US 10564005

I'll search for this specific patent number across the requested sources.

Let me run a couple more targeted searches to confirm litigation/docket status.

US 10,564,005 B2 — "Position determining sensor unit"

Sources searched: Google Patents (authoritative full text supplied), Justia Patents, PatentCenter-derived data via Google Patents family/legal-events table, IP Verse (TDK-Micronas litigation/opposition tracker), and general web search for CAFC 2026 dockets. No CAFC 2026 docket and no US district-court or PTAB proceeding naming US 10,564,005 was found. (IP Verse's TDK-Micronas tracker shows 1 EPO opposition filed by TDK-Micronas against a Melexis application and zero PTAB cases for the company.) I cannot rule out a recently filed, not-yet-indexed action.

Bibliographic data

Note the two German priorities are distinct applications (…487 vs …488), yielding two separate US patents. Do not conflate them.

Abstract (as issued)

A position determining sensor unit has N sensors at predetermined positions along a path, plus a transducer with a first end movable along the entire path and a length parallel to the path. Each sensor has first and second supply-voltage connections and a switching output switched On/Off depending on whether a sensor-signal threshold is exceeded or undershot. Each sensor's first supply connection goes to a supply voltage; a first sensor sits at the path start and a last sensor at the path end; the first sensor's second supply connection goes to reference potential and the first sensor draws a power consumption. (Continuation is in the claim body — the abstract truncates before the daisy-chain limitation.)

Independent claim 1 — plain-language overview

Claim 1 is a device claim to "a position determining sensor" (the granted text says "sensor," not "sensor unit"). Elements:

1. A current measuring unit.
2. A plurality of sensors spaced at predetermined intervals along a path, each with (a) a first supply-voltage connection, (b) a second supply-voltage connection, and (c) a switching output.
3. A transducer with a first end movable along the entire path and a length extending from that end parallel to the path.
4. The switching output toggles On/Off based on the sensor signal exceeding or falling below a threshold value.
5. Every sensor's first supply connection ties to a supply voltage; the first sensor is at the path beginning and the last sensor at the path end.
6. The first sensor's second supply connection goes to reference potential; the first sensor draws power and is always in the switched-on state when the device is in operation (a limitation added during prosecution relative to the published application).
7. Each further sensor's second supply connection is tied to the switching output of its immediately preceding sensor, so each further sensor is switched on/off by that preceding output, and draws power when switched on. (This is the cascade / "daisy-chain" — the core of the invention.)
8. The current measuring unit is placed either before the first sensor's first supply connection or before the first sensor's second supply connection — i.e., wherever it can see the whole summed current.
9. The sum of the power consumption of all switched-on sensors forms an aggregate current through the measuring unit, and that aggregate current is a multiple of the single-sensor power consumption and is proportional to the position of the transducer's first end.

Point of novelty: position is encoded in how many identical sensors are energized (plus, in PWM variants, a fractional current), so no addressing, no bus, and no reading of individual sensor outputs is required — only one aggregate current measurement.

Dependent claims (2–18), grouped

• Claims 2–5 — operating modes and resolution.
  • 2 (non-inverting, digital): On when a first threshold is exceeded, Off when it falls short; transducer extends from its end at least to S1; position over the m-th sensor ⇒ Isum = (m+1)·Isup.
  • 3 (non-inverting, PWM fine interpolation): Two thresholds per sensor; PWM of the switching output between them, duty cycle proportional to the sensor signal; a sensor fed by a PWM output draws Ipwm < Isup; position over the m-th sensor ⇒ Isum = [m]·Isup + Ipwm. ⚠️ Literal-text warning: the granted claim as displayed reads Isum = I/Isup + Ipwm, which appears to be a transcription artifact; the specification states Isum = 3·Isup + Ipwm and describes an integer multiple of Isup plus Ipwm, so the mathematically coherent reading is m·Isup + Ipwm. I flag this as an apparent textual defect rather than auto-correcting it.
  • 4 (inverting, digital): Off when the first threshold is exceeded, On when undershot; transducer only needs to span a partial region; position over the m-th sensor ⇒ Isum = m·Isup.
  • 5 (inverting, PWM): As claim 4 plus inverse-proportional PWM between the two thresholds ⇒ Isum = (m−1)·Isup + Ipwm.
• Claims 6–8 — practical/accuracy: sensor power consumption variance ≤10% (6); consumption stabilized or trimmed (7); sensor intervals substantially identical (8).
• Claims 9–10 — transducer geometry: first end formed as a tip or edge (9); transducer-to-path distance constant, or increasing near the first end (10) — used to linearize the PWM transfer characteristic.
• Claims 11–12 — sensor type: magnetic field sensors (11); Hall sensors with laterally or vertically measuring Hall plates (12).
• Claims 13–15 — switching output: open-drain transistor (13), with ≥100 mA current-carrying capacity (14) and/or ≤100 mΩ input resistance (15).
• Claim 16 — the last sensor may be replaced by a resistor.
• Claim 17 — sensors may be capacitive, inductive, temperature, force, or pressure sensors.
• Claim 18 — sensors are identical or substantially identical.

Prior art of record

Of the 29–30 cited references, the closest framing references appear to be WO 2015/192965 A1 (Caterpillar Global Mining Europe — Hall switches in a row switchably grounding nodes between series resistors) and US 8,122,159 B2 / DE 11 2009 004 394 B4 (Allegro MicroSystems — daisy-chain address determination). Also cited: US 5,140,225 (barber-pole mark/space ratio displacement), US 2015/0084619 A1 (Tyco — series magnetic field sensors), US 2016/0176215 A1, and US 2017/0254669 A1 (the sibling TDK-Micronas "Position determining unit").

Sibling application granted as a separate patent

US 2017/0254669 A1 (app. 15/449,650, priority DE 10 2016 002 488.7) issued as US 10,429,209 B2 on 2019-10-01, same inventors and assignee, EP 3214412 B1. It is not the same document as US 10,564,005.

Confidence and uncertainty

• High confidence: all bibliographic fields, the abstract, claim text, mode structure (non-inverting/inverting × digital/PWM), and family data — these come from the authoritative full text you supplied.
• Uncertainty: (a) the arithmetic form of claim 3 ("I/Isup") is incoherent as literally rendered and conflicts with the specification; (b) no CAFC 2026 docket, PTAB proceeding, or US litigation could be located for this patent — treat "no proceedings exist" as unresolved rather than established, since I can only search indexed public sources and not the courts' live PACER/CM-ECF dockets directly; (c) I did not verify the current fee/status entry against the live USPTO PatentCenter record itself — the status shown derives from Google Patents' captured legal-events data.

I'll search for litigation involving US patent 10564005.

Let me refine my searches with more targeted queries.

Let me try more specific searches targeting litigation databases.

Litigation Search — U.S. Patent No. 10,564,005

Bottom line

I found no evidence of any litigation involving U.S. Patent No. 10,564,005. Based on the searches I was able to run (see below), there are no known district court cases, Federal Circuit appeals, ITC investigations, or PTAB proceedings (IPR/PGR/CBM) in which the '005 patent is identified as an asserted or challenged patent.

Please treat this as a "no known litigation found" result rather than a definitive negative — I was unable to query the paid/comprehensive litigation databases (Docket Navigator, Lex Machina, Westlaw/Law360 Litigation, PACER docket search) directly, and those would be the authoritative sources for a conclusive answer.

What I searched

What I did establish about the patent's posture

• Assignee/Patent Owner: TDK‑Micronas GmbH (Freiburg, Germany). Assignment recorded 2017‑04‑07; inventors Joerg Franke and Klaus Heberle.
• Priority: DE 10 2016 002 487.9 (2016‑03‑03); filed 2017‑03‑03; granted 2020‑02‑18.
• Status: Active, adjusted expiration 2038‑04‑20; 4th‑year maintenance fee paid August 2023. There is no recorded reissue, reexamination, or adverse legal event in the file that would typically accompany assertion litigation.
• Related family members: U.S. Publ. No. US20170254668A1, EP3214411B1, DE102016002487B3, and a sibling U.S. application published as US20170254669A1 ("Position determining unit," which issued as U.S. 10,429,209).
• Third‑party patent-practice activity around TDK‑Micronas iis patent-office-only: I saw a TDK‑Micronas European opposition against a Melexis application (dated Apr. 30, 2026), and commentary indicating TDK‑Micronas has filed PTAB cases and faced none — none of which involve the '005 patent.

Case-by-case listing

None. There is no case for which I can supply a plaintiff, defendant, jurisdiction, case number, filing date, or status, because none was identified.

Caveats and recommended verification

1. Coverage limits. The patent is only ~6 years old (granted 2020) and, as a component-level automotive/industrial sensor patent, may be asserted against customers or competitors under NDAs or in settled suits that never generate public filings I could surface. Litigation could also exist in foreign jurisdictions (Germany/EPO opposition at the European Patent Office is a real possibility given EP3214411B1).
2. Confirm against authoritative sources. To close this out definitively, check:
   • PACER / CourtListener RECAP — search party name "TDK-Micronas" and the patent number in complaints.
   • Unified Patents Portal — https://portal.unifiedpatents.com/litigation/caselist and the PTAB case list.
   • USPTO Patent Center / Global Dossier for the '005 patent — "Litigation" and "Proceedings" tabs.
   • EPO Register for EP3214411B1 to check for a filed opposition.

I did not fabricate any case details; where I could not find litigation, I have said so explicitly rather than supplying placeholder case numbers.

I'll verify the structured "no proceedings" default against live sources before writing the memo.

Those results were noise. Let me run tighter, party-focused queries.

One more corroboration pass focused on the patent number and possible challengers.

PTAB / AIA Trial Proceedings — U.S. Patent No. 10,564,005

Canonical source used: the structured "PTAB proceedings on file" block (USPTO Open Data Portal ingest), which returns zero AIA trial proceedings. Web-search corroboration was run and returned no hits for this patent number; note my verification pass was cut short by a tool-call limit, so treat coverage of very recently filed (post-ingest) petitions as incomplete.

Proceedings overview

Zero AIA trial proceedings — 0 IPRs, 0 PGRs, 0 CBMs, 0 derivations — have been filed against US 10,564,005 (no active proceedings, no claims invalidated, no claims sustained, no settlements, no institution denials), so the patent stands completely untested at the Board: every one of claims 1–18 is presumptively valid and available to the patent owner, and a defendant gets no PTAB-based shortcut, no FWD to lean on, and no § 315(e)(2) estoppel to exploit — the patent is an unblemished assertion vehicle, not a hardened one and not a damaged one.

No proceedings to report

There is no proceeding number, petitioner, panel, institution decision, FWD, or appeal to describe. I will not manufacture one. What I can document is the verification trail:

The company-level IP Verse data corroborates the patent-level result: TDK-Micronas has faced no PTAB cases at all, and has filed none. Its only recorded post-grant activity is offensive and European — a 2026-04-30 EPO opposition against Melexis's EP 4 056 956 ("Position Sensor System And Method, Robust Against Disturbance Fields"). That is consistent with the earlier litigation section of this analysis and is a useful behavioral signal: the patent owner litigates its sensor portfolio in Europe, not at the PTAB.

⚠️ Near-miss trap — do not confuse with IPR2024-00083

Search engines and practitioners' notes freely use the shorthand "the '005 patent" for a different patent that was invalidated:

IPR2024-00083 — LifeCore Fitness, LLC d/b/a Assault Fitness v. Woodway USA, Inc. — U.S. Patent No. 11,465,005 B2 ("manually powered treadmill"). FWD entered 2025-05-06 (Paper 61) holding claims 1–4, 6, 7, 9–13, 15, and 16 unpatentable as obvious over Socwell, Magid, and Sclater, with claim 8 surviving; patent owner's notice of appeal to the Federal Circuit was dated 2025-07-08. That is US 11,465,005, a treadmill patent, and has nothing to do with US 10,564,005 (TDK-Micronas, position-determining sensor unit). If a vendor or an AI tool hands you "IPR2024-00083 invalidated the '005 patent," it is wrong. Applying that FWD to a TDK-Micronas demand letter would be a serious error.

Strategic summary

Claim status. All of claims 1–18 are untested — there is no CANCELED set and no SUSTAINED set, because no claim has ever been construed or judged by the Board. Claim 1 (the sole independent claim) is intact as issued, including the limitation added during prosecution that the first sensor "is always in a switched-on state when the position determining sensor is in operation." Note the one caveat flagged in the earlier summary: claim 3's aggregate-current equation is rendered as Isum = I/Isup + Ipwm, which is mathematically incoherent and conflicts with the specification's stated Isum = m·Isup + Ipwm (the spec works the example as 3*Isup + Ipwm). A defendant should determine on the face of the printed patent whether that is a printing/OCR artifact in the Google Patents rendering or a genuine certificate-of-correction issue — but do not assume it helps you: it is a dependent claim, and claim 1 does not depend on it.

Estoppel landscape. There is none. § 315(e)(2) estoppel only attaches to a petitioner that has been party to an instituted IPR and obtained an FWD — no such petitioner exists here. Likewise there is no adverse-judgment or termination history. That cuts both ways:

• For the patent owner: no narrowing amendments, no prosecution-disclaimer statements extracted under Board pressure, no IPR record to mine — you are litigating the patent as examined in 2017–2020, with a fresh § 112 and claim-construction clean slate.
• For the defendant: you have a fully open prior-art field. Every ground the examiner did not consider is fair game, and remember that the '005 family's own prosecution-facing art is unusually rich — WO 2015/192965 A1 (Caterpillar, Hall switches switchably grounding nodes between series resistors — cited on the face) and US 8,122,159 B2 / DE 11 2009 004 394 B4 (Allegro, daisy-chain addressing — cited on the face) are the two references that frame this invention, and the family has since been cited against at least one later TDK-Micronas filing.

Pattern signals. No serial petitioner exists (there is no petitioner). The patent owner has never appealed to the Federal Circuit from any PTAB decision on this patent and has never had to defend it at the Board. There is no defensive aggregator (e.g., Unified Patents) in the chain — Unified's portal returned no hits for this patent. Combined with zero US district-court litigation surfaced in the earlier sections and zero PTAB filings, the profile is that of a portfolio patent that has not been commercialized adversarially — plausibly held for cross-licensing or for enforcement against OEM customers under NDA, where public filings would not appear.

Timing windows. Because the grant date is 2020-02-18:

• PGR under § 321(c) is time-barred — the 9-month window closed on or about 2020-11-18.
• CBM is unavailable twice over: the claims are not directed to a financial product or service, and the transitional CBM program sunset for filings after 2020-09-16.
• IPR remains fully available forever under current law — the § 315(b) one-year bar is triggered only by service of a complaint alleging infringement, and no such service has been identified. If you have not been sued, you can file today.

Recommended next steps

1. If you are a defendant already served with a complaint alleging infringement of US 10,564,005, diary the § 315(b) date immediately. It runs one year from service of the complaint (or from service of a complaint on your RPI/privy), and there is no § 315(b) clock running before that. There is no existing proceeding to join and no estoppel to worry about.
2. Because there is no FWD to cite, do not build a PTAB defense on the assumption of an invalid-claims shortcut. Build the invalidity case from scratch, starting with WO 2015/192965 A1 and US 8,122,159 B2 plus closer Hall-switch/aggregate-current art, and commission a focused search on the two-step abstraction the claim actually performs — (a) daisy-chained power gating from switching outputs, and (b) inferring position from a summed current multiple. That second step is the genuine point of novelty and the place a § 103 combination is most likely to be attacked on motivation-to-combine grounds.
3. Consider § 325(d) risk before filing. The examiner already had WO 2015/192965 A1 and US 8,122,159 B2 of record, so a petition built primarily on those references invites discretionary denial. A materially different primary reference is the safer posture.
4. Verify the record yourself before relying on this memo. My searches were cut short by a tool limit. Close it out with:
   • PTAB E2E / USPTO PatentCenter "Proceedings" tab for US 10,564,005,
   • the ODP API AIA-trial endpoint for application 15/449,505,
   • Unified Patents portal litigation and PTAB lists,
   • CourtListener RECAP for "TDK-Micronas" complaints.
5. Check the sibling patent separately. US 10,429,209 B2 ("Position determining unit," app. 15/449,650, priority DE 10 2016 002 488.7) is a distinct patent. The company-level IP Verse data suggests no PTAB activity on it either, but I did not run a patent-specific check — do not assume its PTAB history mirrors this one.
6. If you need the count to be authoritative and litigation-grade, say so and re-run this pass so I can query the PTAB E2E docket and CourtListener directly rather than relying on the ODP ingest plus general web results.

Bottom line for a defendant: no PTAB proceeding exists on US 10,564,005, so there is no invalidated claim to point at and no estoppel to hide behind — but also no adverse FWD to overcome. Your IPR window is open on all 18 claims, and the absence of any IPR after six-plus years of an active patent is itself a signal that this patent has not yet been worth asserting in a venue that generates public filings.

I'll research this patent's ownership chain, starting with the assignment records and cross-referencing the assignee's corporate status.

I hit the tool-call limit before I could pull the live Assignment Center record directly, so the section below distinguishes carefully between what is verified from the authoritative full text / Google Patents legal-events table and what I could not confirm. I have not fabricated any reel/frame, correspondent, or assignment.

Inventors

• Both inventors are named in the single recorded assignment as assignors to TDK-MICRONAS GMBH, GERMANY (see timeline), which establishes their employment relationship with the assignee at the relevant time.
• No unusual-departure pattern detected. I found no evidence that either inventor left TDK-Micronas within 12 months of the 2016-03-03 priority filing, and no data source surfaced indicating a portfolio fire-sale. (Absence of evidence here, not affirmative evidence of retention — the company does not disclose employee rosters.)
• The same two inventors also name the sibling application US 15/449,650 (issued as US 10,429,209 B2), so this is a two-inventor team filing a family of two applications, not a one-off inventor transfer.

Original assignee

TDK-Micronas GmbH, Hans-Bunte-Straße 19, 79108 Freiburg im Breisgau, Germany.

• Entity type / status: Operating company; an operating subsidiary of TDK Corporation (Tokyo, Japan). Status: active, not dissolved, not in bankruptcy. The predecessor entity Micronas was acquired by TDK Corporation in 2015 and renamed TDK-Micronas — i.e., the corporate-name event predates this application's 2016 priority and is reflected as an internal change of name, not a transfer away from the operating business.
• Primary line of business: Magnetic-field sensors and CMOS integration within the TDK group — Hall-effect and TMR switches, linear sensors, 3D position sensors, current sensors, and embedded motor controllers for automotive and industrial markets.
• Does it ship product embodying the claims? The closest product families are the HAL 37xy 3D position sensor family (linear position up to ~40 mm) and the masterHAL® HAL 39xy family, both marketed for linear/angular position measurement. The company states it has shipped "over six billion Hall sensors." This is a genuine operating company with manufacturing and R&D in Freiburg (~1,000 employees), not a licensing shell. (Source: TDK-Micronas company profile, https://www.micronas.tdk.com/index.php/en/company/profile)

Assignment timeline

I found exactly one recorded assignment in the chain, and no post-issuance assignments whatsoever. The single record is drawn from the Google Patents legal-events table for US 10,564,005 (which mirrors USPTO assignment data):

• 2017-03-07 (executed) / recorded 2017-04-07 — Reel 041929/0860
  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Joerg Franke and Klaus Heberle (the inventors)
  • Assignee: TDK-Micronas GmbH, Germany
  • Correspondent: Not retrievable from the sources available to me. Neither the authoritative full text I was given nor the Google Patents legal-events row exposes the correspondent of record (attorney/agent who filed the recording). I could not open the Assignment Center record directly before exhausting my tool budget. I will not guess a name or firm. To capture the correspondent you must open the record at Assignment Center (search patent 10564005 / reel 041929 frame 0860).
  • Context: Standard pre-issuance inventor-to-employer assignment — the inventors' statutory assignment of the invention to their employer, executed four days after the 2017-03-03 filing date and two days after the 2016-03-03-priority German application. This is origin-transfer, not a change of ownership event.

No other records exist for this patent. There is no security agreement, no merger, no subsequent assignment to any holding/licensing entity, no license record, and no release. The patent therefore remains owned by the original operating-company assignee.

One adjacent record worth flagging (not a link in this patent's chain): I found a PlainSite-indexed assignment "Micronas GmbH → TDK-Micronas GmbH" (https://www.plainsite.org/patents/assignment.html?id=[8338219](/patent/8338219)). Its schedule lists Micronas applications filed in 2006–2009 and does not include US 15/449,505 (filed 2017, i.e., after the TDK acquisition). This is the corporate-name/ownership cleanup from the 2015 TDK acquisition applied to the pre-acquisition portfolio; it is not a link in the '005 chain, and I am not representing it as one.

Timeline diagram

timeline
    title Ownership of US 10564005
    2015 : Micronas renamed TDK-Micronas after TDK acquisition
    2016 : DE priority application filed
    2017 : US application filed 2017-03-03
         : Inventors assign to TDK-Micronas reel 041929/0860
    2020 : Patent issued 2020-02-18
    2023 : 4th year maintenance fee paid

NPE / troll-pattern signals

Verdict

Insufficient data — by the literal criterion ("only the original assignment").

Justification: The entire ownership record consists of the single pre-issuance inventor-to-employer assignment executed 2017-03-07 and recorded 2017-04-07 at reel 041929/0860, naming Joerg Franke and Klaus Heberle as assignors and TDK-Micronas GmbH as assignee. There are no post-issuance assignments, no shell transfers, no known asserters, no cascading LLCs, and no defensive-aggregator terminus — the patent remains with its original, product-shipping operating-company owner (a TDK Corporation subsidiary that manufactures HAL-series Hall position sensors). In other words, every NPE signal is affirmatively absent, and the only reason the verdict is not a clean "not an NPE" is that the framework's non-NPE options presuppose assertion activity that does not exist here.

Verification link (Assignment Center, search by patent number 10564005 or reel 041929 frame 0860): https://assignmentcenter.uspto.gov/ — corroborated by the Google Patents legal-events table at https://patents.google.com/patent/US10564005/en.

Two caveats to close out, stated plainly rather than papered over:

1. I could not retrieve the correspondent of record for reel 041929/0860, and I did not fabricate one. That is the single most useful field for your repeat-attorney analysis, so it is worth pulling manually.
2. I could not query the Assignment Center live; the reel/frame above traces to the USPTO-derived legal-events data in the authoritative text. If the live Assignment Center shows an additional unindexed record (e.g., a 2023+ change of name or intra-TDK transfer), it would post-date the sources I could reach.

I'll research the prior-art references cited against US 10,564,005, verifying bibliographic details and technical content for the references most likely to bear on patentability.

Let me gather the remaining key references.

Prior-Art Analysis — U.S. Patent No. 10,564,005 B2

0. Patent identification (USPTO record, literally interpreted)

The number 10564005 resolves to U.S. Patent No. 10,564,005 B2, titled "Position determining sensor unit", application 15/449,505, filed 2017-03-03, granted 2020-02-18, priority DE 10 2016 002 487.9 (2016-03-03). Assignee TDK-Micronas GmbH; inventors Joerg Franke / Klaus Heberle.

I did not substitute or round this number. Note two adjacent identifiers that must not be conflated:

Caveat on tooling: the authoritative full text (Google Patents, which mirrors the USPTO/Global Dossier record) was supplied and is treated as controlling. My web searches supplemented descriptions of the cited references but did not change any bibliographic field. Where I could not independently verify a reference's substantive content, I say so and rely only on the citation table's title/assignee/date (which is itself from the USPTO-derived record).

1. Critical framing point before the claim-by-claim mapping

Every one of claims 2–18 depends from independent claim 1. Therefore, under §102 a reference can only anticipate a dependent claim if it discloses all of claim 1 plus the added limitation. Claim 1 requires the concurrence of three things no single cited reference discloses together:

1. a transducer with a first end movable along an entire path of spaced sensors;
2. a daisy-chain power architecture in which each further sensor's second supply-voltage connection is wired to the switching output of its immediately preceding sensor; and
3. a single current-measuring unit whose aggregate current is an integer multiple of one sensor's power draw and proportional to transducer position.

As analyzed below, the two closest references each supply only part of this combination (→ §103 material), and no cited reference anticipates claim 1 or, derivatively, any claim. I flag this explicitly rather than manufacture an anticipation theory.

2. The 29–30 cited references, with relevance ranking

Tier 1 — closest art (drives §103, not standalone §102)

Why Tier 1 does not anticipate. Reference A senses position with a transducer over a sensor row but encodes it as a voltage-divider change using series resistors — the opposite architecture to the claimed "second supply connection chained to the preceding sensor's switching output" with a summed current. Reference B has exactly the claimed daisy-chain-with-current-measurement topology, but there is no transducer moving along a sensor path and no position-proportional aggregate current. The '005 invention is the confluence of A's position-sensing front end and B's current-summing daisy-chain back end — a classic §103 combination, not §102 anticipation.

Tier 2 — secondary references (element-level §102 relevance only; cannot anticipate claim 1 alone)

Tier 3 — background / peripheral citations (no realistic §102 bearing on claim 1)

(US 4,367,506 A appears in the table at both the citation and the DE-family-filing lists; the substantive count is the ~29–30 references shown.)

3. §102 finding per claim

Claim 3 textual caveat (carried forward from the summary): the granted claim 3 arithmetic is displayed as Isum = I/Isup + Ipwm, which is dimensionally incoherent and conflicts with the specification (Isum = 3·Isup + Ipwm, i.e. m·Isup + Ipwm). I do not auto-correct it; I flag it as a probable transcription/typo defect. It does not change the anticipation analysis, since no reference discloses the claim-3 combination regardless.

4. Bottom line

• No cited reference anticipates any claim of US 10,564,005. Because all claims depend from claim 1, and no reference discloses the three-part claim-1 combination (transducer-over-sensor-path + daisy-chained second-supply switching + single aggregate-current position readout), §102 anticipation fails for claims 1–18.
• The genuine validity attack is §103: WO 2015/192965 A1 (Caterpillar) supplies the position-sensing front end (row of Hall switches + movable magnet + threshold switching) and US 8,122,159 B2 / DE 11 2009 004 394 B4 (Allegro) supplies the daisy-chained, common-port, current-measured component architecture whose measured current indicates component position. Combined, these two — both cited in the '005 file — render the core claim-1 concept (and dependent claims 2, 6–8, 11, 12, 18) obvious. Secondary references C, D, E, F, G reinforce the PWM-resolved position (claims 3, 5, 9–10) and threshold-monitoring (claims 6–7) subject matter.
• Not prior art: US 2017/0254669 A1 / US 10,429,209 B2 (same family, commonly owned — same inventors, same assignee, shared priority date).

Confidence and uncertainty

• High confidence: all bibliographic fields, the reference list, assignee/title/date data, and the claim-element mapping — these derive from the authoritative full text and the USPTO-derived citation table.
• Moderate confidence: my technical characterizations of references C–I beyond their titles were reconstructed from supplementary searches; the Tier-1 references (A, B) were verified against primary text.
• Uncertainty I will not paper over: (i) §102(a)(2) eligibility of the two post-2016-published references (US 2017/0176215, US 2018/0172475) depends on whether they are US-originating publications with earlier effective filing dates — unverified; (ii) I could not retrieve full text for a few Tier-3 references, so I describe them only to the extent their titles/assignees support. I have not invented any disclosure or mapped any feature a reference does not evidently contain.

I'll ground the analysis by pulling the disclosures of the key prior-art references of record before constructing the §103 combinations.

Let me pull the remaining potentially-closest references, including the liquid-level proximity-sensor array and the sibling TDK reference.

Obviousness Analysis — U.S. Patent No. 10,564,005 B2 under 35 U.S.C. § 103

Scope note. This is a technical invalidity assessment, not legal advice or a litigation opinion. It builds on the earlier sections in this analysis (bibliographic data, claim construction overview, litigation posture) and does not repeat them. Because the application was filed 2017-03-03 with a 2016-03-03 DE priority, it is an AIA patent; § 102/103 as amended apply. All references below are drawn from the prior-art record captured on the Google Patents page for US 10,564,005.

1. Framework

1.1 Hypothetical PHOSITA

A person having ordinary skill in the art would be an engineer (B.S. in EE, physics, or mechatronics, or equivalent) with roughly 2–5 years' experience designing magnetic-field position sensors and Hall-switch ICs, including sensor chains/arrays for automotive and industrial applications, and familiar with current-sense and daisy-chain ("supply-rail signaling") architectures. This is a mature, incremental art: the '005 patent's own Background concedes the field already contains "Hall switches arranged in a row along the path," movable transducers, series resistor chains, and voltage measurement (WO 2015/192965 A1), plus daisy-chain sensor systems (DE 11 2009 004 394 B4 / US 8,122,159).

1.2 Claim 1 as the analytical anchor

Claim 1 requires, in substance: (i) N sensors each having first supply connection, second supply connection, and switching output; (ii) a transducer with a first end movable along the entire path; (iii) threshold-based On/Off switching; (iv) all first supply connections tied to Vsup; (v) first sensor's second connection to reference potential, and that sensor always on; (vi) each further sensor's second connection tied to the switching output of the immediately preceding sensor (the cascade); (vii) a current measuring unit at the first sensor's supply side; and (viii) the summed consumption of all on-sensors forming an aggregate current that is a multiple of the unit consumption and proportional to the transducer position.

Elements (i), (iii), (iv)–(v) and (vii)–(viii) are individually ordinary. The asserted novelty resides in the cascade + single current-sense node, which the specification summarizes in the litigation/validity-relevant passage: "the aggregate current in connection with the fixed positions of the sensors along the path already contain the information about the position of the transducer. This in particular eliminates the need for an elaborate circuit for address allocation and address output."

2. The of-record prior art, and what each actually discloses

Not prior art — flag: US 2017/0254669 A1 / US 10,429,209 B2 (the sibling "Position determining unit," same inventors/assignee, DE …488.7) shares the same 2016-03-03 effective filing date. It is therefore not "effectively filed before" the '005 date under §102(a)(2), and its 2017-09-07 publication post-dates the '005 filing — it cannot be used in any §103 combination. Any analyst conflating it with the Caterpillar or Allegro art would be making an error.

Citation-status caveat: On the captured Google Patents page, WO 2015/192965 A1 and US 8,122,159 B2 appear in the citations list without the "*" (examiner-cited) marker that other entries carry. I cannot reliably determine from that page whether they were examiner-cited, applicant-cited, or third-party-cited. This matters procedurally (examiner-considered art is harder to attack with "the examiner never saw it"), but not substantively for §103.

3. Grounds of rejection

Ground 1 — WO 2015/192965 A1 in view of US 8,122,159 B2 → claims 1, 2, 4, 6–18

Claim element mapping for claim 1:

Under MPEP 2143 rationales (A), (C) and (D), this combination is a textbook case of "use of a known technique to improve a similar device in the same way" and "applying a known technique to a known device ready for improvement to yield predictable results."

Why the PHOSITA would combine them:

1. Same field, same problem, same physical arrangement. Both references place a serial array of identical switch/sensor elements along a path and ask which element a movable member currently actuates, to derive position. WO '965's own background disparages Hall-sensor measuring rods because they "require calibration of each sensor and may increase the total cost." US '159's background states "an addressing scheme should be simple in order to reduce cost" and that daisy-chaining "reduce[s] the number of wires… extra weight… the number of connections in a car." The two references are directed at the identical cost/wiring problem from complementary angles — a strong, express motivation to combine.

2. US '159 supplies exactly the two missing elements. WO '965's switches already switch a node to ground (the "switching output" of claim 1). US '159 takes that same switch and cascades it — output port of one element feeding the input of the next, with the first element tied between supply and ground and all common ports tied to the other — and measures current in that chain to determine location. A PHOSITA seeking to eliminate WO '965's N+1 resistor string and its multi-node signal-conditioning circuitry (Uo detection) has a ready, published substitute that performs the identical function (encode "which element is active") using one node.

3. The aggregate current as position information is not a new principle. US '159 already teaches that the chain's total supply current is information — the master compares "the total ICC consumption seen by the master component" against a threshold, and each slave measures its own IOUT. Once identical loads are cascaded between Vsup and the first sensor's ground return, Kirchhoff's current law makes the aggregate current at the head of the chain equal to N_on × Isup as a matter of ordinary circuit theory. The '005 patent claims no more than this arithmetic consequence.

4. Predictable results / no new cooperation. Combining a Hall-switch array (WO '965) with a cascaded supply-current topology (US '159) yields nothing more than the expected sum of the individual functions; each reference's function is unchanged and the combination merely adds their known benefits (single sense node + cascade switching). Under KSR v. Teleflex, "familiar elements according to known methods… combined in a way that a person of ordinary skill could predict" is obvious.

Claims 2 and 4 (mode polarities) — strongly met. WO '965 expressly states the switch may be "closed (ON) when no magnetic field is present, and… opened (turned OFF) by the magnetic field." That is the inverting embodiment of claim 4 (Isum = m·Isup) verbatim, and the normally-open behavior already described is the non-inverting embodiment of claim 2 (Isum = (m+1)·Isup). The two threshold conventions and their corresponding formulas are thus a disclosed design choice, not an inventive selection. (Their integer offsets are forced by where "S1 always on" sits relative to the transducer end.)

Claims 6–8, 11–12, 13, 16, 18 — see §4, Notes.

Ground 2 — Ground 1 + US 5,140,225 A (Bosch) [or US 5,594,335 (Siemens)] → claims 3 and 5

Claims 3 and 5 add dual-threshold sensors with PWM of the switching output between the thresholds, the duty cycle being ∝ (claim 3) or inversely ∝ (claim 5) the sensor signal, so that Isum gains a fractional Ipwm term for sub-pitch interpolation.

US 5,140,225 discloses precisely the missing technique: from two adjacent magnetic-sensing elements, an evaluation circuit "produces a cyclical rectangular voltage… in which the mark/space ratio equates to the position of the magnet between the two elements," expressly to avoid the "expensive" separate microcomputer interpolation. That is PWM encoding of intra-pitch position — the analog of claim 3's Ipwm, and (with a straightforward polarity inversion) claim 5's inverse duty cycle.

Motivation: WO '965 teaches that resolution is bounded by switch pitch and by the spatial extension of the magnet's field (the magnet must be strong/wide enough that "at any point at least one Hall switch is actuated"), and that using more elements is the only disclosed way to raise resolution. Bosch teaches a way to obtain sub-pitch resolution without adding elements, at lower cost. A PHOSITA would apply Bosch's mark/space interpolation to WO '965's array — Rationale (C), improving a similar device in the same way. US 5,594,335 (Siemens: "interpolation circuit for interpolating analog output signals of the Hall elements") is a corroborating alternative to the same effect. The low-pass/shunt conversion of PWM to an analog level recited in the '005 specification is elementary.

Caveat on claim 3's text. As flagged in the prior section, the granted claim 3 reads Isum = I/Isup + Ipwm, which is incoherent as literally rendered and contradicts the specification (3·Isup + Ipwm in FIG. 1; "an integer multiple of Isup plus Ipwm"). The obviousness conclusion is the same under either the literal or the coherent reading — the claim adds only PWM interpolation on top of Ground 1 — but the discrepancy is an independent §112 vulnerability worth noting.

Ground 3 — Ground 1 + US 5,565,687 A / EP 0 763 723 A2 (Rolls-Royce) → claim 17 (and supporting claim 16)

Claim 17 permits the sensors to be capacitive, inductive, temperature, force, or pressure sensors.

US 5,565,687 — in the same position-responsive architecture of a series of detecting elements along a path with cumulative array signals — teaches that the detecting elements may be Hall-effect devices, and its companion EP 0 763 723 discloses a capacitance probe with a series of capacitive switches whose binary outputs are "combined with a series of other outputs from a series of such proximity sensors to form an apparatus to determine position of an object, particularly a liquid level." This is a simple substitution of one known sensing element for another to obtain the predictable result of sensing a different physical quantity across the same array — Rationale (B), reinforced by KSR's known-technique doctrine. The '005 specification itself concedes these are mere alternatives ("the sensors… can be capacitive sensors or inductive sensors"). No unexpected result is asserted for any of them.

Ground 4 — Ground 1 + US 2015/0084619 A1 (Tyco) → claims 9, 10, and as an alternative route to claims 3/5

Claims 9 and 10 (transducer end as a tip or edge; transducer-to-path distance constant or increasing near the first end) are directed to linearizing the PWM transfer characteristic — the specification says so outright ("an output characteristic can be linearized").

Tyco US 2015/0084619 is directed to the same objective: making the position signal of a series magnetic-sensor array linear over long ranges, including by overlapping measurement ranges, offset correction and learning routines. Additionally, WO '965 itself discusses how the spatial extent of the magnet's field determines which Hall switches actuate, and the '005 specification concedes the geometry is chosen for linearity. Claims 9–10 are therefore best characterized as optimization of result-effective variables with no asserted criticality — obvious under In re Antonie / In re Boesch and KSR.

This is the weakest of the grounds. I could not verify an of-record reference that expressly discloses a tapered/tip-shaped magnet end with an increasing gap to the sensor row. If the patentee is defending validity, claims 9–10 are where a genuine factual dispute over the scope and teachings of the art (and possibly a "teaching away" argument based on Bosch's stated spacing-independence, see §6) is most likely to arise. Grounds 1–3 are materially stronger.

4. Claim-by-claim notes (dependent claims)

5. Consolidated motivation-to-combine statement (for Ground 1)

A PHOSITA at the 2016 priority date, facing the known problem of determining the position of a magnet along a long row of switch-type Hall sensors at low cost and with minimal wiring, would have:

• started from WO 2015/192965 A1, which discloses the row of interval-spaced Hall switches, the movable magnet, and the concept that the number/location of actuated switches encodes position via a single output signal;
• recognized that WO '965's series resistor chain plus voltage-drop evaluation (a) consumes N+1 discrete resistors and (b) requires monitoring a node whose signal is a voltage division that must be resolved to a discrete switch position;
• turned to US 8,122,159 B2, which in the same field and for the same purpose (cost and wiring reduction) teaches cascading the identical switch/output elements between supply and ground and reading the chain's current to learn where in the chain an element is; and
• combined them by cascading each Hall switch's output as the supply return of the next stage and sensing the summed supply current — yielding, by ordinary circuit analysis, an aggregate current proportional to how many identical sensors are on, hence to transducer position.

The asserted advantage of the '005 patent (no bus, no address allocation) is the very motivation already articulated in US '159's own background and is realized by the combination as a predictable consequence. There is no teaching away, no unexpected result, and no change in the principle of operation of either reference.

6. Counterarguments a patentee could raise (and how they fare)

1. "Neither reference discloses measuring the aggregate current at the head of the chain as the position output." Partially true. US '159's per-element current detection is directed at identifying the last active slave and counting addresses, not at superposing N quiescent currents. But US '159's master does monitor total chain ICC against a threshold, and the arithmetic superposition is elementary. Assessment: weak-to-moderate defense; the ground survives, but this is the strongest factual gap and a patentee would press it.

2. "WO '965 is a voltage-drop system and teaches away from current sensing." WO '965 nowhere disparages current sensing, and silent disclosure is not a teaching away (In re Fulton). It also expressly contemplates normally-closed switches (enabling claim 4) and cascading mounting units. Assessment: fails.

3. "US '159 still requires addressing, which the invention eliminates." Requiring addressing for fault diagnosis (US '159's stated rationale) does not preclude using the chain's aggregate current as the position output; the two uses coexist. Assessment: fails, though it is rhetorically attractive.

4. "Bosch teaches that its output is independent of magnet-to-sensor spacing — i.e., teaching away from claim 10's geometry-dependent transfer characteristic." This is a genuine, if narrow, point as to claims 9–10 only; it does not touch claims 1–8 or 11–18. Assessment: plausible as to claims 9–10; immaterial to the core claims.

5. Secondary considerations. I found no objective evidence in the record — no litigation (consistent with the earlier litigation section), no evidence of commercial success, long-felt need, failure of others, copying, or industry praise, and no unexpected-result data distinct from the express cost/wiring advantages already articulated in the references themselves. Any such evidence would also need a nexus to the cascade/aggregate-current limitation, which the specification attributes to generic cost/simplicity benefits rather than to a surprising technical effect. Assessment: no rebuttal weight presently available.

6. Procedural posture. Both primary references appear in the file's citation list (with an ambiguity as to whether they were examiner- or third-party-cited). If examiner-cited, the patentee will argue the examiner considered and allowed over them. That does not defeat a §103 ground — particularly one built on a different combination rationale (Allegro as the cascade/current-sense teaching rather than merely as daisy-chain addressing art) — but it does mean any challenger must meet the clear-and-convincing standard with a clearly articulated motivation that does not merely re-present the references individually.

7. Bottom line

• Claims 1, 2, 4, 6–8, 11–13, 16, 18 — strong prima facie obviousness over WO 2015/192965 A1 in view of US 8,122,159 B2 (Ground 1; claims 13–15 additionally rely on the FET-to-ground disclosure and result-effective optimization of the current path).
• Claims 3 and 5 — strong with the addition of US 5,140,225 (mark/space ∝ position) or US 5,594,335 (Hall interpolation) (Ground 2). Claim 3 additionally carries an independent textual defect (I/Isup) that undercuts any narrow reading.
• Claim 17 — moderate-to-strong with US 5,565,687 / EP 0 763 723 (Ground 3).
• Claims 14 and 15 (numeric open-drain limits) — moderate, dependent on the patentee failing to show criticality; the specification supplies reasons but no criticality data.
• Claims 9 and 10 — weakest; would require a "result-effective variable / design choice" theory and could be met with a spacing-independence teaching-away argument. These are the most likely survival points if the patent is challenged.

The single most defensible proposition is that the '005 patent's asserted point of novelty — using the aggregate supply current of a cascade of identical sensors as a direct, single-node measure of position, with no addressing and no bus — is the predictable combination of a Hall-switch row with a movable magnet (WO '965) and a current-sensed daisy-chained supply topology (US '159), both of which are in the same field and both of which expressly motivate the combination on the same cost/wiring grounds the '005 patent claims as its advantage.