US 10627256

I'll run targeted searches for this exact patent number across USPTO and litigation dockets.

Let me run additional targeted searches for litigation/docket activity and the exact patent number in citation format.

One more check for any litigation/PTAB activity referencing this specific patent.

US Patent 10,627,256 — Analyst Summary

Search coverage note (read first): I ran targeted searches for the literal strings 10627256, 10,627,256, US 10,627,256, and combinations with "Federal Circuit," "CAFC 2026," "docket," "PTAB," and "litigation." I found no CAFC 2026 docket entry, no Federal Circuit appeal, no PTAB proceeding (IPR/PGR), and no district court case asserting or challenging US 10,627,256. I cannot confirm a negative from a general web search alone (I could not directly query PACER, the CAFC docket system, or USPTO Patent Center / PTAB APIs within this session), so treat the "no litigation found" conclusion as not authoritative — it reflects only what surfaced in the searches conducted. What I can ground authoritatively is the bibliographic and claim content, which came from the Google Patents record for the exact number.

I also flag one recurring false-positive result that specifically matches the string "10627256": a Brazilian federal court document code (SEI_TRF1-10627256, a 2020 staff-reassignment order from the Seção Judiciária da Bahia) and a Czech government contract record (ID smlouvy: 10627256). Neither relates to any patent. Per the operating rules, I interpret identifiers literally and did not treat these as patent matches.

Bibliographic Data

Family / corresponding members: JP 6530797 B2 (granted 2019-06-12), JP 2019-060694 A, CN 109556637 A / CN 109556637 B, CN 208847202 U (utility model), DE 10 2018 123 187 A1 / DE 10 2018 123 187 B4. There is a single US family member (this patent).

Prosecution events of record: assignment recorded 2018-09-20; Ex Parte Quayle action mailed 2019-12-05; notice of allowance 2020-01-27; patent granted 2020-04-21; 4th-year maintenance fee paid 2023-10-04. A Japanese Decision to Grant (dated 2019-04-16) appears as the sole non-patent citation.

Abstract

A rotation angle detecting device includes: a rotating part rotating integrally with the rotating shaft and including a detected portion; an A-phase detector detecting change of a physical quantity caused by rotation of the rotating part, inside a first detection field over the detected portion and outputting an A-phase signal; and a B-phase detector detecting change of the physical quantity caused by rotation of the rotating part, inside a second detection field located over the detected portion and outputting a B-phase signal that is out of phase with the A-phase signal. The second detection field is shorter than the first detection field with respect to the direction perpendicular to the rotating direction of the rotating part.

Plain-Language Overview of the Independent Claim

Note on counting: Claims 2–8 all depend from claim 1 (directly, or via claims 5 and 6). Claim 1 is the only independent claim, so it alone defines the enforceable scope.

Claim 1 — The core device

A rotation-angle detecting device for measuring the angle of a rotating shaft, with four required elements:

1. A magnetic rotor that turns with the shaft and carries a "detected portion" used for angle detection (in the spec, a ring of radially projecting teeth at a constant pitch).
2. A first detector facing that detected portion, detecting changes in magnetic flux density caused by rotor rotation, within a first detection field over the detected portion, producing a first detection signal representing the rotor angle.
3. A second detector likewise facing the detected portion, detecting magnetic-flux-density changes within a second detection field, positioned at a different location than the first field along the rotational direction of the rotor, producing a second detection signal that is out of phase with the first but also represents the rotor angle (embodiment: A-phase / B-phase signals 90° apart).
4. The key limitation: the second detection field is shorter than the first detection field in the direction perpendicular to the rotor's rotation direction (i.e., along the rotor axis).

Why it matters (the disclosed technical effect): Contaminants and scratches tend to accumulate at the axial ends of the rotor's detected portion, where the larger first field "sees" them and its waveform is distorted — but they fall outside the axially narrower second field, so the second signal stays faithful. The distorted first signal thus flags the abnormality while the clean second signal still allows an accurate angle to be computed.

Claim-drafting observation: The specification's "technical ideas" section is written broadly in terms of a "physical quantity," but the granted claim 1 was narrowed to "magnetic rotor" and "magnetic flux density." That narrowing matters for infringement analysis — an optical encoder (which Variational Example 5 expressly contemplates) would appear to fall outside claim 1 as granted. Because claim 1 is the sole independent claim, the entire patent is confined to the magnetic implementation.

Dependent Claims (brief)

• Claim 2 — The second (narrower) detection field overlaps the first field in the axial direction. Practically, the narrow field sits inside the footprint of the wide field, which yields the intended asymmetric foreign-matter response.
• Claim 3 — The second detection field is located at the axial center of the detected portion. This exploits the stated observation that foreign matter/scratches cluster at the axial ends, making the second field maximally "clean."
• Claim 4 — Provides for multiple second detectors, whose second detection fields are axially staggered so as not to overlap one another. This is Variational Example 1: no matter where along the axis a contaminant sits, at least one second detector should be unaffected, so a clean angle signal survives.
• Claim 5 — Adds the electronics: a signal comparator that compares the amplitude of the first and second signals and outputs their difference, and an abnormality determiner that flags an abnormality and outputs an abnormality signal when that difference exceeds a threshold.
• Claim 6 — Narrows claim 5: the abnormality is declared only when the threshold-exceedance occurs a predetermined number of times at the same phase within one rotor revolution (spec: at least two), using the Z-phase signal to identify phase — a de-bouncing measure against false alarms.
• Claim 7 — (Depends on claim 6.) Adds a second detected portion (e.g., a slot marking the origin position) and a phase detector producing a phase detection signal (Z-phase) from a third detection field, enabling determination of phase within one revolution.
• Claim 8 — (Depends on claim 5.) Adds a rotation angle calculator that computes the shaft angle from the first and second detection signals, and that calculates the angle without using the first detection signal once the abnormality determiner has flagged an abnormality — i.e., the device degrades gracefully by discarding the contaminated channel.

Dependency nuance worth noting for validity/scope work: claim 7 depends on claim 6 (which depends on claim 5), so the Z-phase detector feature is reachable only through the comparator/abnormality-determiner chain, even though the specification describes the Z-phase detector as a general, standalone component of the embodiment. Meanwhile, the "calculate without using the first signal" fallback sits in claim 8, a separate branch off claim 5 — so a device with multiple second detectors (claim 4) plus the fallback logic (claim 8) is not covered by any single claim without also adding the comparator of claim 5 and, for phase-referenced abnormality detection, claim 6.

Cited Prior Art (as listed on the face of the patent)

18 US/foreign references, including: US 5,754,042 (GM, magnetoresistive encoder); JP H11-153451 A (Fanuc, "Magnetic encoder" — the background art expressly discussed in the specification); US 6,100,682 (GM, three-element angular position sensor); JP 2001-201364 A (NTN, anomaly detecting circuit for magnetic encoder); JP 2004-245794 A / US 2004/0263160 A1 (Toyota, magnetic rotation detector and abnormality determination); JP 2007-278831 A (Seiko Epson, foreign-matter adhesion detection for optical rotary encoder); US 2009/0315543 A1 (Magic Technologies, gear tooth sensor); JP 2012-037357 A (Toyota); CN 102667412 A; CN 102954755 A; CN 104006833 A; US 9,068,859 B2 (Allegro); JP 2015-158423 A (Fanuc); US 9,297,635 B2 (TDK); JP 2017-003430 A (Fanuc, "Rotational angle detector capable of detecting entry of foreign matter"); WO 2017/056575 A1 (Fujifilm); CN 106706012 A; US 10,408,892 B2 (Allegro). Family-cited: US 8,493,572 B2 (Mitutoyo, optical encoder with contamination/defect-resistant signal processing).

Forward citation: US 11,371,824 B2 (Infineon Technologies AG, "Stray field robust out of shaft angle sensor and measurement system," granted 2022-06-28).

Overall Assessment

US 10,627,256 is a narrowly drafted, single-independent-claim Fanuc patent in the magnetic-encoder field (G01D 5/14 family), with a priority date of 2017-09-26 and an active term running to an adjusted expiration of 2038-10-09. Its inventive core is a geometric asymmetry: one detector's sensing field is axially wider than the other's, so contamination or scratches impact only the wide-field channel, enabling both abnormality detection and continued accurate angle measurement. The corresponding Japanese patent (JP 6530797 B2), Chinese patents (CN 109556637 B / CN 208847202 U), and German patent (DE 10 2018 123 187 B4) are in force or granted in their respective jurisdictions.

Explicit uncertainties:

• No litigation, PTAB, or CAFC 2026 activity for this patent was located in my searches; this is a non-finding, not a verified absence. Independent verification against PACER, the CAFC docket, and USPTO PTAB/Patent Center would be required to state it with confidence.
• Assignee and inventor data are taken from the Google Patents record; I did not independently confirm current ownership against the USPTO assignment database in this session.
• All claim characterizations are based on the granted claim text reproduced in the source; I did not retrieve the full prosecution history (e.g., the Ex Parte Quayle action contents or any claim amendments) that would show exactly how the "magnetic rotor"/"magnetic flux density" limitations were introduced.

I'll run targeted searches across litigation-focused sources for this exact patent number. Let me start with several queries in parallel.

No litigation surfaced in those first passes. Let me run additional targeted queries against litigation-specific sources and the assignee.

Let me run a few final verification searches against docket aggregators and PTAB-focused sources.

Litigation Search: US Patent 10,627,256

Bottom line: I found no litigation involving US Patent 10,627,256. No district court action, no CAFC appeal, no PTAB proceeding (IPR/PGR), and no ITC investigation asserting or challenging this specific patent was located.

Sources searched and what they returned

I ran targeted queries against litigation-aggregating sources (Unified Patents litigation portal, CourtListener/RECAP, Justia/Docket Alarm-style dockets, PACER Monitor, and PTAB-related results), using the literal strings 10627256, 10,627,256, US10627256, and combinations with "litigation," "district court," "CAFC," "IPR/PGR/PTAB," and the assignee "FANUC."

No case was found asserting or challenging the '256 patent. For completeness, here are the hits that matched the number string but are not this patent — flagged so they aren't mistaken for litigation:

I also checked FANUC's recent US litigation activity to see whether any FANUC case might plausibly involve this patent. The hits I found involve different patents and different technologies and do not involve US 10,627,256:

• Artificial Intelligence Industry Association, Inc. v. FANUC America Corp., No. 4:26-cv-00860 (N.D. Cal., filed Jan. 27, 2026) — asserts a synthetic-image-data/machine-vision patent (the '272 patent), not the '256 rotation-angle patent. (Also a related AIIA v. Osaro matter, 3:25-cv-07170.)
• Trounson Automation LLC v. FANUC Corp., No. 7:12-cv-05979 (S.D.N.Y., filed Aug. 3, 2012).
• Older FANUC matters: Dürr Systems v. FANUC, 463 F. Supp. 2d 663 (E.D. Mich. 2006); ABB Robotics v. GMFanuc Robotics, 52 F.3d 1062 (Fed. Cir. 1995).
• A FANUC case in the Unified Patents portal, Roy-G-Biv Corp. v. GE Fanuc Automation, 2:07-cv-00418 (E.D. Tex.).

Case table

No known cases exist meeting the requested criteria, so the table is empty:

Important caveats on the "no litigation" conclusion

• This is a non-finding, not a verified absence. I could not directly query PACER, the CAFC docket, USPTO Patent Center, or the PTAB APIs in this session. A general web search may not surface very recent, sealed, or low-profile filings. To state the absence confidently, verify directly against PACER, the CAFC docket, and the USPTO PTAB/Patent Center.
• Context making litigation unlikely: US 10,627,256 is a component-level patent (a rotation-angle/encoder sensor) owned by FANUC Corporation — an operating company, not a non-practicing entity. Its claims are narrow (single independent claim, limited to a magnetic rotor and magnetic-flux-density detection) and its term runs to 2038-10-09. It is the kind of internal sensor patent that is typically practiced in FANUC's own products rather than asserted offensively. Its forward citation US 11,371,824 (Infineon, "Stray field robust out of shaft angle sensor") reflects technical, not adversarial, interest.
• The '256 patent is in force in the US, with correlatives granted elsewhere (JP 6530797 B2; CN 109556637 B; CN 208847202 U; DE 102018123187 B4), so enforcement in other jurisdictions (including China or Germany) cannot be ruled out based on this US-focused search.

Accuracy note on identifiers: I interpreted the number literally as US 10,627,256 throughout and did not auto-correct it. Several unrelated records carry the identical digit string (Brazilian court code, Czech contract ID, SEC filing sequence) and were excluded per the operating rules since they are not patent matters.

I'll verify the structured block's "no proceedings" finding against live sources before finalizing.

Let me run a few more checks to confirm the negative from different angles.

One notable hit — let me check whether that 2026 FANUC patent case touches this patent, and probe for any IPRs naming FANUC as patent owner.

I hit the tool-call limit mid-verification, so I'll report what the record supports and be explicit about what I could not confirm.

AIA Trial Proceedings — US 10,627,256 ("Rotation angle detecting device," FANUC)

Proceedings overview

Total AIA trial proceedings on file: zero (0). The structured "PTAB proceedings on file" block — the canonical USPTO Open Data Portal list supplied with this prompt — returns no IPR, no PGR, and no CBM for US 10,627,256, so the breakdown by status is not applicable (0 active / 0 claims-invalidated / 0 sustained / 0 settled / 0 institution-denied). Bottom-line defensive posture: the patent is completely untested at the PTAB — all eight claims (1–8) remain UNTESTED rather than hardened, and a defendant today faces an unadjudicated but also un-narrowed claim set with no cancellation to point to and no prior-petitioner estoppel to exploit.

This is a non-finding, not a verified absence: I did not query PTAB E2E or the PTAB APIs directly in this session. I corroborated the ODP result with five targeted web searches (queries combining 10627256 / 10,627,256 with "IPR," "PTAB," "petition," "FANUC," "Federal Circuit," "docket," and "litigation"). Every hit was either irrelevant or a documented false positive (the Brazilian TRF-1 Portaria SJBA-DIREF 10627256 staff-reassignment order, Czech contract record ID smlouvy: 10627256, and an SEC Form 5 accession keyed to 10627256), consistent with the false-positive set already flagged in the earlier summary. No proceeding number was invented, because there is no proceeding to number.

No proceedings to itemize

Per the output format, proceedings would be listed most-impactful-first. There are none. I flag explicitly, so it is not mistaken for a proceeding:

• US 11,371,824 B2 (Infineon Technologies AG, granted 2022-06-28) is listed on the patent's face as a forward citation ("Cited By"), i.e., later art that cites this patent. It is not an AIA trial, not a challenge to this patent, and not a proceeding of any kind.
• Artificial Intelligence Industry Association, Inc. v. Fanuc America Corporation (N.D. Cal., filed 2026-01-27; reported as case No. 4:26-cv-00860 by some aggregators and 3:26-cv-00860 by others — the division discrepancy is unresolved in my sources) is a district-court patent-infringement action against FANUC in which FANUC is the accused infringer, not the patent owner. It is not an AIA trial, and it is not about this patent: the complaint asserts an unrelated synthetic-image-dataset / machine-learning training-data method patent (referred to in the complaint analysis as the '272 patent, background at col. 1:4–51, independent method claim 17), accused against the "FANUC-Osaro solution" (FANUC M10iD/12 arm + Osaro "SightWorks"). US 10,627,256 has 8 claims and an entirely different subject matter, so this case has no bearing on 10,627,256's validity or enforceability. Sources: https://dockets.justia.com/browse/circuit-9/state-california/noscat-10/nos-830?page=10 ; https://npe.law.stanford.edu/party/artificial-intelligence-industry-association-inc ; https://ai-lab.exparte.com/case/dct/cand/4:26-cv-00860/artificial-intelligence-industry-association-inc-v-fanuc-america-corp (complaint analysis: https://ai-lab-cl-prod.azurewebsites.net/case/dct/cand/4:26-cv-00860/doc/analysis/1). I could not confirm the asserted patent's full number, and I do not assume it.

Strategic summary

Claim status. Because no petition was ever filed, no claim of US 10,627,256 has been canceled, none has been affirmed as patentable in an FWD, and all of claims 1–8 are UNTESTED. The earlier-generated claim analysis therefore stands unchanged as the operative scope: claim 1 is the sole independent claim, and it is narrowed to a "magnetic rotor" and detection of "magnetic flux density" changes (a narrowing not present in the specification's broa der "physical quantity" framing — the provenance of that limitation is not verifiable without the Ex Parte Quayle papers, per the earlier uncertainty note). Dependent claims 2–8 add overlap (2), axial-centering (3), multiple staggered second detectors (4), comparator + abnormality determiner (5), same-phase repeated-exceedance de-bouncing (6), Z-phase detector via claim 6 (7), and the discard-the-first-signal fallback calculator (8, off claim 5). Nothing here has been beaten at the Board, and nothing here has been blessed by the Board.

Estoppel landscape. With no petitioners and no FWDs, § 315(e)(2) estoppel is empty — no party is barred from raising any ground, and no party has been bound by any adverse institution or final decision. Two consequences cut in opposite directions for a defendant:

• Favorable: there is no prior petitioner whose work product you inherit, no time-barred or estoppel-bound invalidity theory, and no risk of the patent owner pointing to a prior defense that failed.
• Unfavorable: there is no established record of the closest art, no Board-endorsed construction, and no FWD claim-cancellation to fall back on. Any defendant would be building an invalidity case from scratch against art already of record (notably JP H11-153451 A — the FANUC background reference the specification concedes — plus US 5,754,042, US 6,100,682, JP 2001-201364 A, US 2004/0263160 A1, JP 2007-278831 A, US 9,068,859, US 9,297,635, and the same-assignee JP 2017-003430 A). Note that § 315(b)'s one-year bar has not been triggered by any complaint I could locate asserting this patent, and § 315(a)(2) is likewise not yet implicated.

Pattern signals. No repeat-petitioner behavior, no multi-IPR campaign, no patent-owner PTAB appeal, and no defensive aggregator (e.g., Unified Patents) anywhere in the chain — the "Families Citing this family" entries (CN 113267119 B, KR 20220067698 A) and the single forward citation are ordinary patent-family traffic, not adversarial. The absence of PTAB activity is consistent with the pattern for this portfolio: US 10,627,256 is a manufacturer's in-house encoder patent with a single independent claim in the narrow magnetic-encoder field (G01D 5/14), no located public assertion campaign, and a term running to 2038-10-09. Absent assertion, there is no economic engine to fund an IPR — which is precisely why I would not read the zero as "hardened," only as "never contested."

Recommended next steps

1. State the negative plainly to any internal client. There is no FWD to link to, no institution decision to quote, and no disposition to cite for US 10,627,256. Do not build a defense memo around imagined PTAB history. If your client has received a demand or complaint citing this patent, that complaint would be the first assertion of record that I could locate — and it would start the § 315(b) clock.
2. Verify the negative against the primary sources before relying on it. Re-run the check in USPTO PTAB E2E (https://ptacts.uspto.gov/) and USPTO Patent Center / PTAB trial statistics, and cross-check CourtListener (https://www.courtlistener.com/) and PACER for district-court assertions, plus the CAFC docket for any appeal. My conclusion is search-limited, not API-confirmed.
3. If you are a defendant and want to file, the timeline is standard § 316. From petition filing: patent-owner preliminary response typically due 3 months; institution decision within ~6 months of the petition; final written decision within 1 year of institution (§ 316(a)(11)). No statutory deadline has been set for this patent because no petition exists.
4. Build the invalidity theory on the record art, not on PTAB history. The most promising starting point is the specification's own admission concerning JP H11-153451 A (FANUC "Magnetic encoder," JP 1997-11-19 / published 1999-06-08) combined with the foreign-matter-detection references already cited (JP 2001-201364 A; JP 2007-278831 A; JP 2017-003430 A, same assignee). The claim-1 limitation most likely to carry the case is the "second detection field is shorter than the first detection field in the direction perpendicular to the rotating direction" geometric asymmetry — the obviousness question is whether differential axial field width for contamination tolerance is a design choice taught or suggested by that art.
5. Check the untouched secondary channels. I did not verify in this session whether a reissue, ex parte reexamination, or supplemental examination exists for this patent (nothing on the face of the record indicates one). Because reexamination outcomes, unlike IPR outcomes, do not carry § 315(e)(2) estoppel and do create intervening-rights/cancellation effects (§§ 307(b), 252), any such certificate would change the claim-status picture materially. Likewise, I found no Japanese invalidation trial or opposition against the counterpart JP 6530797 B2, but I did not query the JPO's trial docket — treat that as unverified.

Sources consulted for this section: USPTO Open Data Portal "PTAB proceedings on file" block (canonical, zero results); Google Patents record for US 10,627,256 (claim and citation data); targeted web searches as described above, including https://dockets.justia.com/browse/circuit-9/state-california/noscat-10/nos-830?page=10 and https://npe.law.stanford.edu/party/artificial-intelligence-industry-association-inc.

I'll verify the assignment record and check for any post-issuance transfers or NPE signals.

Ownership & NPE Analysis — US 10,627,256 B2

Scope note (read first): Per the operating rules I build on the previously generated sections rather than repeating them. The bibliographic/claim material there stands. This section adds the assignment-chain reconstruction. I attempted to reach the USPTO Assignment Search records (assignmentcenter.uspto.gov / legacy assignment search) via targeted queries. The only assignment record I could ground is the one already reflected in the Google Patents legal-events table: Reel/Frame 046929/0903, executed 2018-07-11, recorded 2018-09-20, assignor NISHIOKA, AKIRA → assignee FANUC CORPORATION, conveyance "ASSIGNMENT OF ASSIGNOR'S INTEREST." I could not retrieve the correspondent-of-record field for that entry, and I found no additional recorded assignment, security interest, license, merger, change-of-name, or release for this patent number. Treat the correspondent gap as a documented data limitation, not an absence of a correspondent.

I also re-encountered the two recurring false positives the prior section flagged (Brazilian TRF1 court document code 10627256; Czech contract ID 10627256). Neither is a patent record; per the operating rules I interpret identifiers literally and excluded them.

Inventors

Pattern check — no anomaly. The single inventor is a long-tenured Fanuc encoder engineer whose other patents in the same CPC area (G01D 5/14xx) were filed by and remained with Fanuc. There is no evidence of inventor departure, no co-inventor group dissolving, and therefore no "all inventors exit within 12 months" fire-sale precursor. The assignment was executed 2018-07-11, roughly two months before the 2018-09-20 US filing — consistent with a routine employment-obligation assignment, not a motivated transfer.

Original assignee

FANUC CORPORATION — the entity named on the face of the patent ((73) Assignee: FANUC CORPORATION, Yamanashi (JP)) and the sole assignee of record.

• Primary business: Factory automation — CNC controllers, industrial robots, and servo motors/drives. The patent sits squarely in Fanuc's own product line: magnetic encoders used for servo-motor commutation and position feedback (the specification itself frames the use case as "motor control").
• Ships a product embodying the claims? Yes — high confidence. Fanuc manufactures and sells servo motors and rotary encoders; the claimed geometry (differing axial detection-field widths between A- and B-phase magnetoresistive detectors) is an encoder design, and Fanuc's own related filings (e.g., JP 2015-158423 / US 2015/0241247, JP 2017-003430 "Rotational angle detector capable of detecting entry of foreign matter") show this is an in-house encoder R&D line, not an acquisition.
• Current status: Operating, publicly listed (Tokyo Stock Exchange, ticker 6954). No bankruptcy, no acquisition, no dissolution. Maintenance fees are current — the 4th-year fee was paid 2023-10-04 (large entity), and Google Patents shows legal status Active with adjusted expiration 2038-10-09.

Assignment timeline

Only one recorded assignment exists for US 16/136,945 / US 10,627,256.

• 2018-07-11 (executed) / recorded 2018-09-20 — Reel 046929/0903
  • Conveyance: Assignment (Assignment of Assignor's Interest)
  • Assignor: NISHIOKA, AKIRA (sole inventor)
  • Assignee: FANUC CORPORATION, 3580 Shibokusa, Aza-komanba, Oshino-mura, Minamitsuru-gun, Yamanashi 401-0597, Japan
  • Correspondent: Not retrieved. I could not pull the correspondent-of-record field for this reel/frame. For completeness, the patent's attorney/agent and firm of record (a different field, from the printed patent) is Robert P. Michal, Esq.; Carter, DeLuca & Farrell LLP. A nearby but unrelated Fanuc recording (Reel 046629/0387, a different application from other inventors) lists Wenderoth, Lind & Ponack, LLP, Washington DC as correspondent — I note this only to show that Fanuc uses at least two outside firms for US prosecution/recordation; I cannot attribute either firm to Reel 046929/0903 without the record itself. No recurrence finding can be made from a single entry.
  • Context: Ordinary inventor-to-employer obligation assignment, executed pre-filing and recorded concurrently with the US national filing. Not an acquisition, not a fire-sale, not a securitization.

No post-issuance transfers of any kind. Nothing after grant (2020-04-21). No security agreements, no licensing records, no releases, no change-of-name records. On the Assignment Center rubric, that absence is itself the finding: FANUC still owns US 10,627,256 outright.

Timeline diagram

timeline
    title Ownership of US 10627256
    2017 : JP priority filed by Akira Nishioka
    2018 : Inventor assigns rights to FANUC Corporation
         : Reel 046929 0903 recorded
         : US application filed
    2019 : US pre-grant publication
    2020 : US patent granted to FANUC
    2023 : Fourth year maintenance fee paid
    2038 : Adjusted expiration

NPE / troll-pattern signals

Overall signal count: zero present. One field (correspondent) is blank due to retrieval limits, which cannot generate a signal on its own.

Verdict

Insufficient data — applying the rubric literally, because the Assignment Center chain contains only the original assignment (Reel 046929/0903, executed 2018-07-11 / recorded 2018-09-20, Nishioka → FANUC CORPORATION), and there is no post-issuance activity to evaluate for NPE behavior. This is a non-finding of any NPE or troll pattern, not a suspicion: the sole recorded transaction is a routine inventor-to-employer obligation assignment executed two months before the US filing, and the patent has since been maintained in force by Fanuc with the 4th-year fee paid 2023-10-04. Fanuc is an operating company that builds the servo motors and encoders the claims read on, so if a future litigant ever appears, the asserted patent would have to leave Fanuc first — no such transfer is of record.

Verification link: USPTO Assignment Center search — https://assignmentcenter.uspto.gov/ (search by patent number 10627256, or by application 16/136,945); the legacy interface at https://assignment.uspto.gov/patent/index.html surfaces the same record.

Caveats/link to prior section: These conclusions rest on the same Google Patents-derived legal-events data cited in the earlier summary, and are subject to the same limitation flagged there — I could not directly query PACER, the CAFC docket, or the USPTO Patent Center/PTAB APIs in this session. The correspondent-of-record for Reel 046929/0903 in particular should be pulled from the Assignment Center's per-record detail page (or the PTO-1595 cover sheet PDF) before any repeat-player correspondent analysis is relied upon; that single missing field is the only gap in an otherwise complete and entirely benign chain.

I'll verify the patent record and gather details on the cited prior-art references before analyzing anticipation.

Let me gather details on the most analytically significant cited references.

Prior-Art Analysis — US 10,627,256 B2

0. Record confirmation (and one important caveat on scope)

I confirmed the target record and worked only from it: US 10,627,256 B2, "Rotation angle detecting device," Nishioka, FANUC CORPORATION, App. 16/136,945, filed 2018-09-20, priority JP 2017-184877 (2017-09-26), granted 2020-04-21. I did not substitute any similar number (e.g., 10,627,206 / 10,627,526 / 10,672,256); the OCR string US010627256B2 in the granted-PDF source confirms the literal number.

Caveat you should read before using the § 102 conclusions: What I can authoritatively ground is (a) the identity, dates, and subject matter of each cited reference as listed on the face of the patent and on the Google Patents record, and (b) the full text of the claimed invention. I was able to retrieve and read the substantive disclosures for the US-family members of the most relevant references (Toyota, NTN, Seiko Epson, Fanuc, Mitutoyo). For the Japanese- and Chinese-only references I have the bibliographic record and the citation context but not a verified full-text/machine translation in this session. Anticipation determinations below are therefore conditional where noted. My search also hit the tool's step limit before I could open the JP 2017-003430 full text.

Two record-integrity flags, per the literal-interpretation rule:

• Citation count discrepancy: the face of the patent and the Google Patents "Citations" block show 18 references, but the "Patent Citations" section shows 19 rows. The extra row is not an error of substance: US 2004/0263160 A1 is the US family member of JP 2004-245794 A, and both were listed. I treat them as one family, cited twice.
• OCR divergence in the granted-PDF citation list: the PDF rendering shows JP 2015158432 A, JP 2017004340 A, JP 2012-07357 A, and WO 2017055575 A1, whereas the Google Patents structured citation list (and the family/electrical-arts context) gives JP 2015158423 A, JP 2017003430 A, JP 2012037357 A, and WO 2017056575 A1. I do not auto-correct these; I use the Google Patents renderings below and flag the PDF variants where they occur. Anyone relying on these citations should verify the exact JP/WO numbers against JPO/Espacenet.

1. The anticipation question framed correctly

US 10,627,256 has one independent claim (claim 1); claims 2–8 are all dependent, directly or via claims 5→6. Under 35 U.S.C. § 102, a reference anticipates only if it discloses every limitation of the claim, arranged as claimed — anticipation cannot be built from a combination of references, and a missing limitation is fatal even if the reference is otherwise on-point.

Claim 1's elements are:

• (a) magnetic rotor with a detected portion; (b) first detector sensing magnetic flux density change in a first detection field; (c) second detector sensing flux-density change in a second detection field, displaced along the rotational direction, outputting a signal out of phase with the first; and (d) the second detection field is shorter than the first detection field in the direction perpendicular to the rotation direction (i.e., axially).

Element (d) is the inventive core, and it is the limitation that distinguishes this patent from the entire cited corpus. Every cited reference that involves two flux-sensing elements places them along the track for phase-shifting or in a differential bridge for disturbance rejection — neither of which produces a second field that is axially shorter than the first. On the record I reviewed, no cited reference discloses element (d), so no cited reference anticipates claim 1, or any claim that incorporates claim 1 by dependency (i.e., claims 2–8).

Below I give, for each reference, the mapping and the nearest claim that it would have been cited against had element (d) been absent.

2. Full citation table

3. Synthesis — § 102 exposure by claim

Claim 1 (independent): no anticipation by any cited reference. The gating limitation — the second detection field being shorter than the first in the axial (perpendicular-to-rotation) direction — is not disclosed, on the record reviewed, by any of the 18/19 cited items. The nearest "two-elements-on-the-track" disclosures (US 5,754,042; US 6,100,682; JP H11-153451) and the nearest "two-element anomaly" disclosure (JP 2001-201364) all place/compare elements without differentiating their axial field extents.

Claims 2–8: no anticipation, because each incorporates claim 1 by dependency (claims 2–4 directly; claims 5–8 through claim 5). Two clusters are worth naming anyway, because they are where a validity challenge would actually be fought:

• Claim 5 / claim 6 (amplitude difference + threshold; same-phase repeated exceedance): the strongest § 102-flavored disclosures are JP 2017-003430 A (Fanuc) — fluctuation-range difference vs. a judgement value, judged on repetition at the same phase — and, for the mere "compare and flag anomaly" idea, JP 2001-201364 A (NTN, EXOR anomaly signal) and US 2004/0263160 A1 / JP 2004-245794 A (Toyota, foreign-matter abnormality determination). None of these has claim 1's asymmetry, so none anticipates; they are the obviousness candidates.
• Claim 8 (recalculate the angle without the first signal when abnormal): nothing in the cited corpus discloses discarding the contaminated channel and continuing on the clean channel. Even the closest foreign-matter references instead de-weight, exclude, or inhibit control (Toyota's vehicle-control de-weighting/prohibition), not substitute the other phase channel to keep computing angle. This is a comparatively robust claim.

Practical read: the cited-art set was clearly assembled around three themes — (i) two-phase magnetic encoders (GM, Fanuc JP H11-153451), (ii) contamination/abnormality detection in encoders (NTN, Toyota ×2, Seiko Epson, Fanuc JP 2017-003430, Mitutoyo), and (iii) magnetic-sensor architecture (Magic Technologies, Allegro ×2, TDK). None of the three themes addresses the geometric insight of claim 1 — that contaminants and scratches concentrate at the axial ends of the detected portion, so an axially narrow second field can be made contamination-immune while the axially wide first field both detects the fault and gets discarded from the angle calculation. That insight, not the electronics, is what the patent's validity hangs on.

4. Explicit uncertainties

• Not a verified absence of anticipation. I did not obtain full texts/machine translations of CN 102667412 A, CN 102954755 A, CN 104006833 A, CN 106706012 A, JP 2012-037357 A, JP 2015-158423 A, WO 2017/056575 A1, JP 2007-278831 A (full text), or JP 2017-003430 A (full text). The "no anticipation" conclusion for those is based on title/abstract/citation-context, not a complete element-by-element read. The conclusion is strongest for the references I read in substance (US 2004/0263160 A1, US 9,457,794-family Toyota disclosure, JP 2001-201364 A abstract, Fanuc JP 2017-003430 A abstract, JP 4983076 B2).
• Date conventions: all publication dates above are the dates carried in the Google Patents record; JP "publication date" = kokai publication, which is what governs its status as a printed publication under § 102(a)(1). For US 10,408,892 B2, its 2019 grant is after the 2018-09-20 filing, so it is only available as art through the § 102(a)(2) route (earlier effective filing), not as a § 102(a)(1) printed publication.
• Prosecution history gap (carried over): I have not retrieved the Ex Parte Quayle action (mailed 2019-12-05) or the notice of allowance (2020-01-27). Nothing in the record explains why no prior-art rejection of record attaches to these 18 citations; the Quayle/allowance sequence is consistent with allowance over the cited art, but I cannot state the examiner's reasoning from this record.
• Two citation-integrity issues remain uncorrected and should be independently verified: the 18- vs 19-row citation count (resolved above as a family double-listing), and the OCR discrepancies between the granted-PDF citation strings (JP 2015158432 A, JP 2017004340 A, JP 2012-07357 A, WO 2017055575 A1) and the Google Patents structured list (JP 2015158423 A, JP 2017003430 A, JP 2012037357 A, WO 2017056575 A1).

Sources: Google Patents record for US 10,627,256 B2 (https://patents.google.com/patent/US10627256/en) and the granted-PDF OCR (https://patentimages.storage.googleapis.com/b9/2e/34/bde2210443b662/US10627256.pdf); US 2004/0263160 A1 (Toyota) (https://patentimages.storage.googleapis.com/6c/9c/af/c307f3f2fadb7d/US20040263160A1.pdf); JP 2001-201364 A (NTN) abstract (https://www.sumobrain.com/patents/jp/Anomaly-detecting-circuit-magnetic-encoder/JP2001201364A.html); JP 4983076 B2 / JP 2007-278831 (Seiko Epson) (https://patents.google.com/patent/JP4983076B2/en); US 2016/0363469 A1 (Fanuc, JP 2017-003430 family) (https://www.freepatentsonline.com/y2016/0363469.html).

I'll ground the obviousness analysis by checking the content of the most technically pivotal cited references, then build the § 103 case.

Let me verify two more pivotal references — the TDK multi-detection-circuit sensor and the NTN magnetic-encoder anomaly circuit.

Obviousness Analysis — US 10,627,256 B2 under 35 U.S.C. § 103

Scope note. This analysis builds on the bibliographic, claim and prior-art sections already produced. I do not repeat the claim summaries or the reference list. Where I characterize a reference's disclosure, I ground it in text I retrieved this session and cite the URL; where I could only see title-level metadata, I say so.

Evidence-quality caveat up front. I retrieved and read substantive text for JP H11-153451 (Fanuc), US 2004/0263160 A1 (Toyota), US 8,493,572 B2 (Mitutoyo), US 9,297,635 B2 (TDK), and the Fanuc foreign-substance family (US 2016/0363469 A1 / JP 2017-003430 A). I hit a tool step-limit before pulling JP 2001-201364 A (NTN), JP 2007-278831 A (Seiko Epson), US 5,754,042, US 6,100,682 and US 2009/0315543 A1. Those five are characterized below from their titles/abstracts as listed on the patent's face only, and are flagged as such. I also surfaced one highly material reference that is not on the face of the patent: US 4,370,614 (Fujitsu Fanuc, 1983), which I flag as such.

1. Legal framework and the person of ordinary skill

I apply the Graham v. John Deere framework as elaborated by KSR Int'l v. Teleflex (2007): (a) scope and content of the prior art; (b) differences between the prior art and the claims; (c) level of ordinary skill; (d) secondary considerations. Under KSR, a combination is obvious where the improvement is "a predictable use of prior art elements according to their established functions," where there is "a design need or market pressure" with "a finite number of identified, predictable solutions," or where a known technique is applied to improve a similar device in the same way. MPEP 2143 supplies the articulable rationales; MPEP 2144.04 governs teaching-away.

PHOSITA definition I would propose: a person holding a B.S. in electrical or mechanical engineering (or equivalent) with 2–4 years of experience designing magnetic or optical rotary encoders for machine-tool/servo motor feedback, or an M.S. with less experience. That person would be familiar with: magnetoresistive (MR) bridge sensing of gear/tooth rotors; generation of 90°-shifted quadrature channels by offsetting sensors by ¼ tooth pitch; threshold-based fault detection on encoder channels; and MR element geometry as the determinant of "detection field" footprint.

Priority-date-critical point: all five references I would lead with predate the 2017-09-26 priority date by years and are § 102(a)(1)/102(b) art. Notably, JP 2017-003430 A published 2017-01-05 and US 2016/0363469 A1 published 2016-12-15 — both inside the 2017 window but before the priority date, so they are available notwithstanding same-applicant status (they are pre-priority-date publications, not § 102(b)(2) exceptions).

2. Claim 1 — element-by-element mapping

Conclusion on claim 1: novelty appears intact — no reference of record teaches axial asymmetry of the two detection fields. Obviousness of (d) is the battleground.

3. The core § 103 case for claim 1

Combination A — Toyota problem-recognition + Mitutoyo solution-strategy + Fanuc baseline

Primary: JP H11-153451 A. Supplies (a), (b), (c) — and by the patentee's own admission.

Secondary 1: US 2004/0263160 A1 (Toyota). This reference is unusually on-point for the problem. It expressly teaches that a magnetic rotation sensor "cannot be sealed," that "foreign matters inevitably penetrate," that "iron fragments… may adhere to the magnetic bodies disposed on the outer periphery portion of the rotor and may disturb a change in magnetic flux," and it claims "an abnormality determination portion that monitors an apparent fluctuation… and that determines that the magnetic rotor is in an abnormal state." patentimages PDF This is the same failure mode the '256 patent identifies as its motivation.

Secondary 2: US 8,493,572 B2 (Mitutoyo). Supplies the strategy. It teaches (i) that "contaminants such as dust or oils on a scale track may disturb the pattern detected by the readhead, creating errors," (ii) that conventional responses (disabling signals, error flags) "do not provide a means of continuing accurate measurement operations despite the abnormal signals," and (iii) a detector with multiple sub-portions whose "least-similar" outlier is excluded while the remaining sub-portions are used to compute position. It also characterizes prior art (the '803 application) as teaching "two or more photo detectors… each input to respective signal stability judging means," with abnormal signals "excluded from position measurement calculations." Google Patents

Articulated motivation (MPEP 2143): Toyota identifies localized foreign matter on a rotor as a real, recurring failure that corrupts the sensed signal. Mitutoyo teaches that the recognized answer to localized scale contamination is spatial differentiation of sampling regions plus selective exclusion of the affected region. Given those two teachings, a PHOSITA seeking a "clean" reference channel has a finite, identifiable set of options: (i) duplicate the detector at a different axial position (yields claim 4's approach), (ii) duplicate it at the same position (TDK '635, below), or (iii) shrink the sampling footprint of one detector so it views a region less likely to be contaminated. Option (iii) is the predictable use of a known element (an MR element of smaller axial dimension, which necessarily produces a smaller "detection field") according to its established function.

Design-choice rationale: KSR and MPEP 2144.04 expressly list "changes in shape" and dimensional design choices as within the routine artisan's toolkit. The '256 specification concedes the causal chain is mechanical, not surprising: foreign material in the wide field "takes a value different from that when the foreign material EX is not attached," while the same material is "outside the range of" the narrow field. Nothing in the specification alleges an unexpected result, a result contrary to expectation, or a new mode of operation.

Combination B — same-applicant art (strongest motivation, and a § 102 risk for claim 6)

Primary: JP 2017-003430 A / US 2016/0363469 A1 (Fanuc) — "Rotation angle detector capable of detecting entry of foreign matter." Retrieved text shows it discloses: a rotating body with first and second track parts; detection units generating first and second signals; an abnormality detecting unit with a fluctuation-range calculating unit comparing the fluctuation ranges of the two signals and a judgment unit declaring abnormality when the difference exceeds a judgement value; a phase detecting unit for detecting the phase within one revolution; and judgement "when the difference… exceeds a fluctuation range judgement value in the same phase when the rotary shaft rotates more than once." freepatentsonline

• This maps onto claim 5 nearly verbatim (compare the two signals → threshold → abnormality signal) and onto claim 6 nearly verbatim (repeat events at the same phase within one revolution, "more than once" = the specification's "two or greater"). For claim 6 I would assess this as a serious § 102 anticipation risk, not merely § 103.
• It is the same applicant, same technical field, same stated problem ("entry of foreign substances"), filed 2015-06-10 — which supplies an overwhelming motivation to combine under MPEP 2143 (KSR: familiar elements; common assignee; same field of endeavor).

Secondary: US 9,297,635 B2 (TDK). Discloses a magnetic sensor with first and second detection circuits and "a computing unit that generates an abnormal-event determination signal… by computation using the first detection signal and the second detection signal," where the computation "may include determining the sum" or "the difference" of the two signals. Google Patents This is a direct hit on claim 5's "signal comparator… output a difference."

Combination C — the "quadrature baseline" enhancement. Adding US 4,370,614 (¼-pitch offset → 90° phase lag/lead for direction detection) confirms (c) was a 1983-era commonplace, and JP H11-153451 itself discloses an untoothed portion used as the "origin position" and "one-rotation signal" detection — i.e., claim 7's "second detected portion used for detecting the phase in one revolution" plus a phase detector.

4. Dependent claims

5. Counterarguments the patentee would raise (and how strong they are)

1. No reference teaches, suggests, or renders obvious axial asymmetry of the two fields. This is the patentee's best argument and it is genuinely strong on the record as cited. None of the 18 cited references, on the text I could retrieve, teaches deliberately making the two channels geometrically different. All of the cited encoder art (H11-153451, GM '042, GM '682, TDK '635, TDK '884) teaches identical or near-identical detectors.
2. Teaching away / symmetry expectation. The second channel exists to be a phase-shifted twin of the first (for direction, interpolation, and cross-checking). A PHOSITA would expect the two to be matched; deliberately de-matching them risks amplitude/phase mismatch that degrades the very angle computation the device exists to perform. This is a KSR-era "change in function / teaching away" argument. Strength: moderate. It is blunted by the fact that the '256 specification asserts the A- and B-signals "differ only in phase but have substantially the same waveform" despite the size difference — the patentee's own disclosure shows the mismatch is manageable, which cuts against unexpectedness.
3. The end-concentration premise is an unproven assertion. The patent asserts contaminants/scratches cluster at axial ends but points to no evidence. If a petitioner cannot locate that premise in the art, the placement/geometry argument weakens and claim 3 in particular becomes the hardest to invalidate. Strength: moderate-to-strong for claim 3; weak for claim 1, since claim 1 does not require centering.
4. Prosecution signal. The application issued after an Ex Parte Quayle action and a Notice of Allowance rather than a merits § 103 rejection, meaning the examiner evidently did not locate art teaching the asymmetry. That is mild evidence of patentability, non-binding on a later challenger.

6. Secondary considerations

The record I have shows no objective indicia: no evidence of commercial success with nexus, no licensing, no copying, no industry praise, no unexpected results alleged in the specification, and — per the earlier section — no litigation or PTAB activity that would have developed such a record (that earlier finding was a non-finding, not a verified absence). Absent objective indicia, the Graham factor (d) is neutral-to-favorable for a challenger.

7. Overall assessment

• Claim 1 would likely survive a § 103 challenge on the art of record as I can verify it, because element (d) — the asymmetric detection-field width — is not taught. It is vulnerable only via a well-articulated KSR "predictable design choice / obvious to try" theory anchored in Toyota '160 (problem) + Mitutoyo '572 (spatial differentiation and channel exclusion). That theory is defensible but is the patenteestrongest rebuttal point.
• Claims 5 and 6 are the most vulnerable — claim 6 in particular approaches anticipation by the same applicant's own JP 2017-003430 A / US 2016/0363469 A1 (published 2017-01-05 / 2016-12-15, both pre-priority), which recites judging abnormality when the two-signal difference exceeds a threshold in the same phase when the shaft rotates more than once. A petitioner should lead with this reference; the fact that the examiner listed it as cited art but did not reject over it is worth investigating in the prosecution history.
• Claims 4, 7 and 8 are readily obvious over Mitutoyo '572 (+'803), JP H11-153451, and Toyota '160.
• Recommended next evidentiary steps (outside this session's scope): (i) retrieve full text of US 5,754,042, US 6,100,682 and US 2009/0315543 A1 to check for any asymmetric MR-element geometry — GM '682's "three-element" arrangement is the most plausible latent source of a size-difference teaching; (ii) retrieve JP 2001-201364 A (NTN) and JP 2007-278831 A (Seiko Epson) full text to confirm their threshold/exclusion disclosures; (iii) obtain the Ex Parte Quayle action contents to see whether claim 1's "magnetic" narrowing was made for § 112 or § 103 reasons.