DCT

1:25-cv-00316

Credo Semiconductor Inc v. Molex LLC

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 1:25-cv-00316, D. Del., 03/18/2025
  • Venue Allegations: Venue is alleged to be proper in the District of Delaware because Defendant Molex, LLC is a Delaware corporation and therefore resides in the district.
  • Core Dispute: Plaintiff alleges that Defendant’s Active Electrical Cable (AEC) products, used for high-speed data transfer, infringe three patents related to signal conditioning and pre-equalization technology.
  • Technical Context: The technology at issue involves Active Electrical Cables, which contain powered electronics to regenerate high-speed signals, enabling the use of copper cabling over longer distances in data centers as a cost-effective alternative to optical fiber.
  • Key Procedural History: The complaint alleges that Plaintiff provided Defendant with notice of the asserted patents in a letter dated September 1, 2023, more than a year before filing the lawsuit. This alleged pre-suit notice forms the basis of the willfulness allegations.

Case Timeline

Date Event
2018-08-28 Earliest Priority Date for ’111 Patent
2019-03-01 Earliest Priority Date for ’252 Patent
2019-11-27 Earliest Priority Date for ’233 Patent
2020-12-29 ’233 Patent Issued
2021-05-18 ’252 Patent Issued
2021-06-08 ’111 Patent Issued
2022-11-01 Accused Product referenced in Molex social media post
2023-09-01 Plaintiff allegedly provided Defendant with notice of patents-in-suit
2025-03-18 Complaint Filed

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

U.S. Patent No. 10,877,233, “Active Ethernet Cable with Preset Pre-Equalization” (Issued Dec. 29, 2020)

The Invention Explained

  • Problem Addressed: As data transmission rates in networks increase beyond 50 Gbps, standard copper cables suffer from signal degradation (attenuation and dispersion) over distances of more than a few meters, while relying solely on receiver-side equalization to fix the signal can lead to prohibitive power consumption (’233 Patent, col. 1:15-24, 1:39-42).
  • The Patented Solution: The patent describes an "Active Ethernet Cable" (AEC) that embeds powered "data recovery and re-modulation" (DRR) devices within the connectors at each end of the cable (’233 Patent, col. 8:52-61). Instead of waiting for the receiver to correct a degraded signal, these DRR devices apply transmit-side "pre-equalization" to the signal before it travels down the cable conductors, using pre-determined filter coefficient values stored in the device's nonvolatile memory (’233 Patent, Abstract). This pre-conditioning is intended to counteract the anticipated signal degradation from the cable itself.
  • Technical Importance: This approach aimed to provide a cost-effective alternative to expensive active optical cables for the longer-distance connections (e.g., up to seven meters) increasingly required in data centers (’233 Patent, col. 1:30-36).

Key Claims at a Glance

  • The complaint asserts at least independent claim 1 (Compl. ¶15).
  • Essential elements of claim 1 include:
    • A first and second data recovery and re-modulation (DRR) device, each located in a connector plug at opposite ends of a cable.
    • Electrical conductors connecting the two DRR devices.
    • The DRR devices converting signals between the host interfaces and the "electrical transit signals" that traverse the conductors.
    • The first and second DRR devices providing "pre-equalization of the electrical transit signals using transmit filter coefficient values stored in nonvolatile memories."

U.S. Patent No. 11,012,252, “Active Ethernet Cable” (Issued May 18, 2021)

The Invention Explained

  • Problem Addressed: Similar to the ’233 Patent, the invention addresses the difficulty of ensuring robust, high-speed data transmission in mass-manufactured cables where the electronics of the connected host devices may be unknown or vary (’252 Patent, col. 1:15-24, 4:5-19).
  • The Patented Solution: The patent discloses an active cable architecture where transceivers in each connector perform clock and data recovery (CDR) and re-modulation on signals traveling in both directions. A key aspect is the dual-function signal regeneration: the transceiver cleans up the "inbound" signal from the host before sending it down the cable, and also cleans up the "transit" signal arriving from the other end of the cable before sending it out to the host as an "outbound" signal (’252 Patent, Abstract). The invention specifically claims the use of "fixed, cable-independent, equalization parameters" for the host-facing signal processing (’252 Patent, col. 9:31-36).
  • Technical Importance: By actively managing the signal integrity at both ends and in both directions, the cable assembly itself is designed to guarantee a robust link, independent of the equalization capabilities of the host systems it connects (’252 Patent, col. 4:19-25).

Key Claims at a Glance

  • The complaint asserts at least independent claim 1 (Compl. ¶22).
  • Essential elements of claim 1 include:
    • An active Ethernet cable with two connectors and electrical conductors.
    • Each connector includes a transceiver that performs clock and data recovery (CDR) on the electrical input signal from the host to extract and re-modulate an "inbound data stream" for transit through the cable.
    • Each transceiver also performs CDR on the "electrical transit signal" from the other end of the cable to extract and re-modulate it as an "outbound data stream" to the host.
    • The transceivers employ "fixed, cable-independent, equalization parameters" for both the CDR on the host input signal and the re-modulation of the outbound data stream.

Multi-Patent Capsule: U.S. Patent No. 11,032,111, “SerDes Pre-Equalizer Having Adaptable Preset Coefficient Registers” (Issued June 8, 2021)

  • Technology Synopsis: This patent addresses the challenge of efficiently training a transmitter's pre-equalizer in a high-speed serial link. The invention proposes a transceiver that contains multiple registers, each storing a different set of pre-programmed initial coefficient values for the pre-equalizer. These sets correspond to different anticipated channel conditions (e.g., different cable lengths or types), allowing the system to quickly select an optimal starting point for adaptation rather than training the equalizer from a default or unknown state (’111 Patent, Abstract; col. 2:40-48).
  • Asserted Claims: The complaint asserts at least independent claim 8 (Compl. ¶29).
  • Accused Features: The complaint alleges that the Accused Products' method for selecting and adapting pre-equalization settings infringes this patent (Compl. ¶29).

III. The Accused Instrumentality

  • Product Identification: The accused instrumentalities are Molex’s "AEC OSFP to OSFP Straight Cable with Broadcom BCM 87854 Re-Timer" and "AEC QSFP-DD to QSFP-DD Straight Cable with Broadcom BCM 87850 Re-Timer" (collectively, the "Accused Products") (Compl. ¶15).
  • Functionality and Market Context: The Accused Products are active electrical cables (AECs) that incorporate "Re-Timer" integrated circuits from Broadcom. The complaint alleges these products are designed for high-speed data transmission within data centers and other demanding environments (Compl. ¶¶ 9, 11). The complaint highlights the products' use in advanced server infrastructure, citing their demonstration at the OCP Summit 2022 in a server rack for Microsoft (Compl. ¶17). The complaint provides a link to an article containing a photograph of the Accused Products demonstrated in a server rack at a 2022 trade show (Compl. ¶17).

IV. Analysis of Infringement Allegations

The complaint references claim chart exhibits demonstrating infringement (Exhibits 4, 5, and 6) but does not attach them; therefore, a tabular analysis is not possible. The narrative allegations suggest the following infringement theories.

For the ’233 Patent, the infringement theory centers on the allegation that the Broadcom Re-Timers in the Accused Products function as the claimed "DRR devices." The complaint alleges these devices perform transmit-side "pre-equalization" on signals traveling through the cable by using stored "transmit filter coefficient values" to condition the signal (Compl. ¶15).

For the ’252 Patent, the theory is that the Re-Timers in the Accused Products perform the claimed two-way signal regeneration. This includes performing clock and data recovery (CDR) on signals received from the host and on signals received from the other end of the cable. A core part of the allegation is that this process uses "fixed, cable-independent, equalization parameters" for the specified signal processing steps directed toward the host interface (Compl. ¶22).

  • Identified Points of Contention:
    • Technical Questions: A primary factual question will be whether the operational characteristics of the Broadcom "Re-Timers" in the Accused Products align with the specific functions claimed in the patents. This includes determining if they perform pre-equalization as described in the ’233 Patent and the dual-direction CDR and re-modulation described in the ’252 Patent.
    • Scope Questions: A central dispute may arise over whether the accused "Re-Timers" can be properly characterized as the claimed "DRR device" (’233 Patent) or "transceiver" (’252 Patent) as those terms are used in the patent specifications. Further, the evidence required to prove that the equalization parameters used are "fixed" and "cable-independent" (’252 Patent) will be a key point of contention.

V. Key Claim Terms for Construction

  • The Term: "pre-equalization ... using transmit filter coefficient values stored in nonvolatile memories" (’233 Patent, Claim 1)

    • Context and Importance: This phrase captures the essence of the ’233 invention. The outcome of the case may depend on whether the accused Re-Timers are found to perform this precise function. Practitioners may focus on this term because it links a specific action ("pre-equalization") with a specific implementation detail ("stored in nonvolatile memories").
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The specification suggests flexibility, noting that coefficient values can be determined and stored after the cable is assembled, implying a factory-set process, but also discusses a "training phase" that can occur during operation, which could support a more dynamic interpretation (’233 Patent, col. 2:40-43, col. 10:55-59).
      • Evidence for a Narrower Interpretation: The detailed description and Figure 6 emphasize a manufacturing method where the cable is tested, parameters are determined, and then those parameters are set, suggesting the "preset" nature of the coefficients is a key feature (’233 Patent, Fig. 6; col. 10:10-18).

  • The Term: "fixed, cable-independent, equalization parameters" (’252 Patent, Claim 1)

    • Context and Importance: This limitation is crucial for distinguishing the invention from other equalization schemes. Infringement of claim 1 of the ’252 Patent hinges on whether the accused transceivers use parameters with these specific characteristics for the claimed host-facing operations.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The patent does not provide an explicit definition for "cable-independent." A party could argue it broadly means any equalization setting not specifically tuned to the unique, measured electrical properties of the internal cable conductors.
      • Evidence for a Narrower Interpretation: The specification explicitly contrasts these parameters with "cable-dependent equalization parameters" used for the transceivers processing signals that have traversed the cable (’252 Patent, col. 8:60-65). This contrast suggests "cable-independent" has a specific technical meaning. Furthermore, dependent claim 3 adds the concept of adapting parameters during usage, which may imply that "fixed" in independent claim 1 means non-adaptive.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges induced infringement, stating that Defendant provides instructions, marketing materials, and customer support with the specific intent that its customers and distributors use the Accused Products in an infringing manner (Compl. ¶¶ 17, 24, 31). It also alleges contributory infringement, asserting that the Accused Products and their components are especially made for infringement and are not staple articles of commerce with substantial non-infringing uses (Compl. ¶¶ 18, 25, 32).
  • Willful Infringement: Willfulness is alleged based on pre-suit knowledge. The complaint asserts that Plaintiff sent a letter to Defendant on September 1, 2023, identifying the patents-in-suit. The complaint alleges that Defendant continued its infringing conduct unabated despite this notice, constituting willful and deliberate infringement (Compl. ¶¶ 20, 27, 34).

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

  • Technical Equivalence: A central issue will be one of technical equivalence: What is the precise mode of operation of the Broadcom "Re-Timers" within the accused Molex cables? The case will require a deep evidentiary dive into whether this third-party component performs the specific, two-way signal regeneration and pre-equalization using stored coefficients as recited in the asserted claims.
  • Definitional Scope: The case will also likely turn on a question of definitional scope: How will the court construe the term "fixed, cable-independent, equalization parameters"? The resolution of this claim construction issue will be critical for determining the boundary of the ’252 patent’s claims and whether the Accused Products, which are designed to function in a variety of host systems, fall within that scope.
  • Timing and Knowledge: A key question for damages will be one of timing and knowledge: Can Plaintiff prove that the notice letter of September 1, 2023 provided Defendant with actual knowledge of infringement, and did Defendant's subsequent conduct rise to the level of willfulness, potentially exposing it to enhanced damages?