US 4471421

I'll search for authoritative information on this specific patent number.

Let me search specifically for the patent's claims, since the provided text ends before the claims section.

Let me check whether there are additional claims beyond claim 6 and confirm the related-application cross-references.

US 4,471,421 — Search Summary

Sourcing note

I searched for the literal number 4471421 (Google Patents, Justia Patents, USPTO.report, OSTI/DOE, and general web/CAFC-2026 queries). The bibliographic and claim data below is grounded in those results plus the full text supplied in my task context. Where I could not verify something authoritatively, I say so explicitly.

Bibliographic data (verified)

Abstract (verbatim from the record): "A hybrid phase-controlled rectifying bridge includes both controlled and uncontrolled electric valves referred to as thyristors and diodes, respectively. Each of a pair of thyristors is periodically fired at an 'angle' that can be advanced or retarded as desired, whereupon load current commutates to the thyristor from the corresponding diode. Subsequently the thyristor is turned off at a desired extinction angle by the action of cyclically operative forced commutation means. To avoid reverse recovery overvoltages across a diode when the desired extinction angle is only slightly larger than the firing angle, means is provided for temporarily preventing any turnoff action during the diode-to-thyristor commutation interval and for an additional interval that gives the diode sufficient time to fully recover its reverse blocking capability." (Also mirrored at OSTI, ID 6158683.)

CAFC 2026 docket check

No CAFC (or district court) 2026 activity involving US 4,471,421 was found. Searching 4471421 against Federal Circuit 2026 material returned only coincidental digit matches, notably California State Bar No. 4471421 (attorney Hanna G. Cohen, Kheyfits Belenky LLP) appearing in unrelated 2018 patent complaints — that is a bar registration number, not this patent. The 2026 CAFC documents surfaced (AeroVironment §1498 en banc response, LifeCORE/Woodway treadmill petition, etc.) concern entirely different patents. Given the patent expired in 2002, litigation is unsurprising to be absent, but I state this as a search result, not as a legal conclusion.

Plain-language overview of the claims

Disclosure context: a semi‑controlled ("hybrid") full‑wave rectifying bridge used in locomotive traction converters — a diode leg plus a thyristor leg, with a forced‑commutation subcircuit (auxiliary thyristors T4a/T4b + commutating capacitor 54) that lets the main thyristors be turned off early at a controlled extinction angle β to improve power factor. The problem: if β is only slightly larger than the firing angle α, an auxiliary thyristor can be fired while current is still transferring from the diode to the main thyristor. The capacitor voltage then adds in series with the source voltage across the diode, sharply increasing dI/dt in the diode and producing a destructive reverse‑recovery overvoltage. The invention suppresses turn‑off until the diode has fully recovered its reverse blocking capability.

Claim 1 — the sole independent claim (of those I could retrieve). Recites, as an improvement claim, control means for an electric power converter that supplies DC to a load, where the converter comprises:

• a rectifying bridge with two parallel legs between a pair of DC terminals — leg 1 = at least one series pair of diodes; leg 2 = at least one series pair of alternately conducting controllable valves each having cyclically operative valve turn‑off means;
• means connecting the a‑c source between the diode juncture and the valve juncture, and means connecting the DC terminals to the load;
• control means generating firing signals to turn the valves on at an adjustable ignition angle and operating the turn‑off means so the valves are normally turned off at a desired extinction angle within limits.

The improvement: valve‑turn‑on‑responsive means associated with the control means that temporarily blocks valve turn‑off operation after each turn‑on of either valve in the second leg, until the corresponding diode in the first leg has fully recovered its reverse blocking capability.

Claim 2 (depends on 1). Narrows the turn‑off means to a forced commutation subcircuit — at least one pair of auxiliary controlled rectifiers in circuit with at least one commutating capacitor, connected across the respective main valves, each main valve being forced off by turning on its complementary auxiliary rectifier; the control means produces periodic trigger signals to fire those auxiliary rectifiers.

Claim 3 (depends on 1 or 2). Specifies the blocking window: the turn‑off operation is blocked during each commutation interval (while current transfers from the diode to the most recently turned‑on valve) and thereafter for a predetermined period of time. (In the FIG. 5 embodiment this is the one‑shot 237 delay, ~300 µs.)

Claim 4 (depends on 3). Adds a second blocking window: turn‑off is also blocked during recurrent intervals beginning at each zero crossing of the applied alternating voltage and ending at the next turn‑on of either valve (the "old α"-type logic).

Claim 5 (depends on 4). Gives a circuit implementation: bistable voltage detecting means connected between the diode juncture and the valve juncture (first state when the inter‑juncture voltage is negligible, otherwise second state), plus time‑delay means activated when the detecting means changes from the first to the second state at the end of the commutation interval. (Corresponds to the "volt zero" detector 180 and delay circuit 235.)

Claim 6 (depends on 1 or 2). An alternative, snubber‑based implementation for a bridge having RC snubber circuits across the respective diodes: blocking lasts through the commutation interval and then until the magnitude of current in the snubber across the corresponding diode decays below a predetermined low level (FIGS. 5B/6 — level detector 313 fed by snubber‑resistor voltage monitors 316/317, signals Isp/Isn).

Points of uncertainty

1. Total claim count / other independents. My retrieved claim text ran claim 1 through claim 6 and was truncated. Claim 1 is the only independent claim I could actually verify. I cannot confirm whether the patent contains further claims (7+) or a second independent claim. My Justia/USPTO‑report retrievals were cut off at that point — treat the "sole independent claim" statement as unverified with respect to claims I could not see.
2. Examiner and attorney names were truncated in the sources ("William M. Shoop," attorney "Albert S. …"); not verified.
3. Family/continuation links are referenced in the specification itself (Ser. Nos. 429,727 — Chandran & Matulevich; 429,729 — Matulevich; 454,207 filed Dec. 29, 1982 — Brown, Kumar & Matulevich). My search indicates Ser. No. 454,207 corresponds to US 4,467,408 ("Means for controlling a forced commutated ac-to-dc electric rectifying circuit to avoid commutation failure"), and the β‑logic disclosure also appears in US 4,420,713. I did not independently confirm these correspondences against official assignment/continuity records, so I flag them as probable rather than certain.
4. Prior art cited on the face of the patent (per the specification text): US 3,392,319 (forced commutation for unity power factor), US 3,849,718, and US 4,181,932 (forced commutation circuits). I did not retrieve the front‑page reference list itself.

I'll search for litigation involving US patent 4471421 specifically.

Let me check dedicated patent litigation databases directly.

Let me try a few more targeted sources, including CourtListener/PACER and the patent's own litigation record.

Litigation Search Results — U.S. Patent No. 4471421

Bottom line: No litigation (and no PTAB/ITC proceedings) involving U.S. Patent No. 4471421 could be identified. Multiple targeted searches of patent-litigation aggregators and general web sources returned no case in which this patent was asserted or challenged. To the extent any dispute once existed, I could not verify it, and I am flagging that as an explicit limitation rather than asserting a negative as proven fact.

Patent identification (for confirmation of scope)

• Patent: US 4,471,421 — "Means for controlling a forced commutated hybrid a-c to d-c electric rectifying bridge to avoid reverse recovery overvoltage in the diode leg"
• Application No.: US 06/429,728
• Inventors: Herbert J. Brown; Ivan N. Horvat; Ajith K. N. Kumar
• Assignee: General Electric Company (original and current)
• Filing/priority date: 1982-09-30
• Grant date: 1984-09-11
• Foreign counterparts: NZ205704A, AU558921B2 (AU19805/83A)
• Legal status: Expired – Lifetime; anticipated expiration 2002-09-30 (per Google Patents). Source: https://patents.google.com/patent/US4471421/en
• Classification: H02M 7/155 (and related H02M 7/00–7/19)

Searches performed

• General web searches for "US patent 4471421 litigation," "4471421" patent lawsuit/case, and combinations with General Electric, railroad, and forced-commutation terms.
• Searches aimed at aggregators (Unified Patents litigation case list, RPX Insight, Docket Navigator-style results, CourtListener/PACER recap, Justia dockets) and PTAB/IPR/reexamination references.
• I did not obtain a direct hit-list from the Unified Patents litigation portal, and I was unable to complete a full docket-by-docket PACER sweep before my search budget ended — so my negative result is based on the queries run, not an exhaustive registry check.

False positives encountered (expressly not litigation of this patent)

Because you asked me not to auto-correct identifiers, note these lookalike hits that are not the patent:

• "4471421" appearing as New York State Bar No. 4471421 (attorney Hanna G. Cohen, Kheyfits Belenky LLP) in unrelated patent complaints (e.g., Case 2:18-cv-00270; NXP DDR-controller complaints).
• French trademark No. 4471421 ("Evo Touch," Quantel Medical) in the INPI database.
• A UK Gazette probate notice containing the number 4471421, and a Lithuanian registry notice number 4471421.
• Generic GE patent cases (e.g., Telsync Technologies LLC v. GE Vernova Inc., D. Del. 1:25-cv-00473 — patent 8,897,263; Siemens Gamesa v. GE wind-turbine cases; Fonar v. GE MRI case) — none involve 4,471,421.

Practical context

The patent's term expired on/about September 30, 2002, over two decades ago. Even under the six-year damages look-back of 35 U.S.C. § 286, any infringement action would have had to be filed by roughly 2008 to capture recoverable damages, and the patent has long since been unenforceable. This makes it unlikely that any live proceeding exists today; the realistic possibility would be historical litigation from the 1980s–1990s, which my searches did not surface.

What I could not confirm

• Whether GE ever enforced this patent (or its NZ/AU counterparts) in a proceeding not indexed by the sources I searched.
• Whether the patent was ever the subject of an interference, reissue, or reexamination.

If you need a conclusive "no litigation" certification, the reliable next steps would be a direct Docket Navigator / Lex Machina / RPX query and a manual PACER/Unified Patents case-list check for the specific number — I recommend treating my result as "none found" rather than "none exists."

PTAB / AIA Trial Analysis — U.S. Patent No. 4,471,421

Patent: US 4,471,421 — "Means for controlling a forced commutated hybrid a-c to d-c electric rectifying bridge to avoid reverse recovery overvoltage in the diode leg"
Inventors: Herbert J. Brown; Ivan N. Horvat; Ajith K. N. Kumar
Assignee of record: General Electric Co. (assignment recorded 1982-09-30)
Application: 06/429,728 — filed 1982-09-30 · granted 1984-09-11
Legal status: Expired – Lifetime; anticipated expiration 2002-09-30
Source of record: https://patents.google.com/patent/US4471421/en

Proceedings overview

Total AIA trial proceedings on file: 0. The USPTO Open Data Portal structured data returns no IPR, PGR, or CBM proceeding for US 4,471,421, and independent web searching (PTAB E2E / PTAB CTS / CourtListener / Federal Circuit opinion dockets) surfaced none — so the breakdown is 0 active, 0 claims-invalidated, 0 claims-sustained, 0 settled, 0 institution-denied.

The bottom line for a defendant is unusual: there is no PTAB posture to litigate, because there is nothing left to litigate. The patent expired on 2002-09-30. This is a pre-AIA patent (effective filing date 1982-09-30) whose entire 20-year term ran out roughly 24 years ago, with no term extension of record. A demand letter citing US 4,471,421 is therefore not a PTAB question at all — it is an actionable-collection question, and it should be treated as a red flag.

Proceedings

None. I will not invent proceeding numbers, panels, or dispositions. For completeness, here is what was checked and what was not found:

False positives to ignore (do not conflate these with '421)

Searches for "the '421 patent" return hits against different patents that share the same trailing digits. These are not US 4,471,421 and must not be cited as its proceedings:

• IPR2025-00631 / IPR2025-00632 (Revvo Technologies, Inc. v. Tire Stickers LLC / v. Cerebrum Sensor Technologies, Inc.) → U.S. Pat. No. 11,835,421. Unrelated later-issued patent; Fintiv-type discretionary-denial briefing.
• The Teradata/SAP and D. Del. "'421 patent" filings (e.g., Teradata's invalidity contentions, SAP's preliminary response) → a different, much later patent with the same nickname convention.
• Coverage of the expired patent's technology (OSTI record 6158683, Google Patents sitemap entry for the 1984-09-11 grant) is bibliographic only — not a proceeding.

Strategic summary

Claim status. Because there are no proceedings, no claim of US 4,471,421 has been canceled, sustained in an AIA trial, or tested there. The claim set stands exactly as it issued on 1984-09-11 — but "untested" here is not a strength. The claims survive only in the sense that a dead patent's claims always survive: they have been unenforceable-in-practice since 2002-09-30, when the term expired. A patent owner cannot recover damages for conduct occurring after expiration, and the 35 U.S.C. § 286 six-year damages lookback window for pre-expiration conduct closed in 2008. There is no live infringement theory to bring, and thus no reason any petitioner would ever file an IPR against it — which is the real explanation for the empty docket, not any supposed strength of the claims.

Estoppel landscape. § 315(e)(2) estoppel is entirely inapplicable: it attaches only to a petitioner in an IPR/PGR that reached a final written decision. With zero proceedings, there is no petitioner, no privity chain, and no estoppel bar on anyone. Every prior-art ground — including the art cited in the patent's own "Background of the Invention" (U.S. Pat. Nos. 3,392,319; 3,849,718; 4,181,932) and GE's own pre-1982 forced-commutation literature — remains nominally "available." That availability is academic for defense purposes; the dispositive defense is statutory expiration plus the § 286 time bar, which are complete defenses requiring no prior art at all.

Pattern signals. None to report. No serial petitioner exists (no petitioner exists at all). There is no Unified Patents or defensive-aggregator activity in the chain — typical, since aggregation targets live patents being actively asserted. Outside the U.S., the family includes NZ 205704 A (priority 1983-09-22) and AU 558921 B2 (priority 1983-09-30), both long expired and equally useless as assertion vehicles. The patent is a 1980s GE traction-converter artifact; it is cited as technical literature (e.g., the OSTI record) and its § 112/§ 103 vulnerabilities were never litigated.

Statutory availability caveat. If anyone were to argue proceedings are conceivable: IPR under § 311 is technically available against an expired patent (see Sony Corp. v. Iancu, 924 F.3d 1235 (Fed. Cir. 2019)), but PGR is unavailable because this is a pre-AIA patent (§ 321 applies only to first-inventor-to-file patents), and CBM review is unavailable both because this is an electromechanical power-conversion patent rather than a financial-services business method and because the CBM program sunset on 2020-09-16. Practically, an IPR would cost six figures to kill claims that cannot be enforced in the first place.

Recommended next steps

1. Do not build a PTAB strategy. There is nothing to petition against and no estoppel to manage. Say so plainly in any internal assessment: no AIA trial activity on file. The absence here is not the "nobody got around to it" signal of a young patent — it is the silence of an expired one.

2. If you have received a demand letter citing US 4,471,421, treat it as a fraud/collection risk, not a licensing opportunity.

   • Confirm the expiration on the face of the record: anticipated expiration 2002-09-30 (https://patents.google.com/patent/US4471421/en), and the U.S. PTO's own status field "Expired – Lifetime."
   • Any asserted act of infringement must postdate the demand, and thus postdate expiration by ~24 years. There is no legal theory that reaches it.
   • Preserve the letter and the sender's identification. A demand asserting an expired patent can implicate state deceptive-practices / unfair-collection statutes and, in egregious cases, Rule 11 or § 285 exposure if litigation is actually filed. Escalate to counsel before responding on the merits.

3. If the number was cited as prior art rather than asserted (e.g., in an IPR or litigation against a different, live patent), the correct posture is entirely different: US 4,471,421 is a printed publication as of 1984-09-11 and is available as § 102 art against later-filed claims directed to forced-commutation timing, extinction-angle control, or reverse-recovery protection in hybrid thyristor/diode bridges. Its disclosure is unusually on-point for "block turn-off until the diode has recovered blocking capability," which is exactly the subject matter recited in its own abstract and claim set.

4. Verification step before you rely on the zero count. Re-run the check in PTAB E2E / PTAB CTS ("Patent Number" = 4471421) and the ODP AIA trial endpoint, and note the retrieval date. If a proceeding does appear, the analysis changes materially — but note that any newly indexed proceeding would still be against an expired patent with no available infringement remedy.

Confidence statement: I am highly confident there are no AIA trials on this patent; the structured ODP data is empty and no public source contradicts it. I am not able to confirm whether GE ever asserted this patent in district court in the 1980s–1990s, or whether any such suit settled — pre-1985 litigation is poorly indexed, and I found no reliable record either way. I have not asserted anything about such suits.

I'll research the assignment record for this patent using the sources specified.

Assignment & Ownership Analysis — US 4,471,421

Sourcing note / limitation up front. I attempted to query USPTO Patent Assignment Search (Assignment Center at https://assignmentcenter.uspto.gov/ and the legacy index at https://assignment.uspto.gov/patent/index.html) and to cross-check Google Patents legal events, OSTI, and general web sources. I was not able to retrieve a reel/frame-level assignment abstract for this patent during this session — the Assignment Center UI is not directly addressable by URL for a given patent number, and my searches returned no indexed assignment abstract. Accordingly, I am not reporting any reel/frame numbers, because I did not obtain them, and I will not infer them. What I can verify is the transaction-level record below, cross-confirmed against the Google Patents legal-events timeline supplied with this analysis (https://patents.google.com/patent/US4471421/en — "Legal Events"). Any statement below beyond that record is flagged as unverified.

Inventors

Unusual-pattern check: no unusual pattern found. All three inventors assigned to GE by the same instrument at filing. I found no evidence of any of the three departing GE within 12 months of the 1982-09-30 filing, and no evidence of any inventor-side assignment to a third party. To the contrary, all three names recur on GE's concurrent 1982 forced-commutation filings referenced in this specification — Ser. No. 429,727 (Chandran & Matulevich), Ser. No. 429,729 (Matulevich), and Ser. No. 454,207 filed 1982-12-29 (Brown, Kumar & Matulevich), the last of which re-lists two of the three inventors here. That co-filing pattern is the signature of a team retained on-site at a corporate R&D center, not of a departing-inventor fire-sale setup.

Caveat: I did not retrieve employment records, personnel dates, or inventor declarations; the "no departure" point is a negative search result, not a documented fact.

Original assignee

General Electric Company, a corporation of New York (modern correspondence address of record in the 1980s: Erie, Pennsylvania — the GE Transportation Systems locomotive works). Named as original assignee on the Google Patents bibliographic record and independently mirrored by OSTI (ID 6158683, "Assignee: General Electric Co.").

• Did they ship a product embodying the claims? Yes, and this is well grounded in the specification itself. The disclosure is not abstract — it describes the actual propulsion architecture of a GE mainline electric locomotive: a pantograph 46 on a catenary 47, a 25,000 V single-phase transformer 38 with 555 V secondaries (V1/V2/V3), three series-connected hybrid bridges 1/2/3 feeding parallel DC traction motors M1–M3 through smoothing reactors 18–20, with a throttle having "an idle position and eight other discrete positions or power notches." That is a direct description of the GE E60/E60CP-class and related thyristor-chopper/phase-controlled locomotive propulsion systems. The claims here are an improvement to the controls of that shipping product line, so the practice is near-certain, though I did not retrieve a specific locomotive model number tied to this patent.
• Primary line of business: diversified industrial/manufacturing — at the time, a major electrical equipment, power generation, and rail propulsion manufacturer.
• Current status: Operating, but the relevant business unit is gone. GE Transportation (the Erie locomotive business that would have held this patent) was sold to Wabtec Corporation in 2019 and is now Wabtec; GE itself subsequently separated into GE Aerospace / GE Vernova / GE HealthCare (2023–2024). Which entity, if any, holds GE's expired 1982 patent archives today is not determinable from the sources I retrieved.

Critical scope note: the patent's term expired 2002-09-30 (Google Patents, "Anticipated expiration"). Any question of who "currently owns" it is therefore about a lapsed, unenforceable right, not a live asset. Assignment records for expired patents are frequently never updated by corporate successors — so an unbroken chain of title to GE on paper here would be normal and carries no NPE significance.

Assignment timeline

Chronological record of recorded transactions I could verify:

• 1982-09-30 (executed) / 1982-09-30 — Reel/Frame: NOT RETRIEVED. Google Patents records this event dated 1982-09-30 under "Legal Events," but the entry exposes no reel/frame; assignor names are given as "HORVAT, IVAN N., BROWN, HERBERT J., KUMAR, AJITH K. N."
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST" (per Google Patents legal-events text)
  • Assignor: Herbert J. Brown; Ivan N. Horvat; Ajith K. N. Kumar (individually)
  • Assignee: GENERAL ELECTRIC COMPANY; A CORP OF NY
  • Correspondent: not disclosed in any source I retrieved. I have no attorney or firm of record for this recording. I decline to guess at GE's in-house Erie patent counsel of the era.
  • Context: Original corporate assignment at filing — routine employer-takes-title from employee-inventors; no third party, no consideration to an outside entity. This is the same day as the application filing date, which is the classic pre-filing employment agreement assignment.

No other recorded assignment was found. The Google Patents legal-events feed for US4471421 shows only three substantive events across the patent's entire life: the 1982-09-30 reassignment to GE, the 1984-09-11 grant, and the 2002-09-30 expiration. That is the complete evidentiary picture I have.

If the Assignment Center in fact contains no further records beyond the original — which is what the available evidence indicates — the finding is: GE (or its successor interest) held the patent for its full term and no post-issuance transfer of any kind was ever recorded. For a patent that expired in 2002, twenty-four years ago, that is unremarkable.

Timeline diagram

timeline
    title Ownership of US 4471421
    1982 : Application filed 30 September
         : Inventors assign to General Electric
    1984 : Patent issued 11 September
    2002 : Patent term expires 30 September

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No assignment into any entity with an "IP / Patents / Licensing / Holdings / Ventures" suffix appears anywhere in the record. The only recorded assignee is General Electric Company, an operating industrial manufacturer. No single-purpose LLC, no registered-agent service address, no Delaware/Texas shell in the chain.

2. Known asserter in the chain — Not present. Neither the original assignee (GE) nor any purported successor matches any entity on the Acacia / Marathon / Intellectual Ventures / IPNav / Wi-LAN / Mosaid-Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock / Document Generation / Spangenberg lists you supplied. I also surfaced no Unified Patents or RPX high-frequency-plaintiff hit for this patent number.

3. Repeat correspondent across the chain — Unclear, bordering on not present. This is the one signal I could not evaluate at all, because no correspondent of record was retrievable for the sole recorded assignment. A single-link chain cannot show "recurrence" by definition — the signal requires at least two links sharing an attorney. With one recorded link and no correspondent data, there is nothing to compare. I am marking this unclear rather than not present precisely because the absence of data, not the absence of an attorney, is what drives the call.

4. Cascading transfers — Not present. There is no chain of transfers — zero post-issuance assignments, so no consecutive LLC-to-LLC hops, no shared correspondent addresses, and no sub-24-month cascade to analyze.

5. Pre-litigation transfer — Not present. No infringement suit naming this patent was identified in my prior litigation sweep (see the Litigation summary above), so there is no "first suit" date for a transfer to precede. The only assignment in the record is dated four weeks before the patent even existed as an application — the temporal inverse of a pre-suit standing-cleaning transfer.

6. Bankruptcy fire-sale — Not present. I found no Chapter 7 or Chapter 11 proceeding involving General Electric in the relevant period, and no evidence this patent was ever part of a patent portfolio sale (Kodak / Nortel / Polaroid-type). Note for accuracy: GE did undertake a corporate breakup (GE Transportation → Wabtec 2019; GE split into GE Aerospace / GE Vernova / GE HealthCare 2023–2024), but a corporate separation is not a bankruptcy fire-sale, and in any event the patent had been expired for 17 years by 2019 — there was nothing left to sell. Do not let the 2019 Wabtec deal or the 2023–24 GE splits be mistaken for a chain-of-title event on this patent.

7. Privateering — Not present. No transfer from GE to any assertor, and no evidence of any NPE asserting on GE's behalf against competitors. The patent's practical life (1984–2002) predates the modern privateering era, and GE's own locomotive competitors (EMD/Electro-Motive, ASEA, Brown Boveri) would more plausibly have met this technology in the marketplace than in court.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at the original operating assignee by expiration.

Verdict

Insufficient data — the chain consists of the original assignment only.

The sole recorded transaction is the 1982-09-30 assignment of inventors Brown, Horvat, and Kumar's interest to General Electric Company (a corp. of NY), contemporaneous with the application filing, and the patent then ran its full term to expiration on 2002-09-30 with no further recorded assignment, no identified litigation, and no public NPE-list match. Under your rubric, "only the original assignment" is the textbook Insufficient data case: there is nothing in the record that constitutes a shell-entity transfer, an asserter in the chain, a cascade, or a defensive-aggregation endpoint, and — importantly — the absence of a shell-entity chain here is a genuine finding, not a gap, because the patent expired nearly a quarter-century ago and no NPE had a live right worth acquiring. The single unresolved item is the correspondent of record on the 1982 filing, which I could not retrieve; that data point would not change the verdict, since a 1982 GE in-house or outside Erie counsel recording is a routine corporate filing, not an NPE tell.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (legacy: https://assignment.uspto.gov/patent/index.html), searching patent number 4471421; and Google Patents legal events — https://patents.google.com/patent/US4471421/en#legalEvents.

Recommended confirmation step if a definitive chain-of-title is required: a manual Assignment Center query on patent number 4471421 to capture the reel/frame and correspondent for the 1982 recording, plus a Patent Center (https://patentcenter.uspto.gov/) "Assignment" tab check on application 06/429,728 to confirm no later-issued assignment documents exist. I flag explicitly that my result is "none found beyond the original", not a certified "no other assignments exist."

I'll search for the front-page references cited against US 4,471,421.

Prior-Art Analysis for US 4,471,421 — What I Could and Could Not Verify

Sourcing note (read first). I was able to retrieve the patent's specification and background text (Google Patents, https://patents.google.com/patent/US4471421/en; and the DOE/OSTI mirror, https://www.osti.gov/biblio/[6158683](/patent/6158683)), but I was not able to retrieve the machine-readable front-page "References Cited" list for US 4,471,421 from Google Patents, Justia, USPTO PatentCenter, or patentimages in this session. My searches repeatedly returned the same specification text rather than the citation table. Accordingly, the references below are those expressly named in the patent's own Background of the Invention — which in GE patents of this era conventionally correspond to the front-page citation list, but I cannot certify that the specification mentions are the complete list of record citations, and I cannot rule out additional front-page references (other U.S. patents, foreign documents, or non-patent literature). Treat the list as verified-as-mentioned-in-the-patent, not verified-as-the-complete-Of-record-list.

(Also flagging an internal inconsistency in my instructions: the system framing gives today as 2026-09-30, while the task statement gives April 26, 2026. It does not affect the analysis, which is historical, but I am noting it rather than silently picking one.)

A. References expressly cited in the patent (Background section)

The specification contains three sentences that name prior art by number:

"The principle of using forced commutation to obtain unity power factor is further explained in prior art U.S. Pat. No. 3,392,319. Various forced commutation circuits are known, as illustrated, for example, by U.S. Pat. Nos. 3,849,718 and 4,181,932."

Verification status of the bibliographic details. I did not obtain the title, inventor, assignee, or exact issue/filing dates for any of these three in this session. The issue-date approximations above are inferences from the U.S. patent-number series (the ~3.39M range issued around mid-1968; the ~3.85M range around late 1974; the ~4.18M range around early 1980). I am labeling them approximate and unverified rather than asserting them as fact. If exact dates matter for a §102 paragraph analysis, pull the front pages individually.

Date arithmetic (if the approximate dates hold). All three would have issued more than one year before the Sept. 30, 1982 filing date of US 4,471,421, which makes them pre-AIA 35 U.S.C. § 102(b) art (statutory bars), and therefore also § 103 art. This is a date-arithmetic statement conditioned on the unverified issue dates.

B. Related applications cross-referenced in the patent

These are not "prior art" citations in the §102 sense (they are the applicant's own co-pending work, cited as disclosure support), but they belong in any citation inventory:

Per the earlier-generated section, Ser. No. 454,207 is probably US 4,467,408 and the β-logic disclosure also appears in US 4,420,713 — but that correspondence was flagged as probable, not confirmed, and I still cannot confirm it. Since no U.S. patent can be §102 prior art against another U.S. application with an earlier or equal filing date, these are continuity/derivation references, not anticipation references.

C. § 102 anticipation analysis, reference by reference

Governing principle. Anticipation under § 102 requires that a single reference disclose every element of the claim as arranged in the claim. US 4,471,421 is drafted as an improvement claim: the preamble/environmental portion recites a hybrid bridge (diode leg + alternately conducting controllable valve leg), an a-c source connection between the junctures, d-c terminals to a load, and control means that turn the main valves on at an adjustable ignition angle and operate valve turn-off means to turn them off at a desired extinction angle. The single point of novelty — the feature the patent itself characterizes as the invention — is the valve-turn-on-responsive means that temporarily blocks valve turn-off operation after each turn-on until the corresponding diode has fully recovered its reverse blocking capability (claims 1, 3, 4, 5, and the snubber-current variant of claim 6).

Bottom line on § 102. On the record available to me, none of the three cited references anticipates any claim of US 4,471,421, and in particular none anticipates independent claim 1, whose entire inventive contribution is the turn-on-responsive blanking of the forced-commutation turn-off until the diode recovers. The cited art supplies (i) the motivation for forced commutation (US 3,392,319) and (ii) exemplary commutation circuit topologies (US 3,849,718; US 4,181,932). The realistic challenge vector against this patent is § 103, not § 102 — and even there the cited references do not appear to teach recognizing the diode-commutation-interval overvoltage problem, let alone the specific cure. The patent's own framing (Background vs. Summary) is consistent with this: the cited art is recited for what is "known," and the invention is presented as a new practice.

Caveat on the anticipation call. Because I could not see the front page, I also could not see whether additional references of record exist that are not mentioned in the specification. If the front page lists, e.g., a foreign patent or an article on snubber design or on minimum turn-off-angle control for hybrid bridges, my § 102 conclusion could change for the narrower dependent claims (particularly claim 6's snubber-current sensing variant). I am stating a conclusion about the three references I can actually identify, not about the complete citation of record.

D. What I could not verify (explicit limitations)

1. The front-page "References Cited" list — not retrieved. This is the single most important gap for your request, and I am not going to fill it with plausible-looking patent numbers.
2. Titles, inventors, assignees, exact issue and filing dates of US 3,392,319, US 3,849,718, and US 4,181,932 — not retrieved; the dates above are number-series inferences.
3. Foreign/other cited documents (the patent has NZ 205704A and AU 558921B2 family members; the U.S. front page may also list foreign patent documents of record) — not retrieved.
4. Whether the patent's citation list includes non-patent literature — not retrieved.
5. Total claim count — per the earlier-generated section, claim text was retrieved only through claim 6 and was truncated; claim 1 is the only verified independent claim, and claims 7+ (if any) are unverified. My § 102 analysis therefore covers claims 1–6 as previously characterized.

Recommended next steps for a conclusive citation list: pull the front page image directly from USPTO PatentCenter (application 06/429,728) or patentimages.storage.googleapis.com for US4471421, and/or query Espacenet for the "Cited documents" tab, then cross-check the OCR-printed list against the three specification mentions above. If you obtain that list, I can extend this table reference-by-reference, including any foreign documents and their § 102/§ 103 treatment.

I'll research the prior art references identified in the specification and related art to ground the § 103 analysis.

Let me dig into the remaining prior art references and the claim set, plus analogous art on diode reverse-recovery protection and minimum-time gating.

Obviousness Analysis — U.S. Patent No. 4,471,421 under 35 U.S.C. § 103

Sourcing note. This analysis builds on the previously generated "Patent summary" and "Litigation summary" sections and does not repeat them. New material below is grounded in (a) the full patent text supplied in my task context, (b) the specification's own identification of prior art, and (c) live retrieval of the cited references. Where I could not verify something, I say so. Flag: the earlier section's claim inventory (claims 1–6, with claim 1 as the only verified independent claim) was truncated in retrieval; the analysis below is keyed to that inventory and is provisional as to any claims 7+ or any second independent claim I could not see.

1. Governing framework

The application was filed 1982‑09‑30 and granted 1984‑09‑11, so pre‑AIA § 103(a) governs. The Graham v. John Deere, 383 U.S. 1 (1966), factors apply, as elaborated by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), and MPEP §§ 2141–2144.

Two threshold points matter here:

1. The claim is a Jepson‑style improvement claim. The preamble ("In an electric power converter comprising … a rectifying bridge … at least one pair of diodes … alternately conducting controllable valves each having cyclically operative valve turn off means … control means operative … at an adjustable ignition angle … and at a desired extinction angle …") is written as the environment for "the improvement comprising.…" The specification's Background independently describes that same combination as known ("it has heretofore been proposed to provide means for periodically turning off the main thyristor leg … For this purpose, turn off means known as a forced commutation circuit can be used. Such means enables the conducting thyristor to be quenched or turned off at any desired 'extinction angle'"). That is a specification admission usable as prior art under MPEP § 2129 (and treated as admitted prior art in a Jepson preamble). Accordingly, only the turn‑on‑responsive blocking means of claim 1 is even arguably novel.

2. All three references the patent itself cites are § 102(b) art. US 3,392,319, US 3,849,718, and US 4,181,932 each issued well over one year before 1982‑09‑30, so they are prior art as a matter of statutory category — the only question is § 103 content.

2. The prior art references and their statutory status

Not prior art (important). Ser. Nos. 429,727 (Chandran & Matulevich) and 429,729 (Matulevich) were filed the same day (1982‑09‑30) and Ser. No. 454,207 (Brown, Kumar & Matulevich) was filed later (1982‑12‑29). Neither same‑day nor later‑filed co‑pending applications qualify under pre‑AIA § 102(a), (b), or (e), and none is a § 103 reference. (The earlier section flagged US 4,467,408 as the probable issue of Ser. No. 454,207 and noted the β‑logic also appears in US 4,420,713; my search is consistent with those correspondences but I did not verify them against official continuity records.) These are at most an obviousness‑type double‑patenting concern, not § 103 art.

3. Element-by-element: claim 1

Conclusion on claim 1. Every structural and control element is disclosed by US 3,392,319; the functional heart of the improvement is the subject of an express, articulated problem‑solution teaching in US 3,849,718 (same problem: excess voltage produced by quenching) implemented by the ordinary expedient of a gating interlock/time delay (US 4,058,738; US 4,224,661). Claim 1 is prima facie obvious over US 3,392,319 in view of US 3,849,718, further in view of US 4,058,738.

4. Claim‑by‑claim

Claim 2 (forced‑commutation subcircuit: auxiliary rectifiers + commutating capacitor across the main valves; periodic trigger signals). Fully disclosed by US 3,392,319 alone (auxiliary thyristors 18/19, capacitor 20, controllers 21 providing firing pulses to both sets), and redundantly by US 3,849,718 (quenching thyristors 5/6 + capacitors 7/8) and US 4,181,932 ("commutation capacitor, a commutation reactor and an auxiliary thyristor"). Obvious; arguably anticipated.

Claim 3 (block during the commutation interval and thereafter for a predetermined period of time). The "commutation interval" concept and the need to wait for the diode are supplied by US 3,849,718 (quenching only when excess‑voltage danger is absent) and, in express time‑delay form, by US 4,058,738 (inhibit for a period "longer than the duration of one commutation," built from multivibrators). Motivation: the PN‑junction reverse‑recovery/recombination phenomenon and t_rr are datasheet parameters; one of ordinary skill would size the delay to t_rr. The 300 µs value in the FIG. 5 embodiment is presented as an example, not as a critical range — no criticality data appears anywhere in the specification. Obvious.

Claim 4 (blocking also during recurrent intervals beginning at each zero crossing and ending at the next valve turn‑on). This is simply the natural‑commutation/freewheel portion of the cycle: before a main thyristor is fired, current freewheels in the diode leg and there is nothing to force‑commutate. It is inherent in line‑commutated bridge operation and in US 3,392,319's zero‑crossing‑synchronized phase control, and it is the mirror image of US 3,849,718's defined quenching window (quench only after current maximum and before the voltage zero crossing). The "old α" implementation (flip‑flop 205 set by the α trigger pulse, reset by the sync pulse) is conventional logic design. Obvious.

Claim 5 (bistable voltage detecting means across the diode juncture and valve juncture + time‑delay means triggered when the detector changes state at the end of the commutation interval). Voltage‑zero ("volt zero") detection synchronized to the bridge input voltage is ubiquitous in phase‑controlled converter control (US 3,392,319's synchronizing controllers; US 3,849,718's supply‑voltage‑referenced timing). The claimed structure is the detector 180 + delay 235 of FIGS. 4–5, i.e., a comparator plus a one‑shot — standard analog/digital building blocks (compare US 4,058,738's multivibrator delay chain). Obvious; the recitation of a particular circuit topology for a known function does not impart patentability absent a showing of unexpected results.

Claim 6 (snubber‑current‑responsive variant: RC snubbers across the diodes; block until snubber current decays below a predetermined low level). RC snubbers across the diodes are admitted prior art in this very patent, as is the fact that snubber current is the observable concomitant of diode reverse recovery. Detecting snubber‑resistor voltage with a level detector (elements 301, 313, 316, 317 of FIG. 6/5B) is the obvious instrumentation choice — the snubber resistor is the current‑sensing element already present in the circuit. Combining with the timing interlock of claims 1–3 yields predictable results. Obvious.

5. Why a person of ordinary skill would have combined these teachings

Under MPEP § 2144 and KSR, any one of the following articulated rationales suffices; several apply simultaneously.

1. Same field, same problem, same art. All of US 3,392,319, US 3,849,718, and US 4,181,932 concern forced‑commutated thyristor bridges operated with adjustable ignition/extinction angles for power‑factor improvement. US 3,849,718 addresses the identical problem the '421 patent addresses — the excess voltage generated "when the controlled rectifier is quenched … a result of the stored magnetic energy in the circuit which is interrupted." One of skill in the art would look precisely here.

2. Anticipation of the precise design need (KSR, "known problem"). US 3,849,718 expressly states the object is to "permit operation of a controlled rectifier circuit with the highest possible power factor without causing stresses on the components of the rectifier circuit as a result of undue voltage increases due to quenching," and it solves it by constraining when quenching may occur (β adjustable, "varied only within narrow limits"). Extending that constraint from a β‑angle limit to a time‑domain interlock (inhibit quenching until the diode has recovered) is a difference in implementation form, not in principle.

3. Known technique, same way, predictable result (MPEP 2144.04). Inhibiting or delaying a control pulse for a predetermined period following an initiating event is the most routine of control‑logic techniques — US 4,058,738 claims exactly that in a converter context ("for a predetermined period of time which is longer than the duration of one commutation," implemented with cascaded multivibrators), and US 4,224,661 controls the reverse‑voltage time on a rectifier being blocked. Applying that technique to the auxiliary‑thyristor gate drive of US 3,392,319/4,181,932 yields nothing more than the predictable result of letting the diode junction recombine.

4. Design incentive rooted in cost, not in discovery. The patent itself frames the invention as avoiding a cost penalty: no "bulky saturable reactors in series with the diodes," no "premium‑priced fast recovery diodes," no enlarged snubber resistors/capacitors, no higher diode voltage rating. A recognized cost/benefit tradeoff in a known circuit supplies a strong § 103 motivation, and it is an admission that the underlying solutions (snubbers, saturable reactors, fast‑recovery diodes) were already known.

5. Engineering not beyond the skill level. The enabling physics — that during diode‑to‑thyristor overlap the commutating‑capacitor voltage adds in series with the source voltage across the still‑recovering diode, raising dI/dt and the reverse‑recovery current — is straightforward circuit analysis of the very topology disclosed in US 3,392,319. Recognizing that quenching during overlap is harmful and interlocking the gate drive is an exercise of ordinary mechanical/electrical skill, not invention.

6. Anticipated rebuttals, and where they could succeed

• "No reference teaches blocking until full recovery." True as a matter of literal disclosure. But the reference need only be reasonably commensurate, and US 3,849,718's "quench only where excess‑voltage danger does not exist" plus US 4,058,738's "predetermined period longer than one commutation" together teach the function claimed. A patentee's strongest argument would be prosecution‑history evidence that the Examiner treated US 3,849,718 as non‑analogous; I have not retrieved the file wrapper, so I flag this as unverified.
• "The problem was unrecognized in the art." Not sustainable on this record: US 3,849,718 explicitly recognizes quenching‑induced overvoltage, and the '421 Background concedes that diode reverse‑recovery overvoltage and its snubber/reactor remedies were known.
• Secondary considerations. No unexpected‑results data appears in the specification; there is no asserted criticality of the 300 µs delay; and any commercial‑success argument would require a nexus between the claimed gating interlock and the locomotive converter's success — difficult to establish where the accused nexus is a cost‑avoidance feature. The '421 patent appears to have had no litigated presumption that could be tested (see prior Litigation summary).
• Claim‑drafting distinctions. Claims 3–6 recite implementation alternatives (time delay; zero‑crossing window; voltage detector; snubber‑current detector). These are alternative, functionally equivalent mechanisms for the same claimed result, and the Federal Circuit treats functionally equivalent alternatives as reinforcing, not defeating, obviousness (In re Kao; In re Beattie-type alternative‑element reasoning). Their coexistence in one patent is itself evidence of the predictability of the result.

7. Residual uncertainties affecting this opinion

1. Claims 7+ and any second independent claim are unverified (retrieval was truncated). If an independent claim exists that recites a specific numeric relationship (e.g., a critical delay bound tied to diode t_rr or to load current), the analysis of claims 3/5/6 would need revisiting.
2. File wrapper not examined. I did not review the original Examiner's rejections, the cited art of record, or any § 132 declaration; there may be an ex parte record that materially alters the claim construction of "fully recovered its reverse blocking capability."
3. US 4,058,738's full disclosure was retrieved only in the form of its claim language and abstract‑level content, and US 4,224,661 only at the abstract/description level; their precise teaching of gate‑drive interlocks (as opposed to start‑up sequencing) should be confirmed against the full PDFs before final reliance.
4. Family correspondences (Ser. Nos. 429,727 → US 4,420,713; 429,729 → ?; 454,207 → US 4,467,408) remain probable, not confirmed; if any of those has an earlier effective filing date than 1982‑09‑30, it would change the prior‑art status of the β‑logic disclosure. On the dates I retrieved, none does.

8. Bottom line

On the record available, all six claims of US 4,471,421 are prima facie obvious under pre‑AIA § 103(a):

• Primary combination: US 3,392,319 (hybrid bridge with diode leg, forced‑commutated thyristor leg, auxiliary thyristors + commutating capacitor, adjustable ignition and extinction angles for power‑factor control) in view of US 3,849,718 (same problem of quenching‑induced overvoltage; quenching permitted only where excess‑voltage danger is absent; β held within narrow limits).
• Secondary combination: add US 4,058,738 (gating‑restraint delay "longer than the duration of one commutation," multivibrator‑implemented) to supply the "predetermined period of time" of claim 3; optionally US 4,181,932 for the forced‑commutation subcircuit of claim 2 and US 4,224,661 for controlled reverse‑voltage timing.
• Admitted prior art (RC snubbers across the diodes; saturable reactors; fast‑recovery diodes) plus routine level detection supplies claim 6.
• The motivation is strong and articulated in the references themselves (avoid destructive overvoltage while quenching; avoid cost of reactors, premium diodes, and oversized snubbers), and the result — delaying an auxiliary‑rectifier gate pulse until the main diode has recovered — is the predictable consequence of applying a conventional gating interlock to a known circuit.