US 10190899

I'll run targeted searches for this specific patent number, then summarize from the authoritative patent text.

US 10,190,899 B2 — Thermal Flow Sensor

Sourcing note: The summary below is drawn primarily from the full authoritative patent text provided (Google Patents copy of US10190899B2, fetched 2026‑09‑24). I also ran web searches for the specific number; I did not find any USPTO litigation record or CAFC docket activity (2026 or otherwise) for this patent, and I have no authoritative information confirming that any exists. Treat the litigation/docket portion as "no results found," not as proof of absence.

Bibliographic data

Abstract

A thermal flow sensor with improved measurement accuracy. It includes an air flow rate detection element having a diaphragm (thin-film portion in a semiconductor substrate); at least one heat generating resistor on the diaphragm; at least one temperature measuring resistor detecting temperature on each of an upstream and downstream side of the heat generating resistor; and a correction circuit portion that processes the element output signal based on temperature-difference information of the upstream/downstream temperature measuring resistors. The processed waveform is one obtained by cutting a part of a "mountain" (peak) or "valley" part constituting a peak value by outputting an arbitrary predetermined value when the peak value of the waveform exceeds that predetermined value.

Independent claim 1 (plain language)

A thermal flow sensor comprising four elements:

1. An air flow rate detection element with a diaphragm formed as a thin-film portion in a semiconductor substrate;
2. At least one heat generating resistor on the diaphragm;
3. At least one temperature measuring resistor that detects temperature on each of an upstream side and a downstream side of the heat generating resistor; and
4. A correction circuit that includes a maximum-value detection circuit and a minimum-value detection circuit, and that processes the detection element's output signal based on temperature-difference information from the upstream/downstream temperature measuring resistors.

The characterizing feature: the correction circuit cuts off part of a peak (mountain) or trough (valley) of the output waveform by outputting an arbitrary predetermined value — where that predetermined value is derived from the maximum and minimum values found by the max/min detection circuits — when the waveform's peak value exceeds that predetermined value. In the specification this operation is called "clamp correction" (the cutoff value is the "clamp value," D_clp).

Dependent claims (brief)

• Claim 2: The predetermined (clamp) value is determined from two parameters: an amplitude value and an average flow rate, obtained from signals based on the element output.
• Claim 3: The average flow rate and amplitude value are obtained using the values detected by the maximum and minimum value detection circuits.
• Claim 4: The signals based on the element output are corrected by a circuit that compensates response delay of the element output.
• Claim 5: The output signal is subjected to a correcting process by a frequency response circuit (specification's example is an HPF).
• Claim 6: The signal fed into the max/min detection circuits is a signal obtained by correcting the output signal in a linear direction based on the flow-rate-vs-output relationship (linearization).

Technical gist (from the specification)

The patent addresses "ripple error" in engine intake-air measurement. A semiconductor (MEMS diaphragm) thermal flow sensor with forward/backward flow detection suffers a lean error from response delay at high-frequency rippling, while the inertial effect of a detour bypass route used to cancel that lean error causes a rich error at high ripple amplitudes (≥ ~200%, up to ~1000% in VTC engines). The disclosed remedy is a correction LSI that detects max/min of the digitized element signal (D_in), computes average flow rate (D_ave) and amplitude (D_amp), selects a clamp value (D_clp) from a map, and clamps (cuts) the mountain—or valley—of the waveform so the average moves toward the true value. The specification argues clamping is superior to additive correction because, when noise contaminates the input, a clamp only distorts the portion exceeding the clamp value, whereas an additive correction shifts the entire waveform. Examples 2–4 add an HPF response-delay compensation, reduce the correction map from 3-D to 2-D, and add linearization so amplitude is detected accurately across differing average flow rates.

Uncertainties / caveats

• I did not locate any authoritative USPTO litigation or CAFC docket record for US 10,190,899; the web search surfaced no litigation hits for the specific number. The one search result containing digits "10190899" was an unrelated European application-number sequence, not this patent.
• Assignee histories (Hitachi Automotive Systems → Hitachi Astemo → Astemo) are as reflected in the patent's legal-events data; Google Patents notes such statuses are informational, not legal conclusions.
• No information beyond the patent document itself (e.g., current claim status after any post-grant proceedings) is confirmed here.

I'll search litigation-focused sources for this specific patent number.

Let me check litigation-specific databases directly for this exact patent number.

Let me try a couple more targeted litigation-database queries.

Litigation search results for US Patent 10,190,899

Bottom line: I found no known litigation involving US 10,190,899 (the "Thermal flow sensor" patent). Based on the litigation-focused sources I searched, this patent does not appear to have been asserted in any U.S. district court, ITC Section 337 investigation, or PTAB post-grant proceeding.

Sources searched and what they returned

I did not obtain direct PACER or CourtListener docket access for the exact number, so this is a "no results found" conclusion rather than an affirmative certification of absence. Consistent with the previously generated summary section — which also found no USPTO litigation or CAFC docket activity — I have no record of any case.

⚠️ Important disambiguation (do not auto-merge these identifiers)

The searches surfaced two near-miss identifiers that are not this patent and must not be treated as litigation for US 10,190,899:

1. U.S. Patent No. 9,510,899 — asserted in ITC Inv. No. 337-TA-1112, Certain Radio Frequency Micro-Needle Dermatological Treatment Devices (complainants Syneron Medical Ltd.; Candela Corp.; Massachusetts General Hospital Corp., filed April 9, 2018; instituted May 9, 2018). This is a different patent number (9,510,899 ≠ 10,190,899) and a different technology field (dermatological RF devices). Not relevant.
2. "109170899" appearing as a Unified Patents caselist filter value — this is a different patent identifier (format suggests a 9-digit run, not US 10,190,899), from an unrelated case list. Not relevant.

Case summary table

Caveats and confidence

• Confidence: moderate-to-high that no litigation exists, but not absolute. Unified Patents' caselist portal and PACER require interactive/paid access that my searches could not fully query for the exact number; I relied on indexed search results of those databases.
• This patent issued only on 2019-01-29 (from a 2013 priority application) and is owned by Astemo, Ltd. (successor to Hitachi Astemo / Hitachi Automotive Systems). A component supplier's sensor patent of this type is typically asserted, if at all, in supplier-vs-supplier disputes — no such dispute is documented in the sources I could reach.
• If a definitive negative is required (e.g., for a freedom-to-operate or diligence opinion), the authoritative checks would be: a PACER "Nature of Suit 830" search across all districts, the USITC EDIS database, and the PTAB Patent Trial & Appeal Board case docket, each queried on the literal number 10,190,899. I could not complete those direct database queries within this session, so I am flagging that as an unresolved evidentiary gap rather than asserting categorical non-existence.

No litigation is known for US 10,190,899.

Proceedings overview

Total AIA trial proceedings on US 10,190,899: 0. The USPTO Open Data Portal (ODP) structured record supplied in this prompt returns no AIA trial proceedings for this patent, and my independent web searches (PTAB dockets, PTAB E2E-adjacent public filings, CourtListener/CAFC material, and general queries pairing "10,190,899" / "10190899" with IPR/PGR/CBM) surfaced no petition, institution decision, Final Written Decision, termination, or appeal naming this patent. Bottom line for a defendant: the patent reaches you un-challenged and un-narrowed — all six claims are live and untested at the Board. There is no adverse FWD to borrow, no petitioner estoppel to shelter behind, and no cancellation order to point at. If your demand letter cites claim 1, the troll (or, far more likely here, Astemo or a supplier of Astemo's) is citing a claim that nobody has ever tried to invalidate at the PTAB.

No proceedings to enumerate

Because the count is zero, the per-proceeding template (type / filed / panel / grounds / institution / FWD / settlement / appeal / defensive value) has no entries. Rather than pad it with invented docket numbers, here is exactly what was checked and what came back.

What the structured source says. The canonical block is unambiguous: "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest." That is the authoritative list per the task instructions, and it is empty.

Independent verification attempted (2026-09-29):

Explicit non-findings (flagged as such, not as proof of absence):

• I found no district-court litigation asserting US 10,190,899, and therefore no § 315(b) real-party-in-interest trail that would ordinarily accompany an IPR. My search was not a docket-complete litigation search; treat this as "no results found," not a certified negative.
• I found no ex parte reexamination, no supplemental examination, and no reissue on this patent. Those are not AIA trials and would not appear in the ODP trial block anyway, but they are also absent from what I could see.
• ODP ingest lag means a very recently filed petition (filed within roughly the last several weeks) could theoretically exist without appearing in the structured data. Nothing in open web sources suggests one.

Timing note that matters: the patent issued 2019-01-29. The PGR window closed 2019-10-29 (35 U.S.C. § 321(c), nine months post-grant), and CBM review is unavailable both because this is a thermal-flow-sensor patent (not a "financial product or service") and because the CBM program sunset on 2020-09-16. IPR is the only AIA vehicle still open to a challenger today, and it is fully open — no window has run and no petitioner has consumed the patent.

Strategic summary

Claim status of US 10,190,899. All claims are UNTESTED at the PTAB. Nothing is canceled and nothing is sustained — the distinction is meaningless here. Claims 1–6 (claim 1 independent; claims 2–6 depending from it) stand exactly as issued on 2019-01-29, and the patent is in force with an adjusted expiration of 2034-09-28. For a defendant, this means the full claim set is fair game and the full claim set is against you. There is no narrowing claim amendment, no certificate, and no FWD-imposed construction to constrain the patent owner's infringement theory — you will be litigating claim 1's "correction circuit … configured to cut off a part of a mountain part or a valley part constituting a peak value … when the peak value of the waveform exceeds the arbitrary predetermined value" on a blank slate (see the claim-1 plain-language summary already generated earlier in this analysis; I don't repeat it here).

Estoppel landscape. There is no § 315(e)(2) estoppel of any kind because there has never been a petitioner — no IPR, no PGR, no CBM. That cuts both ways:

• Good for you: no prior-art ground has been "used up" by anyone. Every reference, every combination, every § 102/§ 103 theory is available. The patent's own citation list gives a starting corpus — including JPH1183584A (Hitachi, the PTL 1 backward-flow/error-correction reference the specification itself distinguishes), US 5,635,635 (Unisia Jecs), US 2005/0109085 A1 (Mitsubishi Denki, heat-sensitive flow meter and fuel controller), US 7,177,770 B1 (Delphi, mass air flow metering), and US 2010/0095761 A1 (Mitsubishi Electric) — but nothing has been adjudicated on any of them.
• Watch out for: if you are one of multiple defendants, the first defendant to file an IPR will lock in § 315(e)(2) estoppel for itself and its privies, including real parties in interest and those in privity. Because there is no existing petitioner to hide behind, you are the first mover — whoever files first owns the estoppel risk and the § 315(b) clock simultaneously. If several defendants share a supplier, a joint defense group or a single-filer strategy should be evaluated before anyone files, because a joined or follow-on petition does not reset the first filer's estoppel.

Pattern signals. None yet — and that absence is itself informative. This is a captive OEM component patent owned by Astemo, Ltd. (successor to Hitachi Automotive Systems, Ltd. via Hitachi Astemo, Ltd., name change effective 2021-01-01), covering a MEMS thermal air-flow sensor of the kind designed into engine intake systems. Patents in this space are typically cross-licensed among Bosch / Denso / Hitachi-Astemo / Continental under supply agreements rather than litigated, which is a plausible — but unverified — explanation for the zero-IPR record. There is no defensive aggregator (Unified Patents or similar) in the chain that I could find, no serial petitioner, and no history of the patent owner appealing PTAB outcomes (there are none to appeal). Also note the family footprint — JP6106559B2, EP3043155B1, CN105518420B, WO2015029459A1 — none of which showed AIA-side activity relevant here.

One inference worth carrying into your analysis, flagged as inference. Because the claims were granted on a 2014-filed, post-AIA application (PCT filed 2014-02-03, after the 2013-03-16 first-inventor-to-file cutoff), you are dealing with an AIA patent — but that does not expand your options: PGR is time-barred and CBM is unavailable, so IPR is your only PTAB lane, and it is unencumbered. The validity question therefore turns entirely on whether the clamp-correction limitation of claim 1 was genuinely novel over the cited flow-meter art, and on whether a Section 112 written-description/enablement attack on the "arbitrary predetermined value" and the max/min-derived clamp value is worth taking — § 112 grounds are not available in IPR (35 U.S.C. § 311(b) limits IPR to § 102/§ 103 on patents and printed publications), so a § 112 theory would have to be litigated in district court or, if you were still inside the window, in a PGR — and you are not.

Recommended next steps

1. There is no FWD to cite, and no disposition to quote. Because the patent has never been before the Board, do not represent to a court, a client, or opposing counsel that any claim of US 10,190,899 has been canceled, narrowed, or held unpatentable. Any such statement would be unsupportable. The correct defensive posture to communicate internally is: "un-challenged patent, full claim set live, no estoppel, IPR available and pending nothing."

2. Run a proper § 315(b) pre-filing check immediately. Because no petitioner has gone first, a single timely third-party IPR filing (or a coordinated filing) is the highest-leverage move available. Confirm (a) no complaint against you or a privy has been served more than one year ago, and (b) no one in your supply chain has already filed a civil action challenging validity (§ 315(a)(1) bar). Neither of these could be confirmed from public sources because no proceeding exists to reveal it.

3. Assume you are the first mover and plan for estoppel. With no prior petitioner, whichever party files first absorbs full § 315(e)(2) estoppel for its privity group. Coordinate with co-defendants and your sensor supplier before any filing, and decide deliberately whether to file jointly, to designate a real party in interest, or to let a supplier with indemnity exposure take the lead.

4. Seed the prior-art search behind the specification's own admissions. Use the specification's characterization of PTL 1 (JP 11-83584 A) as the rich-error reference, plus US 5,635,635, US 2005/0109085 A1, US 7,177,770 B1, and US 2010/0095761 A1 as the initial corpus, and target claim 1's clamp-correction feature — the "cut off a part of a mountain part or a valley part … when the peak value exceeds the arbitrary predetermined value" element — which is the only thing standing between this patent and a conventional max/min-tracking flow-meter correction circuit.

5. Verify claim status one more time at filing. Patent Center: https://patentcenter.uspto.gov — confirm no post-grant proceeding, terminal disclaimer, or reissue has appeared since this analysis. PTAB E2E for trial status: https://ptacts.uspto.gov/ptabweb/ — currently shows nothing for this patent. If a proceeding does appear between now and your filing, re-run this analysis, because a pending IPR changes your § 315(a)/(b) posture and your ability to file at all.

Caveat on sourcing. Everything above is grounded in (a) the authoritative Google Patents copy of US10190899B2 fetched 2026-09-24, and (b) the structured ODP block stating zero AIA trials. My web searches produced no countervailing evidence and I am not treating search absence as affirmative proof that no petition exists — only that none exists in the sources available to me as of 2026-09-29. Where I have stated a hypothesis (e.g., cross-licensing explaining the zero-IPR record), I have labeled it as an inference, not a finding.

I'll research the assignment record for this patent across the USPTO Assignment Center and related sources.

Let me check the specific assignment records and the assignee's history.

Good progress. Let me search for any further name-change recordings and litigation history.

Ownership & NPE Analysis — US 10,190,899 B2 ("Thermal flow sensor")

Sourcing note: assignmentcenter.uspto.gov is a JavaScript-driven search application and I could not retrieve rendered record pages for this patent directly. The assignment links below are taken from (a) the USPTO-derived legal-events data carried on the Google Patents copy of US 10,190,899B2 (the authoritative full text supplied for this analysis, fetched 2026-09-24) and (b) a mirrored USPTO assignment record on PlainSite. Reel/frame numbers, execution dates and recording dates below are quoted as they appear in those records. Where a field (notably correspondent) was not exposed in the retrieved record, I say so rather than fill it in. Verification link: https://assignmentcenter.uspto.gov/ (search by patent number 10190899).

Inventors

• Employer attribution rests on the assignment itself: all four inventors conveyed to HITACHI AUTOMOTIVE SYSTEMS, LTD. (reel 037837/0888), and the PCT applicant of record was the same company.
• Unusual-pattern check: negative. There is no evidence any inventor departed the assignee within 12 months of filing. The execution date on the inventors' assignment (2016-02-12) sits ~29 months after the 2013-09-02 JP priority filing and ~10 days before recording (2016-02-26) — this is the ordinary timeline for perfecting title at US national-stage entry of a PCT case (PCT filed 2014-02-03), not a fire-sale precursor.
• No non-inventor assignors appear anywhere in the chain.

Original assignee

Hitachi Automotive Systems, Ltd. (2520 Takaba, Hitachinaka-shi, Ibaraki, Japan) — the entity named on the issued patent.

• Line of business: Tier-1 automotive components (engine management, powertrain/ICE, chassis, and physical-quantity sensors, including thermal air-flow meters for engine intake). The patent's own specification describes the device as an intake-air flow sensor for a vehicle internal-combustion engine, i.e. a product line this assignee actually built and sold.
• Shipped a product embodying the claims? Yes, in substance — the specification is written around a production MEMS-diaphragm thermal flow sensor with a bypass-route housing (FIGS. 2 and 4; sensing element 20, support substrate 25, housing member 30, bypass route 31, air intake duct 40), i.e. an engineered part, not a paper asset.
• Current status: Operating, but no longer under that name. It became Hitachi Astemo, Ltd. effective 2021-01-01 via a management integration with Keihin Corporation, Showa Corporation and Nissin Kogyo Co., Ltd.; the Japanese corporate registry (法人番号 7050001007842) records a name change on 2021-01-04 and absorption mergers of Keihin, Showa and Nissin Kogyo on the same date. It became Astemo, Ltd. effective 2025-04-01 (announced 2025-02-20), with headquarters relocated from Hitachinaka, Ibaraki to 2-1 Otemachi 2-chome, Chiyoda-ku, Tokyo. Ownership today: Honda 40% / Hitachi 40% / JIC Capital 20% (since October 2023); a Honda filing of 2025-12-16 discloses Honda acquiring an additional 21% from Hitachi to hold 61% and consolidate Astemo. Not dissolved and not in bankruptcy.

Assignment timeline

Only two assignment records are attributed to this patent, plus (see flag) a third recording that is implied by the current assignee name:

• 2016-02-12 (executed) / recorded 2016-02-26 — Reel 037837/0888

  • Conveyance: Assignment (assignment of assignors' interest)
  • Assignor: DOI, RYOSUKE; TASHIRO, SHINOBU; SUZUKI, KAZUNORI; MATSUMOTO, MASAHIRO (all four named inventors)
  • Assignee: Hitachi Automotive Systems, Ltd. (Japan)
  • Correspondent: not exposed in the retrieved record — I cannot name one without fabricating it.
  • Context: routine inventor-to-employer title perfection at US national-stage entry of PCT/JP2014/052381; no change in beneficial ownership, no third party introduced.

• 2021-01-01 (executed) / recorded 2021-03-25 — Reel 056299/0447

  • Conveyance: Change of Name
  • Assignor: Hitachi Automotive Systems, Ltd.
  • Assignee: Hitachi Astemo, Ltd. (2520 Takaba, Hitachinaka-shi, Ibaraki, JP)
  • Correspondent: not exposed in the retrieved record. (For context only: a separate, portfolio-wide Hitachi Automotive Systems → Hitachi Astemo change-of-name recording — reel 58481/935, executed 2021-01-01, recorded 2021-11-30 — carries correspondent Crowell & Moring LLP, P.O. Box 14300, Washington, DC 20044-4300. I could not confirm US 10,190,899 appears on reel 58481/935; the reel-58481 patent list I retrieved is long and the entries visible do not obviously include it. Treat 58481/935 as a parallel/duplicate recording of the same 2021 name change covering other patents, not as a confirmed link in this patent's chain.)
  • Context: internal corporate reorganization / name change only — the January 2021 Hitachi Automotive Systems + Keihin + Showa + Nissin Kogyo integration. No consideration, no new owner.

• 2025-04-01 (effective) — reel/frame not retrieved

  • Conveyance: Change of Name (inferred from the record, not confirmed by me)
  • Assignor: Hitachi Astemo, Ltd. → Assignee: Astemo, Ltd. (2-1 Otemachi 2-chome, Chiyoda-ku, Tokyo)
  • Correspondent: unknown / not retrieved.
  • Context: change of name and registered address only (Japanese registry entries dated 2025-04-01 and 2025-04-11).
  • Why I list it: the Google Patents record for this patent displays current assignee "Astemo Ltd," yet the last assignment it attributes to the patent is the 2021 change of name to Hitachi Astemo, Ltd. Those two statements are inconsistent unless the 2025 rename has also been recorded against this patent. I could not retrieve the corresponding US reel/frame, so I flag it as unconfirmed. The 2025 rename is corroborated globally — e.g. Vietnam IP office recordals dated 2025-12-31 substituting "Astemo, Ltd." with the new Chiyoda address, and UK IPO recordals of the 2021 merger-based transfers.

If the Assignment Center is searched and returns only the two records above, that is still a benign finding here: both are internal to the same Japanese corporate family and neither introduces a new beneficial owner.

Timeline diagram

timeline
    title Ownership of US 10190899
    2013 : Priority filing in Japan
    2014 : PCT application filed
    2016 : Inventors assign to Hitachi Automotive Systems
    2019 : US patent issued
    2021 : Change of name to Hitachi Astemo Ltd
    2025 : Change of name to Astemo Ltd

NPE / troll-pattern signals

1. Shell-entity transfer — not present. Every assignee is a Japanese operating corporation in a single corporate lineage. No "IP/Licensing/Ventures/Holdings" suffix, no Delaware or Texas single-member LLC, no registered-agent address. The only post-issuance transfers (reels 056299/0447 and the unretrieved 2025 recording) are Change-of-Name conveyances with no consideration.

2. Known asserter in the chain — not present. Neither current nor prior assignees (Hitachi Automotive Systems, Ltd.; Hitachi Astemo, Ltd.; Astemo, Ltd.) appear on any NPE list. They are an 80,000-employee Tier-1 supplier with manufacturing operations across the Americas, Asia, China and Europe.

3. Repeat correspondent across the chain — unclear, leaning not present. The only correspondent name I obtained anywhere near this chain is Crowell & Moring LLP on reel 58481/935 (a Hitachi Automotive Systems → Hitachi Astemo change-of-name batch recorded 2021-11-30). A single large general-practice IP firm handling a portfolio-wide internal name change for a Japanese operating company is not the signal this test is looking for; the signal requires recurrence across chain links on the same patent, and the correspondents for reels 037837/0888 and 056299/0447 were not exposed in the records I retrieved. I cannot call this present or absent on the evidence.

4. Cascading transfers — not present. The two dated events are 5 years apart (2021-01-01 and the 2025 rename), and both are renames, not sales. No chained LLCs, no shared principal addresses, no sub-24-month sequence.

5. Pre-litigation transfer — not present. I found no infringement suit naming US 10,190,899 (consistent with the earlier section's "no results found" caveat). There is therefore no suit against which a 6-month pre-suit transfer window could even be measured.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding involving Hitachi Automotive Systems, Hitachi Astemo or Astemo was located; the 2021 and 2025 events are a management integration and a rebrand, both solvent transactions.

7. Privateering — not present. No operating company → licensing-vehicle transfer, and no SEC/EFF/Patent Progress coverage of Astemo asserting through an NPE.

8. Defensive aggregator (anti-NPE) — not present. The chain terminates at Astemo, Ltd., an operating Tier-1 supplier, not at RPX, AST, LOT Network, Unified Patents or OIN. (Signature worth noting anyway: the office action in the counterpart European case, EP 14840664.8, drew a Rule 114(2) EPC communication dated 2018-08-15 and an extended search report dated 2017-04-12 — ordinary prosecution friction, not an ownership signal.)

Verdict

Operating-company assertion — with the express qualification that the "assertion" half of that label is not evidenced: I found no litigation involving this patent, and the chain contains zero NPE indicia.

Justification in 2–3 sentences: The only recorded links are the inventors' assignment to Hitachi Automotive Systems, Ltd. (executed 2016-02-12, recorded 2016-02-26, reel 037837/0888) and a Change-of-Name conveyance to Hitachi Astemo, Ltd. (executed 2021-01-01, recorded 2021-03-25, reel 056299/0447), with a further rename to Astemo, Ltd. effective 2025-04-01 that I could not tie to a reel/frame. Every link is internal to one Japanese operating corporation; no shell entity, no known asserter, no cascading LLC transfers, no bankruptcy, no privateering, no defensive-aggregator terminus. Because none of the five menu labels describes "operating company owns and works its own patents, asserting nothing," I select the operating-company label rather than an NPE label — and flag that the absence of any assertion means the alternative reading, Insufficient data (only original/nominal assignments), is arguably a fairer description of the assertion question.

Verification: https://assignmentcenter.uspto.gov/ · record mirror consulted: https://www.plainsite.org/patents/assignment.html?id=[10965070](/patent/10965070) · patent record: https://patents.google.com/patent/US10190899/en

I'll search for this specific patent number and its cited prior art references.

The patent is confirmed. Now let me research the key cited prior-art references, starting with those most likely relevant to the clamping/correction subject matter.

Prior-Art Analysis — US 10,190,899 B2 (Thermal Flow Sensor)

1. USPTO database confirmation of the specific number

A search for 10,190,899 returns the granted patent US 10,190,899 B2, "Thermal flow sensor," application US 14/914,709 (national stage of PCT/JP2014/052381), published pre-grant as US 2016/0209255 A1 (2016-07-21), granted 2019-01-29, priority 2013-09-02 (JP 2013-180852), original assignee Hitachi Automotive Systems, Ltd. (now Astemo, Ltd.). This matches the authoritative document supplied.

Identifiers I deliberately excluded as "similar but not the same" (per the strict rule):

• USD 426,534 S — design patent whose application number is 29/101,908, not US 10,190,899.
• EP 10190899.4 — a European application number (Nestlé/SSI-Schäfer family), not US 10,190,899.

Neither is the patent at issue.

Sourcing note: my search returned the granted patent text but I was unable to pull full front-page/USPTO "References Cited" images directly; the citation list below is taken from the authoritative Google Patents text you supplied (the "Citations (13)" / "Patent Citations (17)" tables), supplemented by web snippets for a few references. Descriptions marked "(patent text only)" are inferred from the title and the patent's own discussion, not from reading the reference's specification.

2. Cited prior-art references, with dates and descriptions

All references predate the 2013-09-02 priority date except US 2015/0141774 A1 (see § 102(e) note below). References the specification itself designates as PTL 1 and PTL 2 are flagged.

Additional references appearing only in the longer "Patent Citations (17)" table (e.g., the JP/EP/US family duplicates of the Hitachi 1997–1998 filings) are cumulative to items 5–7 and add nothing new for § 102 purposes.

3. § 102 anticipation assessment, claim by claim

Claim 1 requires, in combination: (a) a MEMS diaphragm air-flow-rate detection element (thin-film portion in a semiconductor substrate); (b) a heat-generating resistor on the diaphragm; (c) upstream/downstream temperature-measuring resistors; and (d) a correction circuit including a maximum-value detection circuit and a minimum-value detection circuit that cuts off (clamps) a part of a mountain/valley of the waveform by outputting an arbitrary predetermined value — determined from the detected max and min — when the peak exceeds that value.

• No single cited reference discloses all of (a)–(d). The MEMS-diaphragm structure (a)–(c) is taught by JP 2012-032247 A / US 2013/0199280 A1 (item 1) and the two-location heater/thermoresistor architecture by US 5,635,635 A (item 3). The max/min detection + correction logic (d, partly) is taught by US 7,177,770 B1 (item 11). But the specific "clamp by outputting an arbitrary predetermined value determined from max/min" limitation is not disclosed by any cited reference on the record I could retrieve. Claim 1 is therefore not anticipated by any single cited reference.
• US 7,177,770 B1 is the most dangerous reference, but it applies a correction factor to the signal (an additive/multiplicative adjustment) rather than truncating the waveform peak at a clamp value; the patent itself distinguishes clamping from additive correction (see its FIGS. 16A/16B/17 discussion). It also lacks the diaphragm element. At most it supports a § 103 obviousness combination with item 1.

Claims 2–3 (predetermined value from amplitude value + average flow rate, derived from the max/min-detection outputs):

• US 7,177,770 B1 (item 11) discloses computing max/min peaks and deriving thresholds/amplitude and using them with the signal to select a correction — this is the closest prior art to the parameter-selection concept of claims 2–3, though not the exact average-flow-rate + amplitude pair.
• US 2010/0095761 A1 (item 12) explicitly defines the amplitude ratio (ΔQ/2Qa, i.e., amplitude vs. average flow rate Qa), which maps directly onto the "amplitude value and average flow rate" parameters of claim 2.

Claim 4 (signals corrected by a response-delay compensation circuit) and claim 5 (frequency response circuit / HPF):

• The patent's own background attributes the response-delay/lean-error problem to the semiconductor sensor of JP 2012-032247 A (item 1); response-delay/frequency characterization is also discussed in US 2010/0095761 A1 (item 12). These are relevant art for a § 103 argument but do not anticipate the claim as a whole.

Claim 6 (linearizing the flow-rate-vs-output signal before the max/min detection):

• I found no cited reference disclosing linearization feeding the max/min detectors. The non-linearity itself is acknowledged as known in the art (it underlies the "lean error" discussion in item 12), but the claimed placement of linearization is not shown.

4. Conclusion — most relevant prior art

1. JP 2012-032247 A (pub. 2012-02-16; family US 2013/0199280 A1) — most relevant for the structural elements of claim 1 (semiconductor thin-film diaphragm sensor with heater and upstream/downstream temperature resistors). This is the patent's own PTL 2.
2. US 7,177,770 B1 (Delphi, issued 2007-02-13) — most relevant for the signal-processing/correction concept (max/min peak detection + correction factor), the closest art to claims 1–3 on the correction side.
3. US 2010/0095761 A1 (Mitsubishi Electric, pub. 2010-04-22) — most relevant background defining the lean/rich (ripple) error and the amplitude-ratio parameter central to claims 2–3.
4. US 5,635,635 A (Unisia Jecs, issued 1997-06-03) — relevant sensor architecture (heater between two thermosensitive resistors, direction detection, averaging).
5. JP 11-083584 A (= PTL 1) and EP 0 902 261 A1 / US 6,672,146 B1 — background on backflow/error correction and thermal-resistor measuring.

Bottom line: On the record available, no cited reference anticipates any claim under § 102 — the claimed "clamp the mountain/valley at an arbitrary predetermined value determined from max/min detection" in combination with a MEMS diaphragm element is not met by a single reference. The strongest § 103 challenge would combine JP 2012-032247 A / US 2013/0199280 A1 (structure) with US 7,177,770 B1 and/or US 2010/0095761 A1 (max/min detection, amplitude-vs-average parameters, ripple-error correction).

Caveats / confidence: I retrieved full text for items 3, 11, and 12 and partial text for the others; descriptions for items 4–10 and 13–16 are based on titles and the patent's own discussion and should be confirmed against the actual specifications before being relied on in a legal filing. All dates above are as listed in the authoritative citation table and Google Patents bibliographic data.

I'll ground this analysis in the actual cited prior art. Let me pull details on the key references from the patent's citation list.

Obviousness Analysis under 35 U.S.C. § 103 — US 10,190,899 B2 ("Thermal flow sensor")

Building on the previously generated summary and claim breakdown above; I do not repeat the bibliographic table, abstract, or the plain-language claim 1 restatement produced there. One minor cross-reference flag: the prior section's caveats concerning litigation and assignee status remain accurate, and nothing in the prior-art record I retrieved contradicts it. (The system prompt and the task prompt give two different "current dates" — 2026‑09‑29 vs. 2026‑04‑26; neither affects the § 103 analysis, since all relevant art predates 2013.)

1. Legal framework applied

The application has an effective filing date of 2014‑02‑03 (PCT/JP2014/052381) with a JP priority of 2013‑09‑02 — i.e., post‑AIA, so AIA § 103 governs. Every reference discussed below published well before the 2013 priority date and is § 102(a)(1)/(a)(2) art.

I apply Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of art; differences; PHOSITA level; secondary considerations) and KSR Int'l v. Teleflex, 550 U.S. 398 (2007) (predictable results; finite number of identified, predictable solutions; design incentives; "obvious to try"), as operationalized in MPEP §§ 2141–2144.

POSITA: a mechanical or electrical engineer (B.S. + ~2–5 years, or equivalent) working on automotive air-flow sensing, familiar with hot-wire/hot-film and MEMS-diaphragm thermal flow sensors, with knowledge of engine intake pulsation ("ripple"), lean/rich ripple error, bypass-route inertial effects, and the sensor-side A/D–DSP–D/A signal chain feeding the engine ECU.

Claim architecture matters here: claim 1 is the only independent claim; claims 2, 3, 4, 5 and 6 all descend from it (2→1, 3→2, 4→3, 5→3, 6→3). Every dependent claim therefore carries the "cut off a part of a mountain part or valley part … when the peak value … exceeds the arbitrary predetermined value" limitation. If claim 1's clamp limitation is shown obvious, the whole claim set falls; if the clamp survives, nothing else in the set is independently patentable on its own.

Claim 1 element breakdown

2. Scope and content of the prior art on the face of the patent

(URLs cited where I actually retrieved text; where I could only verify title/date/assignee, I say so and do not assert disclosure content.)

(a) PTL 2 — JP 2012‑32247 A / JP2012032247A (Hitachi Automotive Systems, 2012‑02‑16), US family member US 2013/0199280 A1 "Thermal Flow Meter." The '899 specification expressly admits this as the quick-response sensor "in which the resistor and the insulating film are provided for the silicon substrate (wafer) through the semiconductor process and the wafer is thinned through the back etching, so that the thermal capacity is suppressed." That is an applicant admission that 1A, 1B and the upstream/downstream resistor pair of 1C are prior art. This admission is dispositive on the structural elements and halves the obviousness inquiry.

(b) PTL 1 — JP 11‑83584 A / JPH1183584A (Hitachi, publ. 1999‑03‑26), "Heating resistor type air flow measuring device, backflow determination method and error correction method." Family confirmed via JP3283800B2 and KR100695982B1 ("Heat resistance resistor air flow measurement device, backflow determination method and error correction method") — https://patents.google.com/patent/JP3283800B2/en. Caveat: I did not retrieve the full specification this session, so I rely on (i) the title, (ii) its family member titles, and (iii) the applicant's own characterization in the '899 background ("the negative error (binarizing phenomenon)… as explicitly described in PTL 1"; and the bobbin-type sensor whose output "changes in the positive direction regardless of whether the flow is the forward flow or the backward flow," producing the rich error). Stated narrowly: PTL 1 discloses backflow determination plus an error-correction method for a heating-resistor air flow meter, and identifies the response-delay-induced lean ("binarizing") error.

(c) US 5,635,635 A (Unisia Jecs, 1997‑06‑03) — https://patents.google.com/patent/[US5635635A](/patent/US5635635A)/en; full text also at http://www.everypatent.com/comp/pat5635635.html. Discloses two thermo-sensitive resistors at upstream and downstream locations with a heating resistor between them, all as films on a single insulating substrate aligned with the flow direction (claims 6, 11, 12, 15, 16, 20, 21); flow-direction detection by comparing the two detection values / the phase of pulsations; and — critically — an averaging/processing circuit that subtracts the reverse-flow component so the computed average is not inflated. It states the problem in the same terms the '899 patent uses: "the computed average Q will be greater than the true average Q, since… the reverse flow component is also detected as being in the normal direction," leading to "a richer mixture, deterioration in exhaust conditions, and poorer fuel consumption." It also discloses comparator, adder, inverter and selector circuits (47–50).

(d) US 5,717,136 (Unisia Jecs, 1998‑02‑10), verified at http://www.everypatent.com/comp/pat5717136.html — first and second temperature-sensitive resistors formed in film upstream/downstream of a heat-generating main resistor; "airflow direction determining means comparing the resistance values"; and "selector means for … inverting said airflow quantity indicative signal" on reverse flow. Not on the face of US 10,190,899 — surfaced in searching; used only as corroborating art.

(e) US 7,177,770 B1 (Delphi Technologies, 2007‑02‑13) — https://patents.google.com/patent/US7177770 and https://www.freepatentsonline.com/[7177770](/patent/7177770).html. This is the most on-point reference for limitations 1D and 1E. Discloses a conventional (uni-directional) hot-film anemometric sensing device plus a custom IC, an on-board microcontroller, and a flow-correction circuit; the signal processor "characterize[s] the input flow signal," determines a flow mode from {non-pulse, pulse, onset of reverse flow, reverse flow}, and selects a flow correction factor based on the determined flow mode and the characterized input signal; the output = the input flow signal adjusted by the flow correction factor. It expressly criticizes conventional filtering ("second-order digital filters… consume substantial amounts of execution time, microprocessor time and computer memory, and [are] not feasible for implementation in a low-cost microcontroller used primarily in an airflow sensing device") and provides a lighter digital filter "capable of handling pulse frequencies observed during pulsing flow conditions, filtering out high frequency noise, while still able to detect onset of reverse flow." The OCR excerpt at https://uspto.report/patent/grant/7,177,770 shows the algorithm is driven by "Minimum Peak," "Reverse Maximum Peak" and like flags — i.e., max/min peak detection of the waveform carried out inside the correction processor of the sensor.

(f) US 2010/0095761 A1 (Mitsubishi Electric, 2010‑04‑22) — https://uspto.report/patent/app/20100095761 and https://www.patentsencyclopedia.com/app/20100095761. Discloses the identical problem statement: "when flow is pulsated due to thermal response delay of a flow rate detector and a non-linear output characteristic, there occurs a lean error that the average flow rate… is smaller than the true average"; and "when pulse flow accompanied by counter flow occurs, the thermal type flow rate measuring device which has no counter flow detecting function detects the pulse flow as forward flow, so that a rich error… larger than the true average occurs." It discusses the known bypass-channel remedy and its limits, and quantifies ripple as standard deviation/average ("output turbulence [%]").

(g) Other cited art (identified by title/date/assignee only — full texts not verified this session, so no disclosure asserted): US 6,672,146 B1 / EP 0 902 261 A1 / EP 0 961 105 A1 (Hitachi — thermal-resistor air flow meters); JP 2006‑058078 A and US 2006/0037390 A1 (Hitachi — thermal air flow meter); US 2005/0109085 A1 (Mitsubishi Denki — heat-sensitive flow meter and fuel controller); US 2012/0291533 A1 and JP 2012‑255768 A (Denso — flow-rate measuring apparatus); JPH07234143A (Unisia Jecs, 1995 — ICE intake air flow rate detection device); US 2011/0257898 A1 (Yamatake — thermal diffusivity measuring system); US 2015/0141774 A1 (Mitsubishi Chemical Holdings — subject information detection; an apparent outlier whose relevance I could not establish).

3. Combinations that render the claims obvious

Combination 1 (primary, two references): JP 2012‑32247 A [PTL 2] + US 7,177,770 B1 [Delphi]

Legal formulation: "The subject matter of claim 1 is obvious over JP 2012‑32247 A in view of US 7,177,770 B1."

Mapping: 1A–1C ← PTL 2 (admitted). 1D ← US 7,177,770 B1 (sensor-side processor + custom IC + microcontroller performing waveform characterization, with peak flags). 1E ← US 7,177,770 B1's flow-correction-factor scheme applied to the characterized pulse/reverse-flow waveform, the choice of clipping the excursion rather than offsetting the whole signal being one of a small, well-known set of signal-conditioning operations.

Motivation (MPEP 2144.01/2144.04): Both references are in the same field (thermal air-flow meters for internal-combustion-engine intake) and address the same defect (pulsation/backflow-induced error in the average flow value). PTL 2 supplies the sensing element the '899 patent itself says is conventional; US 7,177,770 supplies the express teaching that the sensor's own low-cost processor should correct the raw airflow signal based on waveform characterization including reverse-flow onset. A POSITA seeking to improve accuracy of a MEMS meter in a rippling-flow engine would be led directly to the Delphi signal-processing architecture. There is no teaching away, and no new structural interaction is required — the correction circuit is simply added downstream of PTL 2's bridge output, exactly as the '899 specification describes it ("the element output from the sensing element 20 is taken into a correction LSI").

Combination 1A (strongest for the "clamp" limitation itself): US 7,177,770 B1 + US 5,137,001 + PTL 2

I retrieved in this session a reference not on the face of US 10,190,899 that bears directly on limitation 1E: US 5,137,001 (PDF at https://patentimages.storage.googleapis.com/ad/51/71/60a20ab5f1c7f1/US5137001.pdf). Its background states that "the air flow rate A(n), when there occurs a reverse flow of air, shows a fairly large value in comparison with an actual air flow rate in the … low-speed-high-load area," and that the art addressed it by determining "an upper limit value… so that the value of intake air flow rate is clipped"; the patent's own summary provides "a clip means to clip the output of the intake air quantity detecting means at a second value." That is the same remedial concept as the '899 clamp: clip the air-flow-rate signal at an upper bound to remove a reverse-flow-induced positive (rich) error.

Caveat, stated plainly: I have only an OCR snippet of US 5,137,001 (bibliographic details, full claim text of FIG. 5 flowchart, assignee not visible to me in the retrieved text), and it is not among the references cited on the face of US 10,190,899. I therefore present it as an art-discovery lead and a secondary-reference candidate, not as a verified ground of rejection. A full-text pull and an assignee/date check should be done before relying on it.

If verified, the three-way combination is compelling: PTL 2 for 1A–1C; US 7,177,770 B1 for the sensor-side correction circuit and max/min peak characterization (1D); US 5,137,001 for the proposition that clamping/clipping the air-flow signal at a predetermined upper value is a known way to remove reverse-flow-induced positive error (1E) — with the further motivation that US 5,137,001 takes its clip value from stored engine-speed-indexed values, whereas US 7,177,770 derives it from the waveform, so deriving the clamp value from the detected max/min (as claim 1 requires) is a predictable substitution of one known parameter source for another.

Combination 2 (three cited references): US 5,635,635 A + JPH1183584A [PTL 1] + US 7,177,770 B1

• US 5,635,635 A → 1C (upstream/downstream film thermo-sensitive resistors about a heater) and the problem of the reverse-flow-inflated average;
• JPH1183584A [PTL 1] → backflow determination and error correction for a heating-resistor air flow meter;
• US 7,177,770 B1 → implementing the correction in the sensor's own processor using waveform characterization and peak detection (1D).

Motivation: three references from a single, tightly defined art (thermal intake-air flow measurement) addressing a single recognised defect; US 5,635,635 A and PTL 1 come from the same corporate lineage (Unisia Jecs / Hitachi), and the '899 specification itself cites both by number as the starting points. Under KSR, where the art identifies a problem and all elements are known in the field, the combination is obvious absent a teaching away.

Combination 3 (bypass-structure line): US 2010/0095761 A1 + US 7,177,770 B1

US 2010/0095761 A1 supplies the express recognition that (i) lean error has a thermal-response-delay and non-linearity origin, (ii) rich error has a counter-flow origin, and (iii) the bypass remedy's effectiveness "is dependent on the arrangement position of the flow rate detecting element… no attention has been hitherto paid to the point… the effect of reducing the pulsation detection error cannot be sufficiently obtained." That last statement functions as a teaching away from relying on bypass geometry alone and supplies a strong motivation to add electronic correction (US 7,177,770 B1) — precisely the trajectory of the '899 specification's own disclosure.

Combination 4 (MEMS-structure line): US 2006/0037390 A1 or US 6,672,146 B1 / EP 0 961 105 A1 (Hitachi) + US 7,177,770 B1

Equivalent to Combination 1 with a different structural primary reference, useful if PTL 2's disclosure of the diaphragm is contested. (I did not verify these references' texts; this is offered as an alternative formulation, not a completed mapping.)

4. Dependent claims (2–6)

5. Motivation to combine — consolidated rationale

1. Same field, same problem, same solution space (MPEP 2144.01): every reference is from thermal intake-air flow measurement, and each independently identifies the lean/rich ripple-error pair (US 5,635,635 A; US 2010/0095761 A; PTL 1 as characterized in the '899 background; the '899 patent's own background).
2. Analogous art — all are reasonably pertinent to the problem the inventor faced; several are the applicant's own corporate predecessors' work, which is squarely "prior art as a whole."
3. Predictable result, finite solution set (KSR): post-processing the raw air-meter signal to correct a pulsation-mode-dependent error has a small, well-known menu of operations — offset addition, gain scaling, filtering, and clipping/limiting at a bound. Selecting clipping produces, by definition, a change confined to the clipped excursions; the movement of the average toward the true value is an arithmetic consequence, not an unpredictable discovery.
4. Design incentives / regulatory pressure (the '899 background itself): VTC engines generate ripple amplitudes up to ~1000%, and the bypass-inertia remedy fails in the ≥200% region; the market pressure to improve accuracy in exactly that region is the classic KSR "design incentive" fact.
5. Known inadequacy of the alternative — US 2010/0095761 A1 says reliance on bypass arrangement alone is position-dependent and insufficient.

No teaching away identified from clamping, from sensor-side signal processing, or from using waveform maxima/minima as the parameter source. The closest thing to a "teaching away" is US 7,177,770 B1's caution against heavy digital filtering in a low-cost microcontroller — which, if anything, channels a POSITA toward the cheap clamp-and-compare operation the '899 patent claims rather than away from it.

6. Anticipated rebuttals and why they likely fail

7. Bottom line and residual uncertainty

• Claim 1 is the battleground, and its structural elements (1A–1C) are admitted prior art via PTL 2. The correction-circuit architecture (1D) is well taught by US 7,177,770 B1, including peak detection in the sensor's own processor. The clamp limitation (1E) — the crux — is not squarely disclosed by any single reference I was able to read in this session; my confidence rests on the two-/three-reference rationales in Combinations 1 and 1A. Strength of a prima facie § 103 case on claim 1: fair to reasonable, not overwhelming — and it is the kind of case that could be defeated by credible, well-nexused ripple-error test data.
• Claims 2, 3 and 6 are strongly supported (US 7,177,770 B1 + US 2010/0095761 A1 + routine linearization).
• Claims 4 and 5 are the weakest links; a prima facie case needs an additional reference expressly teaching response-delay/HPF compensation in a thermal air-flow meter, which I could not verify among the cited references.
• Highest-value next steps: (i) pull the full text of JPH1183584A and JPH07234143A; (ii) pull the Extended European Search Report (2017‑04‑12) and the Rule 114(2) EPC third-party observation (2018‑08‑15) in EP 14840664.8 — both appear on the face of the US patent and the third-party observation in particular is likely to cite art not on the US face; and (iii) verify US 5,137,001 (full text, assignee, priority), which if confirmed is potentially the most damaging single reference on the clamp limitation because it clips an intake-air-flow-rate value at an upper limit precisely to remove a reverse-flow-induced positive error.

Uncertainty disclosure: Apart from the references for which I quote text and give URLs above (US 5,635,635 A; US 7,177,770 B1; US 2010/0095761 A1; US 5,717,136; the JP3283800B2 family data for JPH1183584A; and the OCR excerpt of US 5,137,001), the disclosure content of the remaining cited references is not verified in this analysis, and I have deliberately not attributed specific teachings to them.