being filled last. ¶106 col. 2:31-34
  • Identified Points of Contention:
    • Scope Question: A central question may be whether assigning a "high cost" to a region is legally equivalent to the claim limitation of "prioritizing the dummy regions such that the dummy regions... are filled with dummy metal last." The court may need to determine if "last" requires a strict temporal sequence of operations or if it can be satisfied by a weighted, cost-based algorithm that effectively deprioritizes those regions.
    • Technical Question: What evidence demonstrates that TI's cost-based assignment actually causes the clock-net-adjacent regions to be filled "last" in practice? The analysis may turn on the specific implementation and operational outcomes of the accused EDA tools.

’807 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
(a) determining an active interconnect feature density for each of a plurality of layout regions of the interconnect layout TI's design tools determine an active interconnect feature density for each of a plurality of layout regions of the interconnect layout of its Accused Products. ¶115 col. 4:26-30
(b) adding dummy fill features to each layout region to obtain a desired density of active interconnect features and dummy fill features... TI's design tools add dummy fill features to each layout region to obtain a desired density of active and dummy features to facilitate uniformity of planarization. ¶116 col. 4:50-54
...the adding comprising defining a minimum dummy fill feature lateral dimension based upon a dielectric layer deposition bias... The adding of dummy fill through TI's design tools is alleged to comprise defining this minimum dimension based on the deposition bias for the dielectric layer. ¶¶116-117 col. 6:1-10
  • Identified Points of Contention:
    • Technical Question: The infringement case for this patent may hinge on the specific limitation of "defining a minimum dummy fill feature lateral dimension based upon a dielectric layer deposition bias." The complaint's assertion is highly technical, and a key point of contention will be what evidence exists to show that TI's process explicitly performs this calculation, rather than simply using a pre-set or empirically derived minimum feature size.
    • Scope Question: How does the patent define "desired density"? The court may need to construe whether this term requires a specific calculation relative to the active feature density, as taught in the patent, or if it could be met by targeting any uniform density value.

V. Key Claim Terms for Construction

  • Term from '259 Patent: "prioritizing the dummy regions such that the dummy regions located adjacent to clock nets are filled with dummy metal last"

    • Context and Importance: This phrase captures the novel technological contribution of the '259 Patent. The outcome of the infringement analysis for this patent will likely depend on whether Defendant’s alleged use of a "high cost" function in its design tools meets this "filled... last" limitation.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The specification's objective is to "minimize[] the negative timing impact" (’259 Patent, col. 2:19-20). A party could argue that any prioritization scheme that achieves this goal, including a cost-based one, falls within the claim's scope.
      • Evidence for a Narrower Interpretation: The patent's flowchart (Fig. 5) depicts a specific sequence: calculating timing factors (step 250), sorting a list based on those factors (step 252), and then inserting metal into the "sorted dummy regions" (step 254). A party could argue this supports a narrower construction requiring a strict sequential filling order.

  • Term from '807 Patent: "defining a minimum dummy fill feature lateral dimension based upon a dielectric layer deposition bias"

    • Context and Importance: This is a precise technical step that distinguishes the claimed method from simply adding fill. Practitioners may focus on this term because proving that a commercial EDA tool performs this specific, physics-based calculation could be an evidentiary challenge.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The phrase "based upon" could be argued to not require a direct, real-time calculation. If an engineer sets a minimum fill size in a tool based on their knowledge of the process's deposition bias, one might argue the size is "based upon" the bias.
      • Evidence for a Narrower Interpretation: The specification provides a concrete example: "the lateral dimension needs to be at least 3 microns to cause a negative bias of -1.5 at the upper surface" (’807 Patent, col. 6:23-26). This suggests a direct, quantitative relationship, potentially supporting a narrower interpretation that requires the accused process to implement a similar rule or calculation.

VI. Other Allegations

  • Indirect Infringement: The complaint focuses on direct infringement under 35 U.S.C. § 271(a) by alleging that TI itself uses the patented methods in its design and manufacturing processes (Compl. ¶103). It does not contain separate counts or detailed factual allegations for indirect infringement (inducement or contributory).
  • Willful Infringement: The complaint alleges that TI's infringement is "exceptional" and seeks enhanced damages and attorneys' fees (Compl. ¶109, ¶120). However, the complaint does not plead any specific facts to support a finding of willfulness, such as pre-suit knowledge of the patents or the infringement.

VII. Analyst’s Conclusion: Key Questions for the Case

  • A central issue will be one of evidentiary proof: The complaint alleges that Texas Instruments' use of general-purpose, third-party EDA tools (from Cadence, Synopsys, etc.) constitutes infringement. A key question for the court will be what technical evidence demonstrates that the generic functions of these tools (e.g., cost-based optimizers, rule checkers) are configured and used by TI in a manner that precisely practices the specific, often multi-step, limitations of the asserted method claims.
  • The case presents a question of functional specificity: Several patents address similar problems in the design workflow, such as handling ECOs (’803 and ’626 patents). The litigation will likely focus on whether TI's accused automated processes perform the exact sequence of steps claimed—for instance, "deleting the intersecting dummy metal objects" ('803 patent) versus creating a "window... bounded by coordinates" for "incremental routing" ('626 patent)—or if there is a fundamental mismatch in their technical operation.
  • Given the assertion of six closely related patents, a core challenge will be establishing the patentability distinctions between them. The court will likely need to scrutinize how the specific contributions of each patent differ, particularly where multiple patents address similar technical challenges like dummy fill placement ('259, '807, and ’760 patents), to resolve potential arguments regarding obviousness-type double patenting and claim differentiation.