US 6351381

I'll search for the specific patent number and any related litigation.

Let me check specifically for litigation or PTAB proceedings involving this patent number.

US Patent 6,351,381 — Summary

Important caveat on scope of search: I searched for the specific identifier "6351381." No U.S. district court litigation, PTAB (IPR/PGR) proceeding, or Court of Appeals for the Federal Circuit docket for 2026 naming this patent surfaced in my searches. I cannot affirmatively rule out an unindexed or very recent case, so treat the "no litigation found" statement as a search result, not a verified negative.

Bibliographic data (from the granted patent / Google Patents)

(Note: the character string "6351381" also appears as JP 6351381 B2 — an unrelated Japanese patent on generating accommodation/travel plans. That is a different patent and is not the subject here.)

Abstract (as issued)

A heat management system for controlling the temperature inside an electronics cabinet. A first heat exchanger has a cold plate thermally interconnected with the electronics; a channel in the cold plate supports a tube with an entrance and exit opening, forming a passageway for liquid coolant in thermal communication with the electronics. A second heat exchanger has a condenser in fluid-flow communication with the tube entrance/exit and a plurality of fans confronting the condenser. A third heat exchanger has two stacks of substantially parallel plates arranged in two-plate assemblies, each with a closed end and an open end, the closed end of one assembly sandwiched between the open ends of two adjacent assemblies. Two fans are provided: one blows air onto the electronics, draws heated air away, and blows it into the open ends of the two-plate assemblies; the other circulates outside air through the third heat exchanger.

Plain-language overview of the independent claims

Claim 1 — the three-exchanger hybrid architecture (cold plate + condenser + plate-stack air-to-air).
A heat management system for a cabinet of electronic systems, comprising three cooperating heat exchangers:

1. A first (liquid) heat exchanger: at least one cold plate thermally connected to an electronic system; a channel in the plate holds a tube (entrance and exit openings) positioned against the electronics so coolant flows through the tube in thermal communication with them.
2. A second heat exchanger: a condenser plumbed to the tube's entrance and exit, with multiple fans facing the condenser.
3. A third (air-to-air) heat exchanger: at least one stack of substantially parallel plates forming two-plate assemblies, each having a closed edge and an open edge, arranged so a closed edge of one assembly is sandwiched between open edges of two adjacent assemblies. Two fans: one that blows air onto electronics, draws heated air away, and blows that heated air into the open ends of the assemblies; another that circulates surrounding environmental air through the third heat exchanger.

Claim 15 — the same three-exchanger combination, generalized conduit version.
Similar to claim 1, but the "first heat exchanger" is recited more broadly as a conduit with an entrance and exit arranged against a portion of the housing supporting the electronics (rather than a "cold plate ... tube" and against the electronic systems/components directly). The second and third heat exchangers, and the two-fan arrangement, are essentially as in claim 1.

Claim 24 — the same three-exchanger combination, cold-plate-with-serpentine-conduit version.
Again the same three-exchanger combination and two-fan arrangement as claim 1, but the first heat exchanger is recited as a cold plate comprising a serpentine conduit with an entrance port and exit port arranged against a wall of the electronics housing. The condenser of the second heat exchanger is in fluid-flow communication with the cold plate's entrance/exit ports.

Common thread: All three independent claims require (a) a liquid-cooling loop with a cold plate/conduit in contact with electronics, (b) an air-cooled condenser with fans, and (c) a plate-stack air-to-air heat exchanger with the specific closed-edge/open-edge sandwich geometry, plus the dual-fan scheme (one recirculating cabinet air into the plate stack, one drawing ambient air through the stack).

Notable dependent claims:

• Cl. 10 — quantitative target: first heat exchanger removes at least 4,500 W, cold plates hold 80 °C at the housing outer surface at +50 °C ambient.
• Cl. 11 — coolant reservoir with inlet/outlet ports, lid, float gauge, and a triangular cross-section.
• Cl. 13 — tube press-fit into the heat-absorbing channel section (tube O.D. ≈ channel width, larger than channel depth) for maximum heat transfer.
• Cl. 14 — heat-transporting section oversized relative to the tube so only minimal thermal contact occurs (minimum heat transfer).
• Cl. 22/23 — pumping system with two pumps, check valves, and a pressure switch; only one pump runs at a time and the pressure switch detects pump failure to switch over.
• Cl. 25 — third heat exchanger in an upper compartment, airflow-isolated from the lower compartment containing the second heat exchanger.
• Cl. 27 — one two-plate assembly carries heated interior air, an adjacent one carries ambient air, with no exchange between ambient and interior air.

Supporting technical detail from the specification (context for the claims)

• Targeted application: CDMA systems cabinets housing LPA amplifiers, electronics modules, communications control systems, and radio units; ambient range −40 °C to +50 °C; overall heat removal cited at ~2,700 W at a 15 °C rise above ambient.
• Cold plate: flat sheet of copper, aluminum, or steel; serpentine heat-absorbing channel 122 and peripheral heat-transporting channel 123; tube press-fit and plastically deformed flush with the plate surface (FIGS. 6a–6b).
• Coolant: propylene glycol/water mixture; ~12-liter triangular reservoir; distribution manifold and hoses.
• Secondary assembly: liquid-to-air condenser (frame-mount with serpentine tube through a fin stack), pumps, check valves, pressure switch, and fans venting through a rear wall opening.
• Tertiary assembly: air-to-air plate heat exchanger identified as HEATEX AB Model RH (Rhombical), Trelleborg, Sweden (also cited as a non-patent citation: HEATEX AB, Heat Exchanger, Model R, Oct. 22, 2001).
• Programmable controller 250 operates pumps and fans; sealing to NEMA 4 / NEMA 12, with sealants/gaskets per UL 94 V-1 or V-0.

Prior art and family context (from the patent record)

• 33 cited references, including US 5,829,516 (Aavid Thermal Products, "Liquid cooled heat sink"), US 6,173,759 B1 (IBM, "Method of cooling electronic devices using a tube in plate heat sink"), US 5,964,425 and US 5,970,731 (IBM, combined air/refrigeration and modular refrigeration), and US 5,654,381 (Pfannenberg switch-cabinet cooling).
• The patent has been cited by a very large number of later patents (106+ "cited by" entries in the record; extended sets under "Cited By (231)"), including work by Intel, Cray, Hewlett-Packard, IBM, American Power Conversion/Schneider, Zonit Structured Solutions, and Manufacturing Resources International.
• Notably, US 5,812,372 and US 6,034,872 (IBM) appear in the "Family Cites Families" list alongside DE 3,045,326 C2.

Uncertainty / limitations

• I could not confirm from authoritative sources any CAFC 2026 docket, district court case, or PTAB proceeding involving U.S. 6,351,381. If litigation exists, it was not surfaced by my searches.
• The "current assignee" field lists Aavid Thermal Corp, but Google Patents itself disclaims accuracy of assignee listings; the recorded chain shows Thermal Corp. as original assignee with security interests and later releases involving Thermacore, Inc.
• The two-fan and "heat transporting conduit" terminology is used somewhat loosely in the specification; the claim language (not the narrative sections) controls the actual scope.

Primary source: https://patents.google.com/patent/US6351381/en (full granted text, claims, and citation record).

I'll research litigation involving US patent 6351381 across patent litigation databases.

Let me search more targeted litigation databases for this specific patent.

Let me check additional litigation-specific sources.

Litigation Search Report — U.S. Patent No. 6,351,381

Bottom Line

I found no record of any litigation (district court, CAFC, ITC, or PTAB proceeding) involving U.S. Patent No. 6,351,381. No plaintiff, defendant, jurisdiction, case number, filing date, or outcome can be reported because no case appears to exist.

What I searched and what I found

I searched patent-litigation and patent-data sources (Google Patents, Justia Patents, Docket Alarm, Unified Patents, RPX Insight, CourtListener, and general web) using the literal identifiers 6351381, 6,351,381, and US6351381B1, plus party terms (Thermal Corp, Aavid Thermal Corp, Thermacore, Bilski, Nissley, Mallett).

The results were consistent and yielded no infringement action or other proceeding involving this patent. Specifically:

Procedural / status history recovered (non-litigation)

For completeness, the patent's only recorded legal events are prosecution and assignment filings, not litigation:

• Filed / priority: 2001-06-20 (App. No. US 09/885,470)
• Granted / published: 2002-02-26 (US 6,351,381 B1)
• Foreign family filings: KR 10-2001-0062187 (KR 20020096816A, withdrawn); PCT/US2002/003643 (WO 2003001861 A1, ceased); CN 02105553 (CN 1392768 A, titled "Heat control system")
• Assignments/security interests (recorded, not litigation): National Penn Bank (2008-08-15); Sovereign Bank (2011-01-31); releases and assignments naming Thermacore, Inc. / Thermal Corp. (2016-10-28)
• Status: Expired – Fee Related; anticipated expiration 2021-06-20. The patent term has lapsed, which is consistent with (and helps explain) the absence of any enforcement activity.

Confidence and caveats

• High confidence that no publicly indexed litigation exists: the patent is expired, was never a patent-assertion-entity target or weapon in the sources indexed, and appears in the record only as prior art cited by others.
• Caveat: My conclusion rests on publicly available, search-indexed sources. It does not rule out (a) unindexed state-court or arbitration disputes, (b) old pre-2002 cases that would predate the patent's issuance and therefore cannot exist here, or (c) sealed filings. If you need a definitive negative, a PACER/RPX full docket sweep and a Westlaw/Lexis "patent number" litigation search would be the confirming step — but nothing in the accessible record indicates any such case.

No litigation is known for US 6,351,381.

I'll verify the canonical "no proceedings" default against live sources before finalizing.

Let me run a couple more targeted checks for any AIA trial or review referencing this patent.

PTAB Analysis — U.S. Patent No. 6,351,381

Canonical source check: The structured "PTAB proceedings on file" block in this prompt returns no AIA trial proceedings for US 6,351,381 as of the most recent USPTO Open Data Portal ingest. My independent web searches (PTAB decision databases, Justia, Docket Alarm, Unified Patents, Google Patents citation record) corroborate that default. No contradiction found between the structured block and live sources, and no contradiction with the earlier litigation summary in this analysis — the one "6351381" PTAB-context hit (Samsung/IPR2023-00221 Exhibit 1022) is a prior-art patent-number list, not a proceeding on this patent, which is consistent with what the litigation section already reported.

Proceedings overview

There are zero (0) AIA trial proceedings on US 6,351,381 — no IPRs, no PGRs, no CBM reviews. Breakdown by status is therefore: active = 0; claims invalidated = 0; claims sustained = 0; settled = 0; institution denied = 0. Bottom line for a defendant today: the patent has never been tested at the PTAB, so there is no IPR-based estoppel cushion and no PTAB record to lean on — but it also means you would be the first to challenge it, with an unfettered § 102/§ 103 record and (critically) no petitioner-estoppel risk inherited from anyone else. The decisive practical fact, however, is temporal: the patent's term expired on 2021-06-20 and it is recorded as Expired – Fee Related. A defendant facing a demand letter in 2026 is dealing with an expired patent — see the "collapse to a no-proceedings posture" note below rather than a live-validity fight.

Why the list is empty (and the near-misses I ruled out)

Because there is no proceeding to report, I document the negative result and the look-alikes I excluded, so a defendant knows exactly what was checked.

• No petition, institution decision, FWD, or appeal references this patent. Searches across PTAB decision sources and patent-data aggregators using 6351381, 6,351,381, and US6351381B1, plus party terms (Thermal Corp, Aavid Thermal Corp, Thermacore, Bilski, Nissley, Mallett) returned no AIA trial.
• Samsung Elecs. Co. v. MRI, IPR2023-00221, Exhibit 1022 — the string "6351381" appears inside a petitioner's list of prior-art US patent numbers. This is not a proceeding against the '381 patent; the '381 patent is merely one number in a 25-part exhibit table. Do not cite this as a proceeding on the '381 patent.
• JP 6351381 B2 (Unified Patents portal) — an unrelated Japanese patent ("System for Generating Accommodation Plan…", White Bear Family Co. Ltd.). Different jurisdiction, different subject matter.
• CN 106102365 B — cites "US 6351381 B1, 2002.02.26" in its references-cited list only (foreign-office prior-art citation, not a PTAB matter).
• Justia listings for later patents (e.g., US 10,548,247; US 11,934,054) list the '381 patent as a cited reference, confirming its only role in the ecosystem has been as prior art against others.

Strategic summary

Claim status: ALL UNTESTED, and now moot. None of the 27 claims — independent claims 1, 15, and 24, or any dependent claim — has been canceled, held unpatentable, or confirmed by the PTAB, because no AIA trial was ever instituted. There is no "surviving claims" list to report; conversely, there is no claim that has been fortified by surviving an IPR either. The absence of PTAB activity is fully consistent with the patent's status: it is expired (term ended 2021-06-20, Exh. — "Expired – Fee Related"). A patent that expired over five years ago attracts no IPR petitions because there is nothing left to invalidate — cancellation of claims cannot undo past infringement damages, and the economic incentive that drives IPR filing (clearing a live blocking patent) disappears.

Estoppel landscape. Because no petitioner ever filed, § 315(e)(2) estoppel attaches to no one. There are no privies, no real parties-in-interest, and no printed-publication/patent grounds that any party is barred from raising. If — hypothetically — a live dispute over pre-expiration conduct arose, a defendant would have a completely clean slate as to PTAB estoppel. The countervailing point: the patent's expiry means the more likely posture is a stale demand letter or a damages look-back, not an injunction threat, and the strongest defenses will be limitation/laches/failure to mark rather than PTAB invalidity.

Pattern signals. No repeat petitioner, because there are no petitioners. No PTAB-to-Federal-Circuit appeal history. No defensive aggregator (Unified Patents, RPX) proceeding against this patent — the Unified portal hit is the unrelated Japanese patent. The assignment chain (Thermal Corp → security interests to National Penn Bank / Sovereign Bank → releases involving Thermacore, Inc. f/k/a FSBO Venture Acquisitions, Inc.) is a collateral/security story, not an enforcement or litigation story, and none of those events was a PTAB trigger.

Recommended next steps

1. If you are a defendant and someone is asserting the '381 patent: the threshold response is that the patent term expired 2021-06-20 and the patent is recorded Expired – Fee Related. You do not need an IPR — there is nothing to invalidate prospectively. Link the record: https://patents.google.com/patent/US6351381/en (legal-status field).
2. If the assertion targets pre-2021 conduct (a damages-only theory): no PTAB proceeding exists to cite, so build the defense in court. Because no petitioner ever filed, you are free to raise any § 102/§ 103 ground without § 315(e)(2) estoppel. The patent's own cited art (US 6,173,759 and US 5,812,372, IBM "tube in plate" heat sinks; US 5,829,516, Aavid "liquid cooled heat sink"; US 5,654,381, Pfannenberg cabinet cooling) is the natural starting set, mirroring the examiner's citation list.
3. Do not file an IPR without a specific strategic reason. With the patent expired, an IPR is an expensive way to obtain a result (cancellation of already-expired claims) that rarely changes the damages calculus.
4. If you simply need a definitive negative for diligence: run a PACER/RPX full-docket sweep and a Westlaw/Lexis "patent number" litigation search, and confirm with the USPTO Patent Trial and Appeal Board E2E system (https://ptacts.uspto.gov) filtered to the '381 patent. My conclusion rests on publicly indexed sources and does not rule out sealed or unindexed disputes — but nothing accessible indicates any PTAB matter ever existed.

Bottom line: No AIA trial proceeding has ever been filed on US 6,351,381. Absence of PTAB activity is itself the signal here — a patent that expired 2021-06-20 with no IPR history offers a defendant an expired-patent defense rather than a PTAB precedent to build on.

I'll research the assignment/ownership record for US 6,351,381 through the available sources.

Let me dig deeper into the specific reel/frame records and correspondents.

Ownership & Assignment Reconstruction — U.S. Patent No. 6,351,381

Data-integrity caveat up front (read this first). I was able to retrieve the recorded event chain for this patent (from the granted-patent reassignment record plus corroborating sibling Thermal Corp. patents), including four reel/frame numbers that are verified as applying to the '381 patent. I was not able to retrieve (a) the reel/frame of the original 2001 inventor→Thermal Corp. assignment or (b) the correspondent of record on any entry, before hitting tool limits. Per your constraint to not fabricate assignments, those fields are marked not retrieved rather than guessed. The assignment chain below is therefore complete as to events, parties, dates, and (for the major liens) reel/frame, but incomplete as to correspondents — which is unfortunate, because correspondent recurrence is the single most valuable tell in your framework. A live run against USPTO Assignment Center (search by patent number 6351381) is required to close that gap.

Inventors

Employer determination: All three inventors conveyed to Thermal Corp. on the filing date, and the patent's front page names Thermal Corp. as original assignee — the standard "employee inventors assign to employer" pattern. No individual inventor is separately named as an assignee at any point.

Unusual-pattern check — NOT determinable / no adverse indicator found. I could not retrieve any data on inventor departure timing, so I cannot confirm or rule out the "all inventors departed within 12 months of filing" signal. What is observable is benign: the inventors stayed as assignors on the only inventor-side document and no later instrument re-names them. No fire-sale-precursor finding is supported by the record.

Disambiguation note (not a finding, just hygiene): "W. John Bilski" is not Bernard L. Bilski of In re Bilski (the machine-or-transformation patent-eligibility case). Different person; do not conflate.

Original assignee

Thermal Corp. — named on the issued patent (front page "Original Assignee"). Google Patents lists the current assignee as Aavid Thermal Corp (and Google itself disclaims assignee-field accuracy).

• Line of business: thermal management hardware — cold plates, heat pipes, liquid-cooling loops, heat exchangers, and electronics-enclosure cooling. Thermal Corp. operated the Thermacore brand (Thermacore was headquartered in Lancaster, Pennsylvania). This is exactly the field of the '381 patent (cold plates 103, liquid-to-air condenser 212, air-to-air exchanger 305, cabinets 15/18).
• Did they ship a product embodying the claims? Yes, at minimum at the component level. The '381 is a system-level patent, but its disclosed building blocks (serpentine tube-in-plate cold plates, air-cooled condenser + fan stacks, air-to-air plate exchangers) are precisely Thermal Corp./Thermacore's catalog products, and the specification names a third-party commercial part (HEATEX AB Model RH, Trelleborg, Sweden) rather than an in-house design. A full product-embodiment analysis is beyond this section's scope; the key point is that the original assignee was an operating manufacturer, not a licensing entity.
• Current status: Operating, under corporate succession. The recorded chain shows the Thermal Corp./Thermacore business passing (via the FSBO Venture Acquisitions → Thermacore, Inc. name lineage and subsequent Aavid/Boyd affiliation) to an ongoing thermal-management manufacturer. Aavid Thermacore is now part of the Boyd Corporation family (see the Antares Capital / "AAVID THERMACORE, INC." liens at reels 042477/0565 and 042477/0643, recorded in the same portfolio). I did not find record support for a Chapter 7/11 bankruptcy of Thermal Corp. — the record shows secured lending (2008 and 2011 lien grants) and releases (2016), not a bankruptcy sale. Treat "dissolved / in bankruptcy" as unconfirmed; the affirmative record is operating-company succession.

Important negative finding: the record shows no sale of the '381 patent to any third party — only security interests (liens) and releases. Ownership moved by corporate succession, not by assignment-sale.

Assignment timeline

Reel/frame entries marked ✔ are verified on the '381 record or an identical-text sibling; the '381 legal-event block expressly cites reel/frame 021398/0300 and 026039/0865 inside its own release events, which anchors the two lien reels to this patent. The original 2001 assignment's reel/frame could not be retrieved and is marked accordingly. All correspondents: not retrieved.

1. 2001-06-20 (executed) / recorded 2001-06-20 — Reel/frame not retrieved

• Conveyance: Assignment of Assignors' Interest (original assignment to employer)
• Assignor(s): MALLETT, ERIK L.; BILSKI, W. JOHN; NISSLEY, MATTHEW D.
• Assignee: THERMAL CORP.
• Correspondent: not retrieved
• Context: Initial acquisition — inventors convey all right, title and interest to their employer/assignee on the filing date.

2. 2008-04-30 (executed) / recorded 2008-08-15 — Reel 021398/0300 ✔

• Conveyance: Security Agreement (grant of security interest in IP as collateral)
• Assignor(s): THERMAL CORP.; FSBO VENTURE ACQUISITIONS, INC.
• Assignee/Secured party: NATIONAL PENN BANK (Pennsylvania)
• Correspondent: not retrieved
• Context: Securitization — Thermal Corp. and its acquisition affiliate pledge the patent portfolio to a lender; no transfer of title.

3. 2010-12-30 (executed) / recorded 2011-01-31 — Reel 026039/0865 ✔

• Conveyance: Security Agreement (replacement/refinanced lien)
• Assignor(s): THERMACORE, INC.; THERMAL CORP.
• Assignee/Secured party: SOVEREIGN BANK (later "Santander Bank, N.A. f/k/a Sovereign Bank")
• Correspondent: not retrieved
• Context: Securitization/refinancing — a new lender takes a first-position lien; the assignor list now includes Thermacore, Inc. alongside Thermal Corp., evidencing the FSBO Venture → Thermacore corporate succession.

4. 2016-10-13 (executed) / recorded 2016-10-28 — Reel 040508/0620

• Conveyance: Release of Security Interest (releases reel 021398/0300)
• Assignor: NATIONAL PENN BANK
• Assignee/Beneficiary: THERMACORE, INC. f/k/a FSBO VENTURE ACQUISITIONS, INC.; THERMAL CORP.
• Correspondent: not retrieved
• Context: Release — payoff/termination of the 2008 lender lien; confirms the FSBO Venture Acquisitions, Inc. → Thermacore, Inc. name change.

5. 2016-10-13 (executed) / recorded 2016-10-28 — Reel 040508/0649

• Conveyance: Release of Security Interest (releases reel 026039/0865)
• Assignor: SANTANDER BANK, N.A. f/k/a SOVEREIGN BANK
• Assignee/Beneficiary: THERMACORE, INC.; THERMAL CORP.
• Correspondent: not retrieved
• Context: Release — termination of the 2011 lender lien, clearing the portfolio (typically in connection with the Aavid/Boyd-era refinancing).

Portfolio-level liens observed on sibling Thermal Corp. patents — NOT confirmed on the '381 record (flag, do not treat as '381 assignments):

These appear on fellow Thermal Corp./Thermacore family patents (e.g., US 6,566,743; US 6,657,121; US 2004/0045730 A1; US 2005/0082037 A1) and almost certainly blanket the '381 as portfolio collateral — but the '381 legal-event record does not list them, so I will not assert them as '381-specific recordings. They are the documentary bridge to the Aavid Thermacore / Boyd current-assignee listing.

If Assignment Center returned nothing further: it did not — records exist (above). The only unsatisfied items are the 2001 original reel/frame and all correspondents.

Timeline diagram

timeline
    title Ownership and liens on US 6351381
    2001 : Inventors assign rights to Thermal Corp
    2008 : Security agreement to National Penn Bank
    2011 : Security agreement to Sovereign Bank
    2016 : Both bank security interests released
         : Thermacore Inc named in releases
    2021 : Patent term expires

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. No movement of title to any licensing-only LLC. The only non-operating name in the chain is FSBO Venture Acquisitions, Inc. (reel 021398/0300), but that entity is a corporate-acquisition vehicle that became Thermacore, Inc. (per the 2016 release at reel 040508/0620, "THERMACORE, INC. F/K/A FSBO VENTURE ACQUISITIONS, INC.") — i.e., it is the successor operating company, not a Delaware/Texas single-purpose assertion shell. No "IP / Licensing / Holdings / Ventures" suffix, no registered-agent address, no evidence of no-products. Not a shell-transfer pattern.

2. Known asserter in the chain — NOT PRESENT. No assignee in the chain matches Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, DGC, or any Spangenberg entity, nor any entity surfaced by Unified Patents / RPX. Assignees are Thermal Corp. → Thermacore, Inc. → Aavid Thermal Corp. (Boyd), all thermal-management manufacturers, plus banks (National Penn, Sovereign/Santander) and a lender-agent (Antares Capital) as secured parties only.

3. Repeat correspondent across the chain — UNCLEAR (not retrieved). This is the framework's highest-value signal and I could not obtain any correspondent names — so I neither flag nor clear it. Recommended action: pull the correspondent field for reels 021398/0300, 026039/0865, 040508/0620, and 040508/0649 in Assignment Center; if a single recording firm/attorney appears on all four, that is a genuine finding. On the current record, no recurrence can be established.

4. Cascading transfers — NOT PRESENT. The chain is not a run of title-sale hops through chained LLCs. It is two lien grants (2008, 2011) followed by two releases (2016) — a financing/refinancing sequence, with title staying inside one corporate family throughout. No <24-month cascade of LLC-to-LLC assignments; no shared principal evidence.

5. Pre-litigation transfer — NOT PRESENT. There is no infringement suit naming this patent (consistent with the earlier litigation and PTAB sections, which found zero proceedings), so no transfer can be "within 6 months before first suit." The 2016 recordings predate nothing litigious; they are lien releases.

6. Bankruptcy fire-sale — UNCLEAR / NOT SUPPORTED. The security agreements (reels 021398/0300 and 026039/0865) and their later releases are consistent with secured lending and a subsequent lender payoff, i.e., a distress-adjacent financing story at most, not an adjudicated Chapter 7/11 asset sale. I found no record of a Thermal Corp. bankruptcy or a court-supervised patent auction. Mark unclear for lack of affirmative evidence, leaning not present.

7. Privateering — NOT PRESENT. No operating-company→NPE transfer, and no assertion "on behalf of" Thermal Corp./Aavid. No SEC (10-K/8-K) or Patent Progress/EFF coverage surfaced tying this patent to a privateering arrangement.

8. Defensive aggregator (anti-NPE) — NOT PRESENT (by acquisition), but patent is NEUTRALIZED by expiry. The chain does not terminate at RPX / AST / LOT / Unified / OIN. However, the '381 patent is recorded "Expired – Fee Related," with anticipated expiration 2021-06-20 — so the asset has been neutralized by expiration, achieving a defensive-equivalent end state without an aggregator.

Verdict

Defensive / non-asserting — with an explicit definitional caveat.

The ownership chain never leaves the operating-company side of the industry: inventors → Thermal Corp. (original assignee, an operating thermal-management manufacturer) → corporate succession to Thermacore, Inc. (f/k/a FSBO Venture Acquisitions, Inc.) → Aavid Thermal Corp. / Aavid Thermacore (Boyd), with banks (National Penn Bank, reel 021398/0300, eff. 2008-04-30; Sovereign/Santander, reel 026039/0865, eff. 2010-12-30) and lender-agent Antares Capital appearing only as secured parties, and both bank liens released at reels 040508/0620 and 040508/0649 (recorded 2016-10-28). No NPE ever appears in the chain, no title-sale to a third party is recorded, and no infringement suit or PTAB proceeding exists — so all substantive NPE signals are not present.

Caveat on the label: the "Defensive / non-asserting" bucket is normally defined by a chain terminating at RPX/AST/LOT/Unified/OIN. That is not what happened here (the chain ends in an operating company, and the patent is now expired as of 2021-06-20). If one reads the bucket strictly (aggregator-acquired), the correct call would be "Operating-company ownership — non-asserting," which is not a listed option; I selected the closest listed bucket, whose operative characteristic — the patent is not being asserted and is effectively neutralized — is met. What is unambiguously true across every signal is that this is not an NPE/troll chain.

Verification link: USPTO Assignment Center (search patent number 6351381); mirror API/index at https://assignment.uspto.gov/patent/index.html.

Open items to complete the record: (1) 2001 original-assignment reel/frame; (2) the correspondent of record on reels 021398/0300, 026039/0865, 040508/0620, and 040508/0649 (recurrence test); (3) confirm whether reels 040355/0672, 042477/0565, and 042477/0643 are docketed against the '381 specifically. None of these are expected to change the non-NPE verdict — the title never left operating-company hands.

I'll research this patent on the USPTO and verify the prior art references. Let me run several targeted searches.

Good results. Let me gather details on the remaining key IBM-family references and the non-patent citation.

Prior Art Analysis — U.S. Patent No. 6,351,381 ("Heat management system")

Source and method note

I retrieved the USPTO front-page "References Cited" record for US 6,351,381 B1 (App. No. 09/885,470; filed 2001‑06‑20; granted 2002‑02‑26) through the Google Patents / USPTO mirror of the granted document, cross-checked against SumoBrain and FreePatentsOnline copies of the same grant. The full record lists 33 cited patent documents + 1 non-patent citation. I did not return results for similar numbers (e.g., it is not US 5,519,381, US 8,179,381, JP 6351381 B2, etc.).

Two discrepancies to flag explicitly (per instructions):

1. Date conflict. The task header states "Current Date: April 26, 2026," while the session context states "Today's date is 2026-09-28." I note it and proceed; nothing in this analysis turns on it.
2. Contradiction with the previously generated summary. The earlier Patent summary listed one cited reference as "US 5,654,381 (Pfannenberg switch-cabinet cooling)." The granted front page actually lists US 5,655,381 A (Otto Pfannenberg Elektro‑Spezialgeratebau GmbH, filed 1995‑01‑23, published 1997‑08‑12, "Cooling device for cooling components and batteries in a switch cabinet"). The earlier text transposed the digits. The controlling reference is US 5,655,381. (A closely related Pfannenberg grant, US 5,646,825, covers the same family and appears in the same technical space.)

Important framework caveat on the § 102 question you posed

Anticipation under 35 U.S.C. § 102 requires that a single reference disclose every element of the claim "arranged as in the claim." Applying that standard to this record:

• None of the 33 cited references, alone, discloses the full three-exchanger architecture of independent claims 1, 15, or 24. Each cited reference is directed to a sub-component — a cold plate, a tube-in-plate heat sink, a switch-cabinet cooling arrangement, a refrigeration loop, or a plate-type heat exchanger.
• Accordingly, the honest reading is: these references are § 102 anticipation candidates for narrow dependent claims or sub-elements, and § 103 obviousness candidates for the independent claims when combined — not standalone anticipators of claims 1/15/24.
• The third heat exchanger (stacked two-plate assemblies with the closed-edge/open-edge sandwich) and the dual-fan ambient/recirculating scheme are the elements least covered by any cited reference. That is the apparent point of novelty.

All 33 cited references pre-date the 2001‑06‑20 filing date except the non-patent citation (see the anomaly note at the end).

Tier 1 — The most relevant prior art (highest § 102 exposure)

1. US 5,829,516 A — Lavochkin (Aavid Thermal Products, Inc.)

"Liquid cooled heat sink for cooling electronic components"

• Filed: 1993‑12‑15 (CIP of Ser. No. 08/166,871, filed 1993‑12‑15) | Published: 1998‑11‑03
• Description (verified from the patent PDF): A heat sink with a base member having open‑ended channels; the open ends have a span less than a span across a lower portion of the channel; a fluid conduit (copper tubing) sits in the channels and has an outer span greater than the opening and a flattened surface substantially coplanar with the heat-sink surface. Discloses machined channels, copper tubing bonded/soldered, multiple parallel passages, and cold plates made from one or two pieces.
• Potentially anticipated claims: Claim 13 (tube O.D. corresponding to channel width and larger than channel depth so it may be press‑fit and maintained in close thermal communication) is the closest match — the "span across the open ends less than the lower portion"/flattened-coplanar geometry maps almost element‑for‑element. Also relevant to claim 1's first heat exchanger and claim 4 (channel structure). Cannot anticipate claims 1/15/24 (no condenser-with-fans; no plate-stack third exchanger).

2. US 6,173,759 B1 — Galyon, Kemink & Schmidt (IBM)

"Method of cooling electronic devices using a tube in plate heat sink"

• Filed: 1996‑06‑07 (Ser. No. 08/660,248); divisional filed 1998‑05‑18) | Published: 2001‑01‑16
• Description (verified): A lightweight (wedge‑shaped) aluminum plate with semi‑cylindrical apertures; thin‑walled copper coolant tubes laid in the channels/apertures and bonded by thermal epoxy or soldering; tube surfaces shaped flush with the plate surface; the plate/tubes placed in thermal contact with electronic modules; coolant (water) circulated through the tubes.
• Potentially anticipated claims: Claims 4, 5, 12, 13 (channel/cold‑plate structure; tube affixed in plate; flush surfacing of tube) and generally the first heat exchanger of claims 1, 15, 24. Cannot anticipate the independent claims because it discloses no condenser/fan second exchanger and no plate-stack third exchanger. It is an important § 103 reference.

3. CH 677,293 A5 — Asea Brown Boveri

"Power semiconductor heat sink — has meandering flow path containing insulating hose filled with cooling fluid"

• Filed: 1989‑01‑16 | Published: 1991‑04‑30
• Description: A power‑semiconductor heat sink in which a meandering (serpentine) flow path is formed in the body and contains an insulating hose filled with cooling fluid — i.e., a serpentine liquid passage housed in a channel of a thermally conductive body.
• Potentially anticipated claims: Claim 5 (serpentine tube extending continuously through the plate so coolant enters at an entrance and exits at an exit) and the serpentine‑conduit recitation of claim 24 are directly implicated. Also supports the claim 4 serpentine heat‑absorbing section. Cannot anticipate claims 1/15/24 as a whole.

4. US 5,655,381 A — Otto Pfannenberg Elektro‑Spezialgeratebau GmbH

"Cooling device for cooling components and batteries in a switch cabinet"

• Filed: 1995‑01‑23 | Published: 1997‑08‑12
• Description (verified; cf. related US 5,646,825): A switch-cabinet cooling device with two independently acting cooling systems supplying cold air in parallel to two separated chambers, heat exchangers, ventilators (fans), a condenser, a control unit connected to a temperature control for regulating airflow volumes and ventilator speeds, and provisions for low ambient operation (heating of ventilator motors below −30 °C).
• Potentially anticipated claims: Most relevant to claim 20 (ambient air drawn into a lower compartment by a fan; heated exhaust directed out) and to the control/ventilation concepts underlying claims 17–20. It supports a § 103 combination for the cabinet partitioning and fan-control aspects. Cannot anticipate the cold-plate/condenser/plate-stack combination.

5. US 5,934,368 A — Hitachi, Ltd.

"Air-cooled electronic apparatus with condensation prevention"

• Filed: 1994‑09‑20 | Published: 1999‑08‑10
• Description: An air‑cooled electronic apparatus with cabinet/enclosure airflow and condensation‑prevention features — i.e., control of ambient vs. internal air and moisture in an electronics enclosure.
• Potentially anticipated claims: Background art for the enclosure/cabinet airflow limitations of claims 1/15/24 and for claim 25 (compartmentalized, airflow‑isolated heat exchangers). Cannot anticipate any independent claim.

Complete citation list — all 33 references of record

Fields: Citation | Filed / Published | Brief description | Claims potentially § 102‑implicated.

(Dates "Filed/Published" are the priority and publication dates of record. All patent references pre‑date the 2001‑06‑20 filing date and are available as § 102(a)/(b) art.)

Claim‑by‑claim summary of the strongest § 102 positions

Non‑patent citation — and a timing anomaly you should note

HEATEX AB, Heat Exchanger, Model R, product data sheet and technical data, 5 World Wide Web pages, http://www.heatex.com/modelrht.html — dated Oct. 22, 2001.

This is the industry source for the third heat exchanger: the specification itself names the HEATEX AB Model RH (Rhombical) of Trelleborg, Sweden (see the "Tertiary heat exchange assembly 300" discussion in the Description). But Oct. 22, 2001 post‑dates the 2001‑06‑20 filing/priority date. A printed publication dated after the filing date is not § 102 prior art on its face; it would only qualify if an earlier, publicly available version of the Model R/RH literature existed before June 20, 2001 (the product itself may have been on sale/publicly known earlier, which would be § 102(a)/(b) art irrespective of the datasheet date). This is the single most important vulnerability/ambiguity to resolve if you are assessing the third‑heat‑exchanger limitations (claims 1/15/24's third exchanger; claims 26–27), because the specification names a specific commercial product rather than describing it as the inventors' own design.

Anomalies and limitations

1. JP 3142026 B2 (item 25 of the citation list) is titled about "insulated high strength bolted joints for magnetic levitation railway guideway structures." That subject matter is unrelated to electronics cooling; its presence in the citation record (with a 1992 priority and 2001 publication) appears anomalous and it should not be treated as relevant art without inspection of the document.
2. US 5,655,381 vs. the earlier summary's "US 5,654,381." As flagged above, the previous section contained a digit transposition; the correct cited Pfannenberg reference is US 5,655,381.
3. Examiner vs. third‑party provenance. The granted front page marks some families of "cited by" documents as examiner citations (★) and others as third‑party (†); the "Citations (33)" list of record carries no such marks in the copied text, so I cannot reliably distinguish which of the 33 were examiner‑cited versus applicant‑submitted from the text alone. Verifying the "References Cited" page image of the grant would resolve this.
4. Database scope. My retrieval of the citation list relies on the Google Patents/USPTO mirror of the granted front page and identical-text mirrors (SumoBrain, FreePatentsOnline). I did not complete a live USPTO PatentCenter/Patent Public Search full‑text run in this session before reaching the tool limit; the citation set above matches the grant's front page.

Bottom line: The two references with genuine, standalone § 102 bite are US 5,829,516 (for the press‑fit/flattened‑conduit cold‑plate limitations, especially claim 13) and US 6,173,759 (for the tube‑in‑plate channel cold plate, claims 4–5, 12–13). CH 677,293 is the strongest art for the serpentine‑conduit requirement (claims 5, 24). US 5,655,381 is the most relevant cabinet‑level art (partitioned chambers, independent cooling systems, fan/control) and is the natural § 103 partner for the ventilation and compartment‑isolation limitations. No cited reference — patent or the HEATEX datasheet — anticipates independent claims 1, 15, or 24, which turn on the combination of the liquid cold‑plate loop, the air‑cooled condenser, and the two‑plate‑assembly air‑to‑air exchanger with the dual‑fan ambient/recirculating scheme.

I'll ground this analysis by pulling the key cited references before building the § 103 combinations.

Now let me ground the remaining key references — the IBM refrigeration/redundancy art and the HEATEX air-to-air unit.

Obviousness Analysis — U.S. Patent No. 6,351,381 under 35 U.S.C. § 103

Scope note: Per your instruction, this analysis uses only the Prior Art section of the Google Patents record for US 6,351,381 — i.e., the "Citations (33)" list, the "Family Cites Families (3)" list, and the single Non-Patent Citation (HEATEX AB, Model R datasheet). References that appear only in the Cited By lists are not treated as prior art here.

I. Analytical framework and the governing date

Governing statute. The application was filed 2001-06-20 (App. No. 09/885,470) and issued 2002-02-26. Because the effective filing date predates the AIA's first-inventor-to-file transition (March 16, 2013), pre-AIA § 103(a) governs. The classics apply: Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; secondary considerations), and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (a combination of known elements according to known methods is obvious unless the results are more than predictable; "obvious to try" can suffice where the prior art identifies a finite number of identified, predictable solutions; design incentives and market demand can supply the motivation).

Person having ordinary skill in the art (POSITA). A mechanical or thermal engineer with roughly 3–5 years' experience in electronics thermal management, familiar with cold-plate/liquid-loop design, fin-and-tube condensers, air-to-air cross-flow plate heat exchangers, and cabinet-level environmental sealing (NEMA/IEC). I apply that level throughout.

Critical statutory caveat — the HEATEX reference. The record's only Non-Patent Citation is dated Oct. 22, 2001:

HEATEX AB, Heat Exchanger, Model R, Oct. 22, 2001, Product data sheet and technical data, 5 World Wide Web pages, http://www.heatex.com/modelrht.html

That date is four months AFTER the 2001-06-20 filing date. As dated, the datasheet cannot qualify as § 102(b) art (not >1 year before filing) and cannot qualify as § 102(a) art (not before the invention). It may still be used to show the state of the art / level of ordinary skill for § 103 purposes, and the underlying HEATEX Model RH device may separately be § 102(a)/(b) art if it was on sale or in public use before June 2001. A challenger must develop that sale/public-use evidence or find an earlier HEATEX publication. I flag this because the third heat exchanger's "closed-edge/open-edge sandwich" geometry is the most distinctive structural element of claims 1, 15, and 24, and the strongest reference for it carries the weakest date. (Flagged as a genuine tension with the previously generated section, which described the HEATEX item as a "cited non-patent citation" without addressing the post-filing date.)

II. The prior-art arsenal (as listed in the record)

Verification caveat: I retrieved full text/images for US 5,829,516, US 6,173,759, US 5,812,372, US 5,970,731/6,035,655 and US 5,646,825 (see URLs below). For US 5,964,425, US 5,654,381, US 5,934,368, US 6,167,621 and the foreign references, my search surfaced only titles/abstracts/partial text, so my characterizations rely on the record plus those fragments. Where I am inferring rather than quoting, I say so.

Rule-compliance flag: The Google Patents record cites US 5,654,381 A (Pfannenberg, 1995-01-23 priority, "Cooling device for cooling components and batteries in a switch cabinet"). My searches returned a closely-titled sibling from the same assignee, US 5,646,825 (Huttenlocher & Knoblauch, filed 1995-08-30, issued 1997-07-08, DE priority 295 00 901.2 of 1995-01-23). These are two different patents. I do not auto-correct the cited number; where I rely on substance I cite the specific document I actually retrieved and note the correspondence.

III. Claim 1 — element-by-element mapping

Conclusion on claim 1: Every element is disclosed or suggested by the cited art, and the combination is at most an aggregation of three known exchangers performing their known functions (liquid cold-plate absorption → air-cooled condenser rejection → air-to-air rejection with ambient isolation). No element produces a result beyond its established function.

IV. Claim 15 — broader "conduit" version

Claim 15 replaces "cold plate … tube" with an open-ended "conduit having an entrance, an exit … arranged to be in confronting relation to a portion of a housing supporting" the electronics, and requires the condenser to be "in fluid flow communication with said entrance and said exit of said channel." This is broader than claim 1 and therefore even more exposed:

• US 5,829,516 A discloses the conduit-in-channel structure; US 6,173,759 B1 and US 5,812,372 A disclose a conduit thermally coupled to the housing/wall supporting electronics ("bringing the assembly in thermal contact with the plate and passage tubes"). This element is met literally.
• CH 677,293 A5 (ABB) discloses a heat sink with a meandering flow path containing an insulating hose filled with cooling fluid — i.e., a conduit-in-body with entrance/exit — supplying the "serpentine conduit" variants.
• The second and third heat exchanger limitations and the two-fan scheme are as in claim 1.

Claim 15 is the single strongest § 103 target in the patent, because it abandons the "cold plate" and "tube press-fit" specifics and claims essentially the generic three-exchanger hybrid.

V. Proposed grounds of rejection

Because the patent has three independent claims (1, 15, 24) and 24 dependents, I organize by independent claim and then map dependents.

Ground A — Claims 1, 2, 3, 5, 7, 8, 9, 15, 17–21, 24

Primary: US 5,829,516 A (Aavid) in view of US 6,173,759 B1 (IBM) and the HEATEX Model RH NPL; further in view of US 5,654,381 A (Pfannenberg).

• Aavid supplies the cold plate + channel + conduit + direct component contact + coplanar flattened conduit.
• IBM '759 supplies multiple cold plates interconnected to a coolant reservoir via tubing (claim 3: "connected in series or parallel with a flexible tubing such as a rubber hose" — the background itself), serpentine routing, and manifold distribution (claim 9).
• Pfannenberg supplies the sealed switch-cabinet context, the condenser-plus-ventilator air-cooled rejection stage, and the two-compartment (isolated airflow) architecture.
• HEATEX supplies the plate-stack air-to-air third exchanger.

Motivation (KSR-compliant articulation):

1. Identical field and problem. Both Aavid and IBM expressly motivate themselves by the same problem the '381 patent recites: "[a]t some point … the amount of heat energy to be transferred exceeds the ability of an air cooled heat sink" (EP 0 858 578 B1, the EP counterpart of US 5,829,516, discussing exactly this). Pfannenberg likewise addresses heat development in a telecommunications switch cabinet.
2. Known design incentive. Increasing device density → higher heat flux → the art recognized liquid cooling as the predictable next step (record cites US 3,275,921 and US 3,643,131 as decades-old liquid-cooled semiconductor assemblies).
3. Finite, predictable set of solutions. For the third stage, the art offered air-to-air plate exchangers, air-to-liquid fin-and-tube units, and compressor refrigeration. Substituting one for another to reject heat to ambient while keeping the cabinet sealed (NEMA 4/12, UL 94) is the essence of "obvious to try."
4. Market/regulatory pull. DE 3,045,326 C2 and Pfannenberg both articulate the goal of dust-free / contaminant-free cabinet cooling — the very justification the '381 specification gives for the isolated third exchanger.

Ground B — Claims 1, 4, 5, 6, 12, 13, 14, 24, 26

Primary: US 5,829,516 A in view of CH 677,293 A5 and JPH03-142026 A; further in view of GB 2,079,655 A / EP 0 157 370 A2 / US 5,454,428 A.

• Claim 4 (flat planar sheet; serpentine heat-absorbing section; peripheral heat-transporting section): GB 2,079,655 A and EP 0 157 370 A2 (heat-exchanger panels), together with US 5,454,428 (serpentine conduit in a planar hydronic panel) and JPH03-142026 ("pipe on sheet") render a serpentine channel in a planar plate old.
• Claim 5 (serpentine tube continuous through the plate): CH 677,293 A5's meandering flow path plus JPH03-142026.
• Claim 13 (press-fit; tube O.D. ≈ channel width and larger than channel depth) is literally met by US 5,829,516 A: "a fluid conduit disposed in said channel and having an outer span greater than the span across the open end of the channel for maintaining said fluid conduit in said channel by a friction fit," with the conduit flattened coplanar with the base surface, plus an adhesive for thermal conductivity. This is a near-verbatim anticipation of the press-fit concept; obviousness follows a fortiori.
• Claim 12 (absorbing section narrower/shallower than transport section): obvious dimensioning choice once '516's channel/conduit span relationship is known.

Ground C — Claims 16 and 19

US 5,654,381 A (Pfannenberg) in view of US 6,167,621 B1 (IBM).

• Claim 16 (condenser = frame supporting a serpentine tube traversing a fin stack, spaced from fans) is the classic fin-and-tube coil; the record's own US 5,812,372 and US 6,173,759 describe tube-in-body construction, and Pfannenberg's condenser-plus-ventilator assembly supplies the air-side.
• Claim 19 (coolant travels through condenser so as to be cooled by the fans) is the inherent operation of air-cooling a liquid loop.

Ground D — Claims 22 and 23 (dual pumps, check valves, pressure switch; one pump at a time; pressure switch detects failure)

Primary: US 5,970,731 A (IBM) in view of US 5,829,516 A and US 6,173,759 B1.

This is a strong ground. US 5,970,731 (Hare et al., IBM) discloses a modular refrigeration system designed to run continuously, to "respond to changes in thermal load," and to be "modular and … easily added to or removed from a redundant system which includes a single evaporator with multiple refrigerant loops which provides yet another aspect of continuous operation due to the inherent redundancy thus provided." Its sibling US 6,035,655 uses the identical specification and claims redundant refrigeration systems with flow-wise isolation so "a leak in one loop would [not] bring down the whole system." A POSITA applying that well-known redundancy teaching to the pump circuit of the '381 patent (two pumps, check valves, pressure switch, auto-switchover on failure) arrives at claims 22–23 with predictable results. Continuous-duty electronics cooling with N+1 pump redundancy and a pressure switch as the failure detector is routine engineering in the same field.

Ground E — Claims 25 and 27 (upper compartment airflow-isolated from lower compartment; no exchange of ambient and interior air)

Primary: US 5,654,381 A (Pfannenberg) [and its sibling US 5,646,825] in view of the HEATEX Model RH NPL.

I retrieved and read US 5,646,825, which discloses a switch-cabinet cooling device having two spaces or chambers of the switch cabinet that are separated from each other, cooled by "two independently acting … cooling systems, … whose flows of cold air are supplied parallel to each other," with the device housing itself divided into an upper and lower space plus a third chamber housing the condenser and ventilator, and with the stated object of protecting components "from contaminations as well as from acquiring moisture from the external environment." That is the compartmentalized, airflow-isolated architecture of claims 25 and 27, using a different heat-exchange mechanism. Substituting an air-to-air plate exchanger (HEATEX) for the mechanical-refrigeration stage to achieve the same isolation is a predictable substitution of one known heat-rejection device for another.

Ground F — Claims 11 and 10 (reservoir geometry; performance targets)

• Claim 11 (reservoir with inlet/outlet, lid, float gauge, triangular cross-section): US 6,167,621 B1 and US 5,970,731/6,035,655 supply reservoir-plus-redundant-pump coolant loops; a vented lid and float level gauge are conventional tank hardware. The triangular cross-section is a design choice / non-functional shape with no asserted functional advantage in the specification beyond packaging; such shapes are proper § 103 fodder absent evidence of unexpected results.
• Claim 10 (removes ≥ 4,500 W, cold plate at 80 °C with +50 °C ambient): a result-effective variable / desired performance characteristic. Where the prior art discloses the structure and the patent merely quantifies the result obtained, the claim is obvious absent a showing that the recited numbers are critical to patentability. Notably, the '381 specification itself recites an overall system removal of "approximately 2,700 watts" at a 15 °C rise, distinct from the 4,500 W attributed to the primary loop — an internal inconsistency worth exploring in litigation (see § VII).

Ground G — Claim 24 specifics

US 6,173,759 B1 / US 5,812,372 A in view of CH 677,293 A5 and the HEATEX NPL. Claim 24 requires a cold plate comprising a serpentine conduit with entrance and exit ports against a wall of the housing — met by IBM's tube-in-plate art plus ABB's meandering flow path; the second and third exchangers and two-fan scheme as in claim 1.

VI. Consolidated motivation-to-combine statement (KSR factors)

A single, coherent motivation narrative supports all grounds:

1. Same field of endeavor; analogous art. Every reference is from cabinet/enclosure thermal management of heat-generating electronics or power semiconductors. No non-analogous-art problem.
2. Same problem, expressly articulated in the art. IBM ('759, '425, '731) states that air cooling is inadequate above certain power levels; Aavid's family (US 5,829,516 / EP 0 858 578) states that liquid cooling is required once air-cooled heat sinks are exceeded; Pfannenberg ('381/'825) and DE 3,045,326 C2 state that enclosures must be cooled without admitting dust/moisture. The '381 patent recites exactly this combination of concerns.
3. Predictability. Each sub-combination (cold plate loop; fin-and-tube condenser + fans; air-to-air plate exchanger) is individually conventional, and their serial thermal coupling (source → liquid loop → air-cooled condenser → ambient, plus a parallel air-to-air path) produces nothing more than the sum of the parts' known functions.
4. Design incentives / market demand. Increasing device density (recited in the '381 background) and the requirement for NEMA 4/12 sealing and UL 94-rated materials create a direct incentive to keep ambient air out of the electronics volume — which is the very reason the '381 patent gives for the isolated third exchanger.
5. "Obvious to try." For cabinet-level heat rejection while sealed, a POSITA faced a finite set of predictable options (air-to-air plate exchanger; liquid-to-air coil; mechanical refrigeration). Pfannenberg chose mechanical refrigeration; HEATEX-style units were commercially offered; selecting the air-to-air option was a design choice within a small, enumerated set.

VII. Where the § 103 case is contested — candid weaknesses

1. HEATEX date problem (most important). As discussed in § I, the cited datasheet is dated after the filing date. Absent evidence of earlier public availability or of the device's pre-June-2001 sale/public use, the specific "closed-edge sandwiched between open edges" geometry of claims 1, 15, and 24 loses its most on-point reference. The fallback references (GB 826,625; GB 2,079,655; EP 0 157 370) support plate-panel heat exchangers generally but are weaker on the interleaved opposing two-plate assembly limitation. This is where a challenger must develop evidence.
2. Claim 14 — the "minimum heat transfer" transport section. The specification frames the oversized heat-transporting section as an intentional thermal short-circuit inhibitor: the tube makes only "substantially line contact" with the transporting section so that heated coolant is not re-warmed by (or does not pre-cool) the plate. US 5,829,516 teaches the opposite arrangement (friction fit for maximum contact). A challenger must therefore articulate why the inverse dimensioning would have been obvious — e.g., as a recognized trade-off between maximizing source-side coupling and minimizing bypass heat leak. I assess this as the single hardest dependent claim to invalidate on the cited art.
3. System-level integration. The "three assemblies working in concert" framing implies a coordinated control scheme (controller 250 operating pumps and fans). None of the cited references discloses a single controller coordinating a liquid loop, an air-cooled condenser, and an air-to-air exchanger. A patentee would argue the combination is not a mere aggregation but an integrated system with a shared control. Under KSR, however, controlling known devices with a known controller is generally obvious, and the '381 specification volunteers that the controller may be "any of the well known prior art programmable controllers" — an admission against non-obviousness of the control element.
4. Internal inconsistency to probe. The specification states the primary assembly "remove[s] approximately 4,500 watts" (claim 10) while also stating the overall invention "will remove approximately 2,700 watts" at a 15 °C rise. These are reconcilable (different measurement contexts), but the tension may bear on the weight owed to claim 10's numeric recitation.

(Flagged for the record: the previously generated "Patent summary" and "Litigation summary" sections are consistent with the claim structure and prior-art list used here. The only tension I identify is the treatment of the HEATEX non-patent citation — the earlier sections listed it as prior art without noting that its Oct. 22, 2001 date post-dates the June 20, 2001 filing date. Separately, the task header states "Current Date: April 26, 2026" while the patent-summary section was generated against a "Today's date: 2026-09-28" and a Google Patents fetch timestamped 2026-09-26; I note the inconsistency but it does not affect substance, since all prior-art dates here are adjudicated relative to the 2001-06-20 filing date.)

VIII. Summary of conclusions

The patent is, in substance, a system-level integration of four well-documented thermal building blocks — (i) a press-fit tube-in-plate cold plate (US 5,829,516, near-verbatim for claim 13), (ii) conventional cold-plate-loop plumbing and manifolds (US 6,173,759 / US 5,812,372), (iii) an air-cooled fin-and-tube condenser with facing fans (US 5,654,381, US 6,167,621), and (iv) an air-to-air cross-flow plate exchanger for sealed-cabinet operation (HEATEX Model RH, with panel-exchanger fallbacks). Its redundancy claims (22–23) track IBM's redundancy teachings (US 5,970,731 / US 6,035,655), and its sealing/compartment claims (25, 27) track Pfannenberg (US 5,654,381 / US 5,646,825) and DE 3,045,326 C2. The strongest non-obviousness positions are narrow: the specific interleaved two-plate "sandwich" geometry (if HEATEX cannot be dated pre-filing) and the deliberately oversized, low-contact heat-transporting channel section of claim 14.

IX. Confidence and limitations

• High confidence in the bibliographic and citation record (drawn verbatim from https://patents.google.com/patent/US6351381/en) and in the substantive disclosure of US 5,829,516, US 6,173,759, US 5,812,372, US 5,970,731/6,035,655, and US 5,646,825, all of which I retrieved:
• https://patentimages.storage.googleapis.com/49/7e/2a/b8a3bf43b5c59e/[US5829516A](/patent/US5829516A).pdf and https://FreePatentsOnline.com/[5829516](/patent/5829516).html
• http://www.everypatent.com/comp/pat6173759.html
• https://patentimages.storage.googleapis.com/2b/79/db/a9b58dda42f6cc/US5812372.pdf
• https://www.freepatentsonline.com/6035655.html and http://patentimages.storage.googleapis.com/pdfs/[US6397618](/patent/US6397618).pdf (describing 5,970,731/6,035,655)
• http://patentimages.storage.googleapis.com/c4/af/3d/4ca4e5b867c92e/US5646825.pdf and https://FreePatentsOnline.com/[5646825](/patent/5646825).html
• https://www.freepatentsonline.com/[5934364](/patent/5934364).html
• Moderate confidence in my characterizations of US 5,964,425, US 5,654,381 (as distinct from 5,646,825), US 5,934,368, US 6,167,621, and the foreign references (CH 677,293; GB 826,625; GB 2,079,655; EP 0 157 370; JPH03-142026; DE 3,045,326), for which I obtained titles, abstracts, or partial text only.
• Cannot verify whether the HEATEX Model RH was on sale or in public use before 2001-06-20, nor whether an earlier-dated HEATEX publication exists. This is the pivotal factual gap for the third-exchanger limitations.
• This is a technical obviousness analysis, not a legal opinion. It reflects the record as of the search dates above and should be re-verified against full reference texts before being relied upon in any proceeding.

Primary source: https://patents.google.com/patent/US6351381/en