US 10190908

I'll verify the patent details against live sources, including any 2026 litigation activity.

Two useful signals emerged — let me verify the actual granted claim set and check for any litigation involving this patent.

Let me confirm the granted claim set structure (method vs. apparatus) since sources appear to differ.

Scope and sourcing note

I ran the searches you asked for, but I must be transparent about method: I could not directly query USPTO PatentCenter, the USPTO assignment database, or the CAFC docket system as such — those are not directly addressable from my tools. What I actually did was a targeted web search for the exact string 10190908 plus the patent's title and the key claim language, and separately a search of CAFC 2026 decision coverage. Everything below is grounded either in the full patent text you supplied (which I treated as authoritative) or in specific search hits, which I cite. Where the two diverge, I flag it rather than resolve it silently.

1. Bibliographic summary — US 10,190,908 B2

Abstract (verbatim from the patent):

"Disclosed are optical devices and methods of manufacturing optical devices. An optical device can include a substrate; an optical emitter chip affixed to the front surface of the substrate; and an optical sensor chip affixed to the front surface of the substrate. The optical sensor chip can include a main sensor and a reference sensor. The optical device can include an opaque dam separating the main optical sensor and the reference sensor. The optical device can include a first transparent encapsulation block encapsulating the optical emitter chip and the reference optical sensor and a second transparent encapsulation block encapsulating the main optical sensor. The optical device can include an opaque encapsulation material encapsulating the first transparent encapsulation block and the second transparent encapsulation block with a first opening above the main optical sensor and a second opening above the optical emitter chip."

2. ⚠️ Important discrepancy I could not fully resolve: apparatus vs. method claims

This is the single most consequential point, and you should treat it as unresolved:

• The full text you supplied presents three "aspects" in the Summary using apparatus language — "In one aspect, there is an optical device. The optical device includes a substrate…". The provided fetch, however, cuts off mid-sentence in the Detailed Description and does not include the claims section at all. So your authoritative source does not actually state the claims for this patent.
• The Justia page for patent 10190908 (patents.justia.com/patent/10190908) reproduces a claims section whose independent claim 1 reads "A method of forming an optical device, the method comprising: forming a substrate comprising a front surface; …" — i.e., method claims, not apparatus claims. Claim 9 likewise begins "A method of forming an optical device…".

A third independent claim (matching the third Summary aspect — opaque coating covering substantially all of the top surface, with the opaque encapsulation material abutting that coating along the block's top/side edges) was not fully retrieved; I saw dependent claims running at least to claim 15. I therefore cannot tell you the total claim count or whether any apparatus claims exist beyond claim 15.

My read: the substance maps 1:1 onto the three Summary aspects, but the claims appear to be drafted as methods of forming, while the Summary — likely boilerplate carried over from the parent US 9,976,894 — retains device language. I am flagging rather than asserting this, because I could not verify the granted claim set against USPTO full-text.

3. Independent claims in plain language

Claim 1 — "Method of forming an optical device" (core structural concept).
Build a device by: providing a substrate; mounting an emitter chip (rear face on the substrate front surface) and a sensor chip (rear face on the substrate) where the sensor chip carries both a main sensor and a reference sensor on its front surface; forming an opaque dam across the sensor chip's front surface and part of the substrate's front face so that it physically separates the main sensor from the reference sensor; forming a first transparent encapsulation block over the emitter chip and the reference sensor; forming a second transparent encapsulation block over the main sensor; and forming an opaque encapsulation material over both transparent blocks, with a first opening above the main sensor and a second opening above the emitter chip — where that opaque material runs between the two transparent blocks and abuts the opaque dam.

Plain-language effect: two optically isolated light channels in one package. The receiver channel (main sensor) and the emitter channel (emitter + reference sensor) are separated by a hard opaque wall, and the outer opaque overmold seals around them while leaving two windows. Notably, the reference sensor sits on the emitter side of the dam and the second opening does not extend over the reference sensor.

Claim 9 — "Method of forming an optical device" (adds on-block optics and coatings).
Same building blocks as claim 1, plus: the first transparent block is formed with a first passive optical element (e.g., a lens) over the emitter chip, and the second transparent block with a second passive optical element over the main sensor; a first opaque coating material is formed on the first block around the first passive optical element, and a second opaque coating material on the second block around the second passive optical element; the opaque encapsulation material then defines a first opening aligned with the first passive optical element and a second opening aligned with the second optical element. The "abuts the opaque dam / extends between the blocks" limitation is retained.

Plain-language effect: the same isolation scheme, but the on-block lens apertures are defined by a precisely patterned opaque coating (the spec describes a photostructurable resist/photoresist developed by laser direct imaging or mask), so the opaque overmold opening doesn't have to define the aperture edge.

Third independent claim (not fully retrieved; likely claim 16 or later).
Per the Summary's third aspect and the description at FIGS. 7–9 / 12A–12F: the first and second opaque coatings cover substantially all of the top surface of their respective transparent blocks, and the opaque encapsulation material encapsulates the plurality of side surfaces of both blocks, abuts each opaque coating along the one or more edges between the top surface and the side surfaces, while still extending between the blocks and abutting the opaque dam.

Plain-language effect: claims the "wrap-around" version where the top is fully blacked out except the apertures and the sidewalls are overmolded in contact with the top coating to close the light-tight seam at the block corners.

Dependent-claim subject matter common to all three (claims 2–8, 10–15):

• the dam defines a channel into which the opaque encapsulation material extends (interlocking), or vice versa;
• the dam is a first opaque epoxy and the encapsulation material a second opaque epoxy, with different viscosities (spec: the higher-viscosity dam material prevents bleed onto the sensitive main/reference sensors during dispense);
• at least one trench formed in the substrate that the opaque encapsulation material fills;
• the dam is thicker over the substrate than over the sensor chip (spec: achieved by slowing/stopping the dispenser over the substrate).

4. Litigation / 2026 CAFC docket check

No results. My searches for 10190908 in connection with litigation, and my review of 2026 CAFC coverage (e.g., the Federal Circuit opinions aggregated by WilmerHale, Gibson Dunn, Oliff, and the CAFC's own posted opinions), returned nothing identifying US 10,190,908 as the subject of any CAFC appeal, IPR, or district court action. It is possible the patent is asserted under a different family member (the parent US 9,976,894 B2, or the sibling applications/family members), which would not surface on a search keyed to 10190908.

Caveat: absence of search hits is not proof of absence. I did not have direct docket-system access. A USPTO PatentCenter "Patent Litigation" view or a PACER/Docket Navigator query on the patent number and on the assignee entities (Heptagon Micro Optics, AMS Sensors Singapore, ams‑OSRAM Asia Pacific) would be the authoritative check.

5. ⚠️ Identifier-collision warning (literal interpretation)

One search hit is a false positive you should not conflate with this patent. An EPO Global Patent Index entry for EP 2375428 A3 lists a priority document rendered as "US 10190908 P 20081001" — that is U.S. provisional application 61/090,908 filed 2008‑10‑01 (a nanoelectromechanical tunneling switch), a completely unrelated document. The digit string 10190908 appears in both, but per your instruction I have not auto-corrected or conflated them; they are different identifiers (61/090,908 vs. 10,190,908).

6. Confidence statement

• High confidence: patent number, title, filing/issue dates, priority chain, inventors, assignee chain, abstract, classifications, and the subject matter described in the specification (all from your authoritative full text plus the Google Patents bibliographic record).
• Moderate confidence: that claim 1 is a method claim as reproduced by Justia — the claim text itself is explicit, but I could not independently verify it against USPTO full-text, and it sits in tension with the apparatus-language Summary.
• Low/unknown: the total number of claims; the exact wording of the third independent claim; whether any apparatus claims exist; and any litigation status.

I'll search for litigation involving US patent 10190908. Let me run several targeted searches.

Initial searches surfaced no litigation for this patent. Let me dig deeper into litigation-specific databases and the assignee's enforcement history.

Answer: No litigation found for US Patent No. 10,190,908

I ran targeted searches across general web sources, litigation-oriented pages, PTAB petition filings, and patent databases. I found no record of any litigation (district court, ITC, or Federal Circuit appeal) involving U.S. Patent No. 10,190,908. I also found no PTAB post-grant proceedings (IPR/PGR) asserting it.

What the patent is

• U.S. Patent No. 10,190,908 B2, titled "Optical devices and methods of making same" (optical emitter + main/reference sensor packaging with an opaque dam and transparent encapsulation blocks).
• Application No. 15/984,278; filed May 18, 2018 as a continuation of Ser. No. 15/353,397; priority to Nov. 18, 2015; granted Jan. 29, 2019.
• Original assignee: Heptagon Micro Optics Pte Ltd → ams Sensors Singapore Pte. Ltd. → current ams-Osram Asia Pacific Pte. Ltd. (per Google Patents assignment history).
• Status: Active; anticipated expiration Nov. 16, 2036.

Case list

No plaintiff, defendant, court, case number, filing date, or outcome can be reported, because no case was located.

Near-misses I encountered (explicitly NOT this patent)

I want to flag these so they aren't mistaken for the target:

1. U.S. Patent No. 9,116,908 — a different patent (Realtime Data LLC data-compression family, held § 101-ineligible in Realtime Data LLC v. Array Networks Inc., Fed. Cir. No. 2021-2251, Aug. 2, 2023). The "908 patent" shorthand in that PTAB petition and litigation refers to 9,116,908, not 10,190,908. Do not conflate them.
2. "US 10190908 P … 20081001" appearing in an EPO Global Patent Index entry (EP 2375428) — this is a priority/provisional application number, not a patent assertion.
3. "10190908" in a Brazilian (JUCISRS) corporate registration list and in a student roster — unrelated identifier collisions, not patents or cases.
4. Other recent patent suits in my search results (e.g., Taction Technology v. Apple, patents 10,659,885 and 10,820,117; Masimo v. Apple, the '776 patent) involve entirely different patent numbers and do not concern 10,190,908.

Important caveats and limitations

• I could not directly query the primary litigation databases you named. I was unable to successfully load results from the Unified Patents litigation portal (https://portal.unifiedpatents.com/litigation/caselist), CourtListener/RECAP, PACER, or the CAFC docket search within this session — searches to those targeted sources returned no attributable results for this patent number.
• No result in any search corroborated a filing, so I am reporting absence of evidence, not proof of non-existence.
• Public litigation databases index by asserted patent number; a patent that has never been asserted (or only appears in confidential/unindexed matters, or in a complaint not yet docketed) may legitimately show no results.

Recommended verification steps before relying on this

1. Search CourtListener/RECAP and PACER (Dockets → "Party" and full-text) for "10,190,908" and for the assignee names Heptagon Micro Optics, ams Sensors Singapore, and ams-Osram Asia Pacific.
2. Search the Unified Patents litigation portal caselist by patent number.
3. Check the USPTO PatentCenter "Litigation" tab and the PTAB case search for IPR/PGR involving the patent or its family (Ser. No. 15/353,397 / parent filings).
4. Check ITC EDIS for Section 337 investigations naming the patent or assignee.

If you'd like, I can attempt a narrower follow-up (e.g., on the assignee's enforcement history specifically) if tool access allows.

Proceedings overview

Total AIA trial proceedings on file for US 10,190,908: 0. The USPTO Open Data Portal structured block supplied for this matter returns no IPR, PGR, or CBM proceedings, and targeted web searches surfaced none either (all apparent "hits" for the string were references to unrelated patents — e.g., IPR2023-00086 on Neo Wireless' U.S. 10,833,908, IPR2024-00006 on U.S. 8,588,033, IPR2025-00373 on U.S. 8,717,101). Breakdown by status: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. Bottom line for a defendant: there is no PTAB record to lean on — neither a cancellation you can free-ride on, nor an institution denial or FWD survivability holding that shows the claims have already been tested. Every claim of the '908 patent is untested at the Board, and whatever validity defense you want has to be built from scratch.

No proceedings to report

There is no proceeding entry to render in the required per-trial format. Per the operating rules, I am not going to manufacture a docket number, petitioner name, panel, or disposition to fill a template. The correct output is the null result, stated plainly.

What I checked and what it does and does not establish:

• USPTO ODP / PTAB structured data (canonical source per the prompt): zero AIA trials on this patent as of the most recent ingest.
• Web search for IPR/PGR/CBM activity naming 10,190,908 or its parent: nothing relevant returned. Confidence that no instituted trial exists: high. Confidence that no petition was ever filed and denied or terminated (which can be harder to surface without direct PTAB E2E access): moderate — a terminated-before-institution petition would still normally appear in ODP, so the null result is likely accurate, but I cannot independently rule it out.
• I did not confirm district court litigation asserting this patent, and I did not exhaustively search the parent, U.S. 9,976,894. Absence of IPRs on either is consistent with the patent family simply not having been asserted in a campaign that provokes IPR filings.

Family context relevant to any future challenge (from the patent's own face): US 10,190,908 is a continuation of application Ser. No. 15/353,397, filed 2016-11-16, which issued as U.S. 9,976,894 B2; the '908 patent claims priority back to provisional 62/257,198 (filed 2015-11-18). Original assignee Heptagon Micro Optics Pte. Ltd., now held by ams-Osram Asia Pacific Pte. Ltd. (via Heptagon → ams Sensors Singapore Pte. Ltd.). Anticipated expiration 2036-11-16, so the patent has roughly a decade of life left on the statutory term — plenty of runway for a challenge to appear later.

Strategic summary

Claims status. Nothing has been canceled, disclaimed, or adjudicated. The patent issued with a single independent claim per embodiment family plus dependents covering the opaque-dam / dual transparent-encapsulation-block architecture, the interlocking wall-portion-and-dam variants, the trench-in-substrate light-tight interface, the passive-optical-element (lens) variants with opaque coating aprons, and the viscosity-differential dam material. All of that is UNTESTED. No claim is CANCELED and no claim has been held SUSTAINED by the Board. If you are building a non-infringement or invalidity position, you are working from the prosecution history and the specification alone.

Estoppel landscape. There is no § 315(e)(2) estoppel on this patent because there is no final written decision. That cuts both ways:

• For you: no prior petitioner has burned any art. Every ground is available — you are not limited by anyone else's petition, and you have no risk that the Board already considered and rejected a reference you might rely on.
• Against you: a first-filer IPR is expensive and, without an earlier institution decision, the Board has no prior merits analysis to anchor to. Petitioners in analogous campaigns sometimes want a first mover to have won institution before they file.

Pattern signals. None yet. No serial petitioner, no defensive aggregator (no Unified Patents-style filer), no PTAB appeal to the Federal Circuit, no Director Review activity. For a portfolio that appears to have been consolidated under ams-Osram, the absence of any AIA trial activity to date is a meaningful signal in itself: this patent has not been asserted in a way that attracted a well-funded challenger. Patents that are asserted broadly against deep-pocketed defendants reliably draw IPRs within about a year of service. That this one hasn't suggests either a narrow assertion history or a licensing posture rather than litigation.

One caveat worth flagging to the client. ams-Osram acquired the Heptagon portfolio and has been an active enforcer in the consumer-optics and behind-OLED sensing space (its own investor materials describe the behind-OLED proximity-sensing product line). If assertion activity picks up, IPRs on this patent and its siblings will likely follow within 9–12 months. A "no proceedings" answer today is a snapshot, not a permanent condition.

Recommended next steps

• Do not represent to anyone that the claims have been invalidated, narrowed, or sustained. There has been no PTAB adjudication. Any statement that "the PTAB killed these claims" would be false and, in litigation, sanctionable under Rule 11 / § 285.
• Build the invalidity case as a first-gen challenge. Because no art has been exhausted, you have maximum freedom in reference selection. Two features worth targeting given the file history: (a) the "opaque dam integrally formed and having varying thickness," where the more-dam-material-over-substrate configuration is a process artifact of syringe dispensing that prior-art wafer-level molding references may render obvious; and (b) the wall-portion/opaque-dam interlock, which is likely the narrow point of novelty that will control claim construction.
• Check the parent, U.S. 9,976,894, and the sibling filings in the family for FWDs. The '908 patent is a continuation, and the operative claim scope may have shifted between the parent and child. If a sibling in this family has been litigated or challenged, its prosecution and adjudication history is the best available proxy for how the Board would treat the '908 claims. I did not find such activity, but this search was not exhaustive.
• Monitor PTAB E2E and the ODP daily. If you are a defendant of record, calendar a standing check. The statutory clock is fast: institution decision within 6 months of the petition's filing date (§ 314(b)), and final written decision within 12 months of institution (§ 316(a)(11)). If a petition is filed against you, you will know the trial trajectory within roughly 18 months end-to-end.
• Directly verify the null result before relying on it. The most reliable confirmation is a records search on USPTO PTAB E2E (https://e2e.uspto.gov/) and the PTAB Decisions page (https://www.uspto.gov/patents/ptab/decisions) for both US 10,190,908 and US 9,976,894. My search did not have authenticated access to those systems.

If you want, I can take the second pass on the parent patent U.S. 9,976,894 and the PCT/SG2015/050224 family to see whether any related patent in the chain has been through an AIA trial — that is the closest thing to a test of this claim family and would materially change the defensive assessment.

I'll research the assignment record and related context before assembling the report.

Let me try to pin down the recorded assignment details and reel/frame numbers.

Ownership & Assignment Analysis — US 10,190,908 B2

Patent: US 10,190,908 B2 — "Optical devices and methods of making same"
Application: US 15/984,278 (continuation of US 15/353,397)
Priority: 2015-11-18 (provisional 62/257,198)
Granted: 2019-01-29 | Anticipated expiry: 2036-11-16 | Status: Active

Data-access caveat, stated up front. This report is built from the Google Patents legal-events record for US 10,190,908 (authoritative text supplied) plus public corporate records. The tooling available for this task could not retrieve the USPTO Assignment Center abstract-of-title detail (reel/frame, execution dates, and correspondent of record) for every link in this chain. Where I have a reel/frame I say so and flag the source; where I do not, I say "not retrieved" rather than guess. Correspondent-of-record data — the single most valuable tell per the brief — was not retrievable and is marked accordingly. Verify at the Assignment Center link in the Verdict section.

Inventors

Six named inventors on the issued patent:

Pattern notes:

• All six are consistent with being Heptagon employees; the assignee of record is Heptagon Micro Optics Pte. Ltd. There is no evidence of an inventor departure cascade. To the contrary, Rudmann and Gubser recur on later Heptagon / ams Sensors Singapore filings in the same technology area, which is the opposite of a fire-sale tell.
• The parent application (US 15/353,397) was filed 2016-11-16, i.e. three weeks after ams AG publicly announced the Heptagon acquisition (2016-10-24) and two months before it closed (2017-01-24). The continuation at issue here was filed 2018-05-18, roughly four months after closing. This timing is a normal prosecution artifact of a corporate acquisition, not a distress signal.
• I could not determine individual inventor city/country of residence with confidence from the sources retrieved; the Singapore/Switzerland split above is inference from naming and from the corporate footprint, and is flagged as such.

Original assignee

Heptagon Micro Optics Pte. Ltd. (Singapore) — the entity named on the issued patent.

• Line of business: Design and high-volume manufacture of wafer-level micro-optics and optical packaging — smartphone camera flash lenses, miniature optical/optoelectronic sensors, gesture/distance-sensing modules. Singapore UEN 200704497W, incorporated 2007-03-19, SSIC 26709 "Manufacture of optical instruments and photographic equipment NEC — manufacture of micro-optics."
• Did it ship a product embodying the claims? Yes — Heptagon was a volume manufacturer and reported supplier of micro-optics into consumer handsets (e.g. the Apple relationship widely reported in trade press), and this patent's claims are directed at the packaging of exactly those proximity/gesture optical modules. This is an operating company, not a holding shell.
• Status: Operating, and absorbed. ams AG (SIX: AMS) announced the acquisition of 100% of Heptagon on 2016-10-24 and completed it on 2017-01-24 (base consideration ≈ USD 570M, up to ≈ USD 285M earn-out; sellers took ≈19.5–20% of ams). Heptagon has since been renamed twice inside the ams/ams-OSRAM group (see timeline): Heptagon Micro Optics Pte. Ltd. → ams Sensors Singapore Pte. Ltd. → ams-OSRAM Asia Pacific Pte. Ltd. The entity remained live (ACRA status: Live Company). In September 2024, ams OSRAM sold optical-component R&D/production assets and related IP (Singapore and Switzerland) to Focuslight Technologies Inc. (SSE STAR: 688167), and the Heptagon brand was relaunched under Focuslight; however, no assignment of this patent to Focuslight was found on the record, and the Google Patents "current assignee" still reads ams-OSRAM Asia Pacific Pte Ltd.

Assignment timeline

Three recorded instruments appear on the legal-events record for this patent. Chronology and details:

1. Effective on/around 2018-05-18 (filing of the continuation) / recorded 2018-09-20 — Reel/frame: not retrieved

• Conveyance: Assignment (inventors → company)
• Assignor: Ng, Kian Siang; Wang, Ji; Yu, Qichuan; Gubser, Simon; Hanselmann, Sonja; Rudmann, Hartmut
• Assignee: Heptagon Micro Optics Pte. Ltd.
• Correspondent: not retrieved
• Context: Routine original assignment of inventors' rights to the operating employer, filed with the continuation. Not an NPE signal.

2. Executed 2018-02-18 (effective) / recorded 2019-02-08 — Reel 048289/0147 (reported for this patent family; confirm for this patent number)

• Conveyance: Change of Name (not a transfer of ownership)
• Assignor: Heptagon Micro Optics Pte. Ltd.
• Assignee: ams Sensors Singapore Pte. Ltd.
• Correspondent: not retrieved. Flag: the correspondent on this instrument is the key data point to pull from Assignment Center — it is the same firm that will reappear on link 3 if a single outside counsel runs the ams/ams-OSRAM IP renamings.
• Context: Internal corporate renaming within the ams group after the 2017 acquisition. No change in ultimate beneficial ownership. The reel 048289/0147 figure is consistent with a recording made in early 2019, matching the 2019-02-08 record date on this patent; it is also cited against sibling Heptagon/ams patents (e.g. US 8,828,174; US 2011/0043923), which supports — but does not by itself prove — that this patent uses the same reel/frame.

3. Executed on/around 2025 / recorded 2025-11-03 — Reel/frame: not retrieved

• Conveyance: Change of Name (not a transfer of ownership)
• Assignor: ams Sensors Singapore Pte. Ltd.
• Assignee: ams-OSRAM Asia Pacific Pte. Ltd.
• Correspondent: not retrieved
• Context: Internal corporate renaming within the ams-OSRAM group (the ams-OSRAM AG rebrand). Again a change of name only — the same Singapore UEN 200704497W persists across all three names.

Total ownership transfers to a third party: zero. Every recorded instrument is either (a) the original inventor→employer assignment, or (b) a change-of-name of the same legal entity (same UEN).

Timeline diagram

timeline
    title Ownership of US 10190908
    2015 : Provisional filed by inventors
    2016 : Parent application filed
         : Named to Heptagon Micro Optics
    2017 : ams AG completes Heptagon acquisition
    2018 : Continuation filed
         : Inventors assign to Heptagon
    2019 : Patent issued
         : Name change to ams Sensors Singapore
    2025 : Name change to ams-OSRAM Asia Pacific

NPE / troll-pattern signals

Verdict

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

This patent's recorded chain contains no NPE signal of any kind: the only true ownership transfer is the original inventor→employer assignment (recorded 2018-09-20), and the two subsequent instruments (recorded 2019-02-08 and 2025-11-03) are Change of Name only, keeping the patent inside the same Singapore entity (UEN 200704497W) that was Heptagon Micro Optics and is now ams-OSRAM Asia Pacific Pte. Ltd., a subsidiary of a listed operating semiconductor manufacturer. There is no shell LLC, no known asserter, no cascading LLC chain, no pre-suit transfer and no litigation.

The caveat: the "Defensive / non-asserting" bucket in the brief is defined by termination at a defensive aggregator, and that parenthetical does not strictly apply here — this is simply an operating-company-owned, non-asserted patent. I chose it over "Operating-company assertion" because the latter requires the owner to be suing actual competitors, and no suit was found. If you insist on strict label fidelity, the closest honest reading is "non-asserting operating-company ownership" rather than any NPE category. Two items should be closed out manually before relying on this: (a) the correspondent of record on all three instruments (signal 3 remains unclear), and (b) whether the 2024 Focuslight asset purchase swept this patent in without a recorded USPTO assignment.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent number 10190908; the Assignment Abstract of Title will return the reel/frame, execution dates and correspondent for each of the three instruments above). Google Patents legal events mirror: https://patents.google.com/patent/US10190908/en

Sources consulted

• Google Patents — US 10,190,908 B2 legal events and description (authoritative text supplied): https://patents.google.com/patent/US10190908/en
• ams press release, "ams announces completion of transaction to acquire Heptagon" (2017-01-24): https://ams-osram.com/ja/news/press-releases/24-01-2017-ams-announces-completion-of-transaction-to-acquire-heptagon-and-related-capital-increase
• Wolf Theiss press release on the Heptagon/ams deal (2016-10-25): https://www.wolftheiss.com/insights/wolf-theiss-advises-sellers-of-heptagon-in-mega-deal-with-ams-ag/
• ACRA / corporate registry data for AMS-OSRAM ASIA PACIFIC PTE. LTD. (UEN 200704497W; former names Heptagon Micro Optics, ams Sensors Singapore): https://opengovsg.com/corporate/200704497W
• Focuslight Technologies press release, acquisition of ams OSRAM optical-component assets (2024-09-02): https://www.focuslight.com/wp-content/uploads/2024/09/Focuslight-Technologies-Inc.-Completes-Acquisition-of-ams-OSRAMs-Optical-Component-Assets.pdf
• Focuslight, "Subsidiary Renaming: Focuslight Singapore is now Heptagon Photonics" (2025-02-05): https://www.focuslight.com/news-events/newslist/subsidiary-renaming-focuslight-singapore-is-now-heptagon-photonics-to-strengthen-branding-alignment/
• Sibling-patent legal events citing reel/frame 048289/0147 for the Heptagon→ams Sensors Singapore change of name: https://patents.google.com/patent/US20110043923A1/en

I'll search for the USPTO record and citation data for US Patent No. 10,190,908 specifically.

The searches didn't surface the citation list. Let me target the patent's front page and citation data directly.

I have the patent's text but not its citation list yet. Let me try to pull the front-page citation data from patent databases.

Bottom line up front

I could not retrieve the USPTO "References Cited" (front‑page / PTO‑892) citation list for US 10,190,908 B2 in this session. My targeted queries returned unrelated documents (unrelated PTO‑892 forms from other applications, unrelated Google Patents pages, and EPO search reports for other patents), and I then hit the tool‑step limit before I could confirm the list. I will not fabricate citations or dates for examiner‑cited references I could not actually retrieve. Below I give (1) what I can verify about this specific patent, (2) the citation‑type relationships I can ground in sources, and (3) a § 102 framework plus the exact steps to obtain the authoritative list.

1. Patent verified (matches the authoritative full text)

⚠️ Literal‑ID flag (consistent with the earlier section): do not conflate this with US 9,116,908 (Realtime Data). They are different patents; the earlier litigation summary already warned about this, and I am preserving that distinction.

2. What I actually could and could not retrieve

• Could not retrieve: the USPTO front‑page "(56) References Cited" list (U.S. patent documents, foreign patent documents, and non‑patent literature) for US 10,190,908, nor the prosecution PTO‑892 "Notice of References Cited" for application 15/984,278 or its parent 15/353,397.
• Queries attempted (for transparency): "US10190908 patent citations / 'References Cited'"; "10,190,908 … prior art cited"; Google Patents front‑page citation scrape; Espacenet cited‑documents; and the parent US 9,976,894 family. None returned the target list; several returned unrelated PTO‑892/EPO documents, which I explicitly decline to present as this patent's citations.
• Could retrieve: the patent's full description/claims (authoritative), the family table, and the specification's own incorporation‑by‑reference statements.

Consequence: I cannot responsibly state, per reference, "this reference potentially anticipates claim(s) X under § 102," because § 102 anticipation requires element‑by‑element comparison to specific claims against specific references — and I do not have the verified reference list.

3. Citation‑type relationships I can ground

3(a) Documents incorporated by reference in the specification itself

These are cited in the patent text (Google Patents, US10190908B2). Note: incorporated‑by‑reference documents are not automatically § 102 prior art — they are only prior art if they independently qualify (date + disclosure). Flagging them because they are the only "citations" I can verify from the document itself:

3(b) Same‑family members (from the family table in the related record)

• US 9,976,894 B2 (filed 2015‑11‑17; granted 2018‑05‑22; Heptagon) and WO 2017/086878 A1 (2015‑11‑17; 2017‑05‑26) — "Cross‑talk prevention structure of electronic device for measuring distance to external object."
• US 2020/0319306 A1 (Heptagon) — related continuation.

⚠️ § 102 caveat: family members sharing the same priority/owner may be disqualified as prior art under AIA § 102(b)(2)(C) (common ownership) and/or excluded where they claim the same invention (§ 102(b)(2)(A–B)). Their only prior‑art bite would be through § 102(b)(2)(C)‑disqualified analysis or as evidence of a cross‑talk/opaque‑dam concept in § 103 combinations — not straightforward § 102 anticipation. I cannot state a claim‑by‑claim § 102 mapping without the actual reference set.

3(c) Related Heptagon background art surfaced (neighboring, NOT the cited list)

These appeared in searches and are topically adjacent, but I did not confirm them as citations on this patent's face: US 2014/0339664 A1 and US 2017/0077162 A1 (Heptagon, "Optical devices and opto‑electronic modules… non‑transparent material on a surrounding portion"); US 2015/0200221 A1 (Heptagon, computational cameras/wafers); US 2008/0054508 A1 (Heptagon, replication tool). Treat as leads only.

4. How the § 102 analysis must be run (framework for the claims at issue)

The claim set has three independent‑claim families (per the authoritative text):

1. Base apparatus — substrate + emitter chip + sensor chip (main + reference) + opaque dam separating them + first/second transparent encapsulation blocks + opaque encapsulation material with first opening over main sensor, second over emitter, extending between blocks and abutting the dam.
2. Plus passive optical elements — first/second transparent blocks include a passive optical element (e.g., lens), with opaque coating around each element and openings aligned therewith.
3. Plus substantial top‑coverage — first/second opaque coatings cover substantially all the top surface; opaque encapsulation material encapsulates side surfaces and abuts the opaque coating along top/side edges.

Dependent‑claim differentiators to test every candidate reference against: channel interlock (dam↔opaque material), trench in substrate, differing dam thickness over substrate vs. over sensor chip, and viscosity difference between two opaque epoxies. Anticipation (§ 102) requires a single reference disclosing all elements as arranged; anything less is a § 103 combination, not § 102.

5. To obtain the verified citation list (authoritative sources)

1. USPTO PatentCenter / Patent Public Search (ppubs.uspto.gov) → Application 15/984,278 (and parent 15/353,397) → "Documents & Transactions": the PTO‑892 "Notice of References Cited" and the IDS/SB‑08 list.
2. USPTO full‑page PDF (front page "(56) References Cited") via PatentCenter "Full pages."
3. Google Patents → patents.google.com/patent/US10190908B2/en → "Citations / Cited By" panels (present in HTML even when the summary scrape truncates).
4. Espacenet → "Cited documents" / "Also published as" for family US15/984,278.

6. Caveats

• I am reporting an inability to verify, not that no citations exist. Every granted U.S. patent has at least a front‑page reference section, so a list does exist; I simply could not load it here.
• Per the operating rules, I have not auto‑corrected any number: "15/984,278," "15/353,397," "62/257,198," "62/256,238," "PCT/SG2015/050224," "US 9,976,894 B2," "US 9,116,908," and "US 10,190,908" are reproduced literally.
• The earlier litigation summary (no litigation/PTAB found) is unaffected by and consistent with this analysis; no contradiction found.
• If you can paste the front‑page "(56)" list or the PTO‑892 text, I will produce the requested per‑reference table (full citation, publication/filing date, brief description, and specific claims potentially anticipated under § 102) precisely.

I'll research the prior art landscape for this patent family before building the obviousness analysis.

Let me try to retrieve the actual cited prior-art references for this patent.

Obviousness Analysis — U.S. Patent No. 10,190,908 (35 U.S.C. § 103)

Patent: US 10,190,908 B2, "Optical devices and methods of making same"
Source: https://patents.google.com/patent/US10190908/en (fetched 2026-09-24)
Priority date per the record: 2015-11-18; app. 15/984,278 filed 2018-05-18 as a continuation of 15/353,397 (which issued as US 9,976,894); anticipated expiration 2036-11-16.

Note on the "Prior Art" section of this page. In the snapshot I was given, the Prior Art section contains only the algorithmic fields — Prior art keywords: opaque, optical, optical sensor, dam, transparent; Prior art date: 2015-11-18 — and the page is truncated inside the Detailed Description, so the patent's own "Citations / Cited By" tables are not present. I therefore reconstructed the prior-art landscape from (a) references the '908 itself expressly incorporates by reference, (b) its parent/sibling family members that surfaced in search, and (c) earlier, independently-published third-party/family art. Where I am relying on partial recall rather than a document I retrieved in-session, I say so explicitly. Do not treat any citation below as verified unless I mark it ✅.

1. What has to be proven

Under § 103, the inquiry is whether the subject matter of each claim as a whole would have been obvious to a PHOSITA at the effective filing date. The '908 has three independent claims (the page's "Definitions" section reproduces them as three "aspects"):

Dependents across the three aspects add: (i) a channel in the dam receiving the opaque encapsulant (or a channel in the encapsulant receiving the dam — i.e., an interlock); (ii) two different opaque epoxies with different viscosities; (iii) trench in the substrate filled by the opaque encapsulant; (iv) dam thicker over the substrate than over the sensor chip; (v) lens elements.

Two claim-construction points drive the whole analysis:

1. "Reference sensor" is not defined in the specification. The '908 only says the IC chip "can control emissions by optical emitter chip 125 and process information received from main optical sensor 117 and reference optical sensor 122." The functional result claimed by the architecture — reference sensor co-encapsulated with the emitter in the same transparent block, main sensor isolated in the other block — is that the reference sensor sees emitter light directly (emitter-output/temperature/crosstalk monitoring), while the main sensor sees only externally reflected light. That architecture is the real point of novelty.
2. "Opening" in the opaque encapsulation must be read as an aperture/passage in the opaque overmold, not as absence of material generally — otherwise the amendment history would be the controlling evidence, which I could not retrieve.

2. Effective filing date and which art qualifies (this is where most § 103 attacks on this patent will live or die)

The '908 is AIA-first-inventor-to-file. Four candidate layers of art:

Practical consequence: the single closest document (R1) is the applicant's own incorporated-by-reference material, and the second-closest (R4) is also Heptagon's. The strongest third-party-neutral § 103 position therefore has to be built on R4 + EP 2 834 683 (R3) + general knowledge, with R1 used as evidence of what the art already taught rather than as the statutory primary reference. I flag this as a defense to expect, not as a settled outcome.

3. Element-by-element mapping — first independent claim (Aspect 1)

Result: every element of Aspect 1 is disclosed, or would be an obvious mechanical consequence of the fabrication sequence, across R1 (with R3/R4 supplying the aperture-through-opaque-layer teaching). The only element not literally recited in R1 is the "opening above the emitter / above the main sensor," which R3 and R4 supply.

4. Grounds of rejection

Ground I — R1 (WO 2016/013977 / US 2016/0229505) in view of R4 (WO 2013/010284) and/or R3 (EP 2 834 683)

Targets: Aspect 1; Aspects 2–3 with additional art.

• R1 teaches the complete light-tight architecture: a dam of black epoxy across the sensor chip separating two light-sensitive regions, a clear encapsulation split into two optically isolated channels, and black epoxy filling the trench between the channels — which is precisely "opaque encapsulation material … extend[ing] between the first transparent encapsulation block and the second transparent encapsulation block and abut[ting] the opaque dam."
• R4/R3 supplement the one missing element, teaching an opaque layer with a transparent window/aperture over each optical channel and a carbon-black-loaded epoxy separation member dedicated to crosstalk suppression in a proximity module.
• Combination result: the claimed module, with no new and unexpected behavior — the openings simply preserve the light path through the already-transparent blocks.

Ground II — R4 (WO 2013/010284) as primary, + R3, + R5

Targets: Aspects 1–3 where the examiner wants a third-party/family-neutral primary reference (avoids the common-ownership issue in Ground I).

• R4 gives: wafer-scale substrate carrying emitters and detectors, spacer wafer, and an optics wafer with transparent portions and opaque blocking portions — i.e., a multi-channel module with channel-to-channel optical isolation and a window per channel.
• R3 gives: the emitter/detector proximity architecture with a non-transparent separating member and structured lens surfaces.
• R5 gives: spacer members with openings and a lens formed in a transparent portion, plus the multi-member stacking that approximates the trench-and-fill geometry.
• The remaining step — placing a reference sensor on the sensor chip on the emitter side and letting the single transparent block cover emitter + reference sensor — is the weakest link in this Ground and is addressed in § 5.

Ground III — Any of the above + the known "reference-channel compensation" technique

Targets: the reference-sensor placement, and the Aspect-2/Aspect-3 coating limitations.

• Co-locating a reference/monitor photodetector with an emitter so that emitter output (and its drift with temperature and age) is sampled and used to normalize the signal channel is a long-standing, textbook practice — monitor photodiodes in optical transmitters and ambient-light-cancellation reference photodiodes in optical sensors are the canonical instances. ⚠️ I could not retrieve a specific citation for this in-session; a grounded reference must be identified before filing. If the examiner can produce one, the reference-sensor element falls.
• For Aspect 2/3 (lens + photostructurable opaque coating apron covering substantially all of the top surface, abutted by the encapsulant along the caps): the '908 itself describes the coating as "a photostructurable material such as, e.g., a resist or photoresist material" applied by spray/spin coating and developed by LDI or mask, and its incorporated R2 (US 62/256,238, "Thin optoelectronic modules with apertures") is directed to exactly that. R4/R5 supply the lens-in-transparent-portion teaching. The "abutting along one or more edges" limitation is a straightforward consequence of applying the coating over the whole top surface before dicing and then overmolding — i.e., the geometry follows from the stated process, not from a separate invention.

Dependent-claim notes

• Interlock (channel in dam / channel in encapsulant): R1's trench-over-foundation-layer, filled with black epoxy, is an interlocking opaque-dam/opaque-encapsulant structure; the '908 in fact obtains it by dicing into the dam. Obvious in view of R1 + the '908's own admitted processing.
• Two opaque epoxies of different viscosity: KSR "predictable variation" — selecting a higher-viscosity dam material to prevent bleed onto the sensor is expressly stated in the '908 as the purpose, and R1 discloses adhesive viscosities in the 400–7,000 MPa·s range. Optimizing a viscosity differential is routine.
• Trench in the substrate filled by the encapsulant: disclosed/obtained by the dicing step common to R1 and the '908's own FIGS. 10A–10F.
• Dam thicker over the substrate than over the sensor chip: an artifact of the syringe-dispense/slow-down process the '908 itself describes; no inventive weight.

5. Motivation to combine (the KSR rationales an examiner would invoke)

1. Same field, same problem, same solution family. R1, R3, R4 and R5 are all proximity-sensor/optoelectronic-module packaging art directed at crosstalk suppression and light-tight channel isolation — the exact problem the '908 states it solves ("improved optical isolation between optical emitter chip and main optical sensor").
2. Known technique, improvement of a similar device (KSR factors 4 & 5). Using a reference channel to normalize an emitter-driven measurement is a known technique used to improve a device in the same way (proximity/distance accuracy). Placing that reference sensor on the sensor chip rather than as a discrete die is an improvement of a similar device in the same field.
3. Design need / predictable results. The '908's own claimed benefit (reference sensor co-encapsulated with the emitter) is exactly the predictable consequence of the architecture: sample the emitter directly, isolate the main channel behind an opaque dam + opaque encapsulant. No new physical principle.
4. Manufacturing pressure (design incentive). Wafer-level, dicing-based singulation with trench fill is the dominant manufacturing paradigm shown across R1/R3/R4/R5; arriving at the '908's structure requires only ordinary process sequencing (dispense dam → clear overmold → dice trench → inject black epoxy → singulate), which the '908 admits at FIGS. 10A–10F.
5. Common ownership / common problem statement. All the Heptagon references share a common specification lineage and problem statement, which strengthens (for a challenger) the showing that the combination was within the ordinary skill—but simultaneously creates the § 102(b)(2)(C) problem in § 6.

6. Counter-considerations, and where a § 103 challenge is weak

1. § 102(b)(2)(C) common ownership. R1, R2 and the parent US 9,976,894 are all Heptagon/ams-Osram. If the assignment records show common ownership no later than the '908's effective filing date, R1/R2 are disqualified as § 102(a)(2) art, and the examiner is forced onto R3/R4/R5 — a materially weaker, less geometrically specific record. This is the single most important thing to verify.
2. § 102(b)(1)(A) grace period for R5. If US 2015/0179827 A1 (published 2015-06-25) shares inventive entity with the '908, the one-year grace period may remove it as § 102(a)(1) art.
3. Priority / written-description pressure point. The express incorporation by reference of R1 (effective 2015-07-22) and R2 (2015-11-17) means the '908's own disclosure is, in part, the incorporated text. If any claim is supported only by incorporated subject matter, a challenger will argue either (a) the '908's effective date reaches back to those documents for that subject matter (removing them as art but confirming they describe it), or (b) the claim lacks written-description support in the 2015-11-18 provisional. Both cut against the patent.
4. Genuine non-obviousness candidates. The strongest defense arguments are: (i) the combination of a reference sensor on the same die as the main sensor with the emitter, in a shared transparent block is a specific architectural choice not shown in any single reference — no reference retrieved so far discloses a two-sensor, two-block, one-dam topology where the block boundary is the dam itself; and (ii) the edge-abutment geometry of Aspect 3 (opaque encapsulant meeting the opaque coating exactly along the top/side edge of each block) is a specific light-tightness solution whose obviousness depends entirely on whether the photostructurable-coating art (R2/R5) is treated as analogous. I have not found affirmative evidence of teaching away in any retrieved reference.
5. Secondary considerations. None evidenced in the record I retrieved (no litigation, no license, no commercial-success evidence surfaced — consistent with the earlier litigation summary finding no litigation for this patent).

7. Bottom line

Conclusion: Every structural feature of the '908 is individually present in the retrieved corpus (R1 is startlingly close, as one would expect for the same applicant's earlier application that the '908 expressly incorporates). A § 103 rejection of at least the broadest independent claim is defensible if and only if the primary reference survives the § 102(b)(2)(C) common-ownership exception. Failing that, the challenger must lean on R4 + R3 + R5 + a reference-channel compensation citation, which is a workable but noticeably weaker combination because the two-sensor/one-dam topology is then only pieced together.

8. Verification checklist before anyone relies on this

1. Retrieve the actual claim text of US 10,190,908 (I used the page's "Definitions"/Summary, which mirrors but may not equal the granted claims) and the file wrapper (Google Patents → USPTO PatentCenter) to see whether the "opening" and "reference sensor" limitations were narrowed during prosecution — that history may be the decisive § 103 datapoint, and I could not retrieve it.
2. Pull the "Citations" and "Cited By" tables for US 10,190,908 and US 9,976,894 directly (they were absent from the snapshot supplied).
3. Confirm the § 102(b)(2)(C) common-ownership facts in the USPTO assignment records for WO 2016/013977 / US 2016/0229505 / US 10,199,412 and US 62/256,238 as of 2015-11-18.
4. Verify EP 2 834 683's exact number, assignee, priority and publication date (I retrieved the EPO document host path but not its bibliographic front page).
5. Verify assignee/inventorship of US 2015/0179827 A1 to assess the § 102(b)(1)(A) grace-period exception.
6. Identify a specific, citable reference for reference-photodetector compensation (proximity sensing, ambient-light cancellation, or monitor-photodiode emitter stabilization). Without it, the reference-sensor element is the most likely point of non-obviousness to survive.

Flagged contradictions / discrepancies

• Priority date: the '908 record states 2015-11-18, but a third-party family table listed US 9,976,894 B2 with a 2015-11-17 date. The '908 itself is a continuation of application 15/353,397 (filed 2016-11-16) and also incorporates US 62/256,238 filed 2015-11-17. These are two different provisional filings one day apart (62/256,238 on 11-17 and 62/257,198 on 11-18). This is consistent with the record but should not be conflated.
• Nothing in this analysis contradicts the earlier litigation summary (no litigation found for US 10,190,908). I found no litigation in the course of this prior-art work either.