US 5999685

I'll run searches against the patent record and any 2026 CAFC/litigation dockets mentioning this number.

Let me run a few more targeted checks on litigation/ownership and any 2026 appellate activity.

US 5,999,685 — Analyst Summary

Sourcing caveat up front: the full patent text you supplied (Google Patents) is the authoritative record here and takes precedence over search results. My searches returned no CAFC 2026 docket and no pending 2026 litigation involving this number. See "Litigation / CAFC 2026" below — I flag that as a genuine gap, not a null result to be over-read.

Bibliographic data

Ownership chain (per Google Patents reassignment records, which carry a stated accuracy disclaimer): Sanyo Electric Co., Ltd. → Foltora Service Limited Liability Company (Sep 17, 2009) → Callahan Cellular L.L.C. (Jan 14, 2016) → Intellectual Ventures Assets 158 LLC (Jan 28, 2020) → Hanger Solutions, LLC (Jan 4, 2020; Google's listing shows Hanger Solutions LLC as current assignee). The intermediate IV-assets step is listed with an effective date (Jan 4, 2020) earlier than the recorded reassignment date (Jan 28, 2020) — a record-keeping artifact, not an error I am correcting.

Abstract (as issued)

A light guide plate includes a first lens portion comprising a plurality of recessed or projected lenticular unit lenses formed of the same resin as that of the plate, on a light emission surface. The lenticular unit lenses are triangular prism portions whose vertexes have an angle of 125° to 165°, arranged with ridge lines nearly parallel. This allows the number of separate lenticular lens sheets to be reduced to one or eliminated. A second lens portion may be provided on the opposite reflective surface. Alternatively, grooves of approximately equal width are formed on the surface opposite the emission surface, parallel to each other, with spacing narrowing with distance from the light-incidence surface, the spacing from the far end of the groove nearest the incidence surface to the near end of the next groove being 5 to 15 times the groove width.

Plain-language overview of each independent claim

• Claim 1 — Light guide plate (single-sided prism lens). A resin light guide plate that emits light from a light-emission surface. On that surface it has a first lens portion made of the same resin as the plate itself (i.e., molded integrally, not a laminated film). That lens portion is a set of recessed or projected lenticular unit lenses shaped as triangular prisms with apex ("vertex") angles of 125°–165°, and their ridge lines run nearly parallel to each other.
• Claim 6 — Surface light source (single-sided). A light source placed adjacent at least one side end surface of a claim-1-type plate, the combination emitting light from the emission surface.
• Claim 8 — Light guide plate (double-sided). Same first lens portion as claim 1, plus a second lens portion on the reflective surface opposite the emission surface — also integrally formed of the same resin, also triangular prism portions with vertex angles of 125°–165° and nearly parallel ridge lines.
• Claim 15 — Surface light source (double-sided). A light source adjacent at least one side end surface of a claim-8-type double-lens plate.
• Claim 16 — Light guide plate (grooved back surface, graded pitch). One side end surface is the light-incidence surface; a surface crossing it is the emission surface. On the surface opposite the emission surface, multiple grooves of approximately equal groove width run nearly parallel to each other, and the groove-to-groove spacing narrows as you move farther from the incidence surface. The key numeric limitation: the spacing measured from the far end of the groove nearest the incidence surface to the near end of the adjacent groove farther away is 5 to 15 times the groove width.
• Claim 19 — Surface light source (grooved type). A light source adjacent at least one side end surface of a claim-16-type grooved plate; emits light from the emission surface.

Dependent-claim notes: claims 2–3 narrow the first lens apex to 135°–165°, then ~150°; claim 4 requires isosceles triangular prisms; claim 5 recites a plate thickness of about 1–10 mm. Claims 9–14 apply analogous apex-angle and geometry limitations to the first and second lens portions, including claim 10's requirement that the two lens portions' ridge-line directions be nearly perpendicular. Claims 17 (groove cross-section is an N-sided polygon with N ≥ 3, an isosceles triangle, or a part of an ellipse/circle), 18 (wedge-shaped plate), and 20 (wedge-shaped light source plate) round out the set.

Technical substance / prosecution context

• The specification frames the invention against the then-conventional approach of stacking two right-angle-crossed lenticular lens sheets over a light guide (FIG. 15; citing Monthly Display, May 1996, pp. 35–39), plus light-diffusion dot-pattern printing on the back surface — both of which the patent aims to eliminate to cut parts and assembly cost.
• The 125°–165° apex range is justified by optical simulation using "CODE-V" (Optical Research Associates), at 600 nm with PMMA (n = 1.49), comparing apex angles of 90°, 120°, 125°, 150°, 165° and flat. The specification states 90°–120° performs worse than flat, 125°–165° is substantially better, 135°–165° significantly better, and ~150° shows a peak (FIGS. 4–5). FIG. 6 reports luminance distribution is insensitive to prism pitch over 10 µm–1000 µm.
• For the grooved embodiment, simulation fixes groove width at 0.2 mm with a 150° isosceles cross-section and varies the nearest groove spacing; the specification reports good uniformity at 5×, 10×, and 15× groove width, and poor results at 2.5× and 20× — the basis for the claimed 5–15× range (FIG. 12).
• Materials disclosed: acrylic/polycarbonate primarily, plus PMMA, polyacrylic acid methyl, methacrylate esters/copolymers, PET/PBT polyesters, polystyrene, polymethylpentene, UV/EB-cured urethane and polyester acrylates, unsaturated polyesters, glass, ceramics.
• Cited prior art of record: US 1,707,965; US 2,347,665; US 3,328,570; US 5,408,344; US 5,420,761; US 5,575,549; US 5,600,455; US 5,735,590; US 5,854,872; and a non-patent citation to "Abstract of Japanese Patent Publ. No. 7120752, dated May 12, 1995." The patent also cross-references applicant's earlier application Ser. No. 08/742,500 (replaceable fine-pattern mold insert and thermal-transfer-foil opaque coating), which is a co-pending-family rather than prior-art reference.
• Classifications: G02B6/0038, G02B6/0033, G02B6/0035, G02B6/0011, Y10S385/901.

Litigation / CAFC 2026 search result — explicit uncertainty

• I found no docket, order, or opinion dated 2026 naming U.S. Patent 5,999,685, and no Federal Circuit appeal for this number.
• My searches for Hanger Solutions, LLC patent suits surfaced cases asserting other patents — e.g., U.S. 6,430,623; 6,609,159; 6,772,227 (Hanger Solutions v. Charter Communications, W.D. Tex. 6:22-cv-00070); U.S. 6,119,171; 6,430,623 (Hanger Solutions v. Cequel Communications, W.D. Tex. 6:22-cv-00330); and Hanger Solutions v. Squarespace (D. Del. 1:21-cv-00774). None of those complaints listed 5,999,685 among the asserted patents in the documents retrieved.
• The PTAB/proceeding hits that appeared in search results (e.g., IPR2014-01097, IPR2015-00490, IPR2015-00360) attach to a different patent record (an Innovative Display Technologies / Rambus-family patent), not to 5,999,685 — I am not attributing them here.
• Two reasons to expect little 2026 activity on this number: the patent's term expired Sep 2, 2017 (anticipated expiration per the record), so new infringement assertions post-date the term; and the Hanger Solutions campaigns visible in the record cluster in 2021–2022. That said, an appeal arising from pre-expiration conduct can be briefed and decided years later, so I cannot rule out a 2026 appeal on this record.
• Bottom line: I do not have authoritative 2026 CAFC or district-court information for U.S. 5,999,685. A PACER/CM-ECF docket query (via the Federal Circuit's own docket or PACER) and a USPTO PatentCenter/Assignment check would be needed to confirm current status definitively; nothing in my retrieved sources contradicts the "Expired – Lifetime" status shown on the patent record.

One note on identifier fidelity

Per your instruction I have not normalized anything: the number is written 5,999,685 / US5999685A, application 08/921,606, and I have kept the JP priority numbers (JP 9-025652 / JP9-099122 / JP9-151180) and publication IDs (JPH10282496A, JPH10339815A) exactly as they appear in the record, including the internal inconsistency where one source lists JP 9-025652 with priority date 1997-02-07 while the family table also lists 1997-04-16 and 1997-06-09 for the later-filed JP applications.

I'll research litigation involving US patent 5999685. Let me run several targeted searches.

Let me search specifically for the current assignee and litigation databases.

Let me check the Unified Patents litigation portal and other databases directly.

Bottom line

Based on the sources I was able to search, I found no litigation that asserts US Patent 5,999,685 ("Light guide plate and surface light source using the light guide plate"). I could not locate any district court complaint, CAFC appeal, ITC action, or PTAB proceeding identifying the '685 patent as a patent‑in‑suit. I am reporting this as a negative finding with the caveats below rather than manufacturing case details.

What the searches did surface (and why it is not the '685 patent)

The patent is currently recorded as assigned to HANGER SOLUTIONS, LLC (via Intellectual Ventures Assets 158 LLC / Callahan Cellular L.L.C. / Foltora Service L.L.C.), and Hanger Solutions is an active patent litigant. However, the Hanger Solutions campaigns that appear in the dockets I reviewed assert different patents, not the '685 patent:

Sources: dockets.justia.com (D. Del. 1:21‑cv‑00040; 1:21‑cv‑00774; W.D. Tex. 6:22‑cv‑00330); unicourt.com case pages for the same; portal.unifiedpatents.com litigation page for 1:21‑cv‑00040; Lexology/Georgia IP Litigation Blog (Mediacom).

Important: every patent‑number exhibit list I retrieved from these Hanger Solutions complaints names the '623, '159, '227 and '171 patents. None named the '685 patent. So while Hanger Solutions holds the '685 patent, these particular suits are not '685‑patent suits.

What I checked

• Google Patents / Justia page for US 5,999,685 (the litigation/assignment metadata and "Cited By" tables list no "Litigation" entries for this patent).
• Unified Patents litigation portal (searched against the number and against the current assignee).
• Hanger Solutions docket records (Justia, UniCourt, CourtListener) — the assignee NPE.
• General searches for the bare numbers "5999685" and "5,999,685" combined with litigation terms, which returned only patent‑prosecution citations (e.g., the number appears as a cited reference in EP applications such as EP 2 372 416, EP 2 280 299, CN 102947764, and in PTAB/IPR exhibit lists) — i.e., prior‑art citations, not assertions.

Caveats on the negative finding

• My search tooling reached its limits before I could exhaustively query PACER, CourtListener's full docket corpus, or Docket Navigator's "patents asserted" filter. District court complaints that were filed and quickly dismissed, or cases where the '685 patent was one of many patents‑in‑suit in a large NPE campaign, could exist without surfacing in these queries.
• The '685 patent expired September 2, 2017 (20 years from the Sep 2, 1997 filing), and the record shows "Anticipated expiration – Expired – Lifetime." All asserted claims would have expired nearly nine years ago, which sharply reduces (though does not eliminate) the likelihood of recent litigation.
• The number overlaps numerically with unrelated litigation documents (e.g., a Medtronic/NuVasive spinal‑implant matter and an Amneal/Purdue IPR referenced "U.S. Patent No. 6,945,933" and similar), which I excluded because they are different patents.

Suggested next steps to confirm definitively

1. Run a targeted Docket Navigator or Lex Machina "patents asserted" query on 5,999,685 (these index every patent‑in‑suit in a complaint).
2. Check PACER for any case naming "5,999,685" in the complaint caption, and the USITC EDIS for Section 337 complaints involving the patent.
3. Query the CAFC docket for appeals referencing the '685 patent.

If you'd like, I can draft the exact search strings for Docket Navigator/Lex Machina, or dig specifically into whether the Foltora→Callahan→IV Assets 158→Hanger chain ever generated a '685‑patent assertion in a specific campaign.

I'll start by verifying the structured data against live sources.

Let me run a few more targeted checks before concluding.

PTAB proceedings on US 5,999,685

Bottom line up front: I found no AIA trial proceedings — no IPR, PGR, or CBM — on US 5,999,685, in either the structured USPTO ODP data supplied with this task or in independent web searching. I want to be explicit that this is a negative finding from a search, not a claim of exhaustive verification: I could not open PTAB E2E / PTAB Center directly in this session, and the ODP ingest the task relies on may lag. But every apparent "hit" I pulled and checked resolved to a different patent that merely cites the '685 as prior art — those lookalikes are set out below so you don't repeat the error.

Proceedings overview

Total AIA trial proceedings on US 5,999,685: 0. No active proceedings, no claims invalidated by the Board, no claims sustained in an FWD, no settlements, no institution denials, no Federal Circuit appeals from a PTAB FWD. This is not a "patent survived two IPRs and is hardened" posture, and it is not a "claims 1–5 are canceled" posture either — the claims carry their original, un-amended scope. Two structural facts dominate the defensive analysis: (1) the patent expired 2017-09-02 (anticipated expiration, per the structured data), so it cannot be asserted for prospective infringement and IPR would be a paper exercise with no live controversy; and (2) it is a pre-AIA patent (priority 1997-02-07, filed 1997-09-02), which means PGR and CBM are unavailable — PGR only reaches patents with an effective filing date on/after 2018-03-16's statutory trigger (AIA first-inventor-to-file, applying to applications filed on/after 2013-03-16), and CBM review sunset for new petitions on 2020-09-16, with the transitional-program window long closed. IPR was the only ever-available AIA vehicle, and nobody used it.

Because there are no proceedings to format, the per-proceeding sections below are not applicable. The remainder of this memo addresses what the absence means and how to handle a current assertion.

False positives I checked and rejected (do not cite these as proceedings on the '685)

These surfaced in search and are all not proceedings on US 5,999,685. Flagging them because "5999685" and similar five-to-six-digit strings are easy to misattribute:

• IPR2014-01431 — VTech v. Spherix (U.S. Pat. No. 5,581,599) — a different patent entirely; the "599" match is coincidental (Spherix cordless-telephone patents). Source: RPX/PTAB FWD excerpt.
• IPR2021-00987 / -01294 / -01458 (U.S. Pat. No. 8,489,599, PARC) — a Palo Alto Research Center/Snap/Xerox dispute over a 2013 "context and activity-driven content" patent. Unlike ours by three orders of magnitude in patent-number space. CAFC lead No. 23-1983. Source: CourtListener CAFC 23-1983 docket.
• IPR2025-00165 (U.S. Pat. No. 9,905,599) and IPR2025-00161-ish / Sony v. Optimum Imaging (U.S. Pat. No. 10,873,685) — unrelated image-processing patents.
• Numerous patent-family and search-report citations to "US 5999685 A (HIBINO KATSUTOSHI ET AL)" — e.g., EP1493051B1 and WO2003085428A1 cite the '685 as prior art, confirming the '685 is a widely-cited reference, not a challenged patent.

If a search vendor or an AI tool tells you there's an "IPR on 5999685," it has almost certainly confused the number with one of the above.

Strategic summary

Claim status: all 20 claims UNTESTED, none canceled, none confirmed. Claims 1–15 (the lens-portion family: unitary triangular-prism lenticular unit lenses at 125°–165° apex angle on the emission surface, optionally with a second ridge-line-perpendicular lens portion on the reflective surface) and claims 16–20 (the groove family: equal-width grooves on the reflective surface with spacing narrowing away from the incidence surface, ratio 1:5 to 1:15) stand exactly as granted on 1999-12-07. There has been no inter partes or ex parte reexamination narrowing either. For a defendant, that cuts both ways: you cannot point to a Board FWD and say "claim 1 is dead," but you also inherit zero § 315(e)(2) estoppel, so every prior-art ground is available to you — including art that the examiner of record never considered and art listed in the IDS-type citations visible in the family documents (e.g., US 5,605,455 and US 5,575,549 to Enplas, US 5,735,590 to Tosoh, and the non-patent citation to Japanese Utility Model/Patent Publ. No. 7-120752, dated 1995-05-12, cited by the examiner). § 325(d) discretionary denial is not a factor you need to worry about, because there is no prior Board record to be "the same or substantially the same" as.

Estoppel landscape: essentially empty, and that is the practical point. Because no petitioner ever filed, § 315(e)(2) estoppel attaches to nobody. There is no defensive aggregator (Unified Patents, RPX, etc.) in this patent's challenge chain — only in its ownership chain. The 1999 Sanyo→Foltra→Callahan→Intellectual Ventures Assets 158→Hanger Solutions assignment sequence (2016-01-14, 2020-01-28, 2020-01-04) is an IV-style monetization chain, and current assignee is Hanger Solutions LLC. That a patent sitting in a monetization portfolio for years attracted no IPR is itself informative — it tends to mean either (a) the patent was never asserted in a suit where an IPR would be worth the filing fee, or (b) it was asserted only against targets who chose to license rather than fight. Either way, that is a factual question you should confirm, because it materially changes the risk read: a licensed-out portfolio patent is a very different beast from one that beat back challengers.

The expiration point is the dominant strategic fact and I want to be careful with it. Per the structured data the patent reached "Anticipated expiration" on 2017-09-02 (20 years from the 1997-09-02 filing). If that is correct — and pre-AIA § 154 terms run 20 years from the earliest U.S. filing date absent a terminal disclaimer of the § 120 chain — the '685 is expired. I did not independently verify the expiration with the USPTO Patent Center fee/expiration screen in this session, so treat the date as sourced from the ODP-derived block rather than confirmed by me. Confirm it before relying on it. If expired, that has these consequences: past damages only within the § 286 six-year lookback of the suit filing; no injunctive relief; and IPR would be of dubious utility because the Board generally requires that a challenged patent be eligible for IPR but the practical value of canceling an expired claim is limited to collateral/troll-defense purposes.

Recommended next steps

1. Confirm the zero-proceeding finding yourself in PTAB E2E / Patent Center before you commit it to a client memo. Search by patent number 5,999,685 and by application 08/921,606 (the trailing "606" continuation of the same application number is a common transcription trap). If you want a second lens, the PTAB "Patent Trial and Appeal Board End-to-End" search and the USPTO PTAB Decisions page are the canonical sources; I was unable to render them directly here, which is the main gap in this analysis.
2. Pull the file wrapper for application 08/921,606 and check for ex parte reexamination, reissue, or certificate-of-correction activity, and for any terminal disclaimer or § 154(b) patent-term adjustment that moves the expiration date off 2017-09-02.
3. Check the litigation docket before you assume no assertion. Hanger Solutions is the current assignee (assignment recorded 2020-01-04). If Hanger (or a predecessor IV entity) asserted the '685 in a district-court complaint, there will be a case number and possibly a covenant or settlement you can use. If there is no suit, no PTAB activity plus no litigation is a strong sign of a paper-only portfolio asset and a demand letter may be low-yield for the sender.
4. If you are currently accused: the honest answer is that there is no PTAB shortcut to hand you. Your leverage is (i) the expiration/limitations defense if the asserted acts predate 2011-09-27 or the six-year § 286 window, and (ii) an IPR petition on the merits if the patent turns out not to be expired — the Enplas '455 and '549 references and Tosoh '590, plus the cited JP 7-120752, are natural starting points, and because no estoppel exists you are free to use them.
5. Do not cite a non-existent proceeding. If opposing counsel or a vendor's AI tool asserts "the IPR on 5999685," ask for the proceeding number and the PTAB paper link. My checked evidence indicates the only proceedings bearing numbers near "5999685" involve the unrelated patents listed in the false-positives section above.

Confidence and limits: I am highly confident there is no PTAB proceeding in the ODP-derived structured data provided (that is stated in the task block itself). From web search I found no IPR/PGR/CBM on this patent and examined and rejected several confusable hits. My residual uncertainty is limited to (a) whether the ODP ingest is current, and (b) whether any 2013–2020 CBM petition existed that predates the database's coverage — though a light-guide-plate patent on a 1997 priority date would have faced a threshold eligibility fight for CBM as a "technological invention," making such a filing unlikely. I found no FWD, no institution decision, no settlement, and no CAFC appeal, and I have not invented any.

I'll research the assignment chain for US 5,999,685, focusing on reel/frame numbers and correspondents of record.

Let me try more targeted searches for the assignment records and the NPE entities.

Assignment record — US Patent 5,999,685

Important sourcing caveat up front. I could not obtain USPTO reel/frame numbers or the correspondent-of-record fields from the tooling available to me. The Google Patents "legal events" feed reproduced in the authoritative patent text mirrors the USPTO reassignment records (it reports the same five reassignment events, with assignor/assignee/conveyance), but it does not expose reel/frame or correspondent. I therefore report the five events as documented and flag every field I could not verify rather than inventing it. To close the gap, query the USPTO Assignment Center by patent number (below) and capture reel/frame + correspondent for each of the five events — the correspondent column is exactly where the repeat-player pattern would show.

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/
• Legacy search UI: https://assignment.uspto.gov/patent/index.html (search "5999685")

Inventors

Seven named inventors, all of whom assigned their rights to the original assignee (see the 1997-09-02 event below), which is the basis for treating them as Sanyo personnel at filing:

Notes / limits:

• The assignment instrument names all seven as assignors to Sanyo, so Sanyo employment is a documented inference, not a guess. Their specific division (Sanyo's display/backlight vs. semiconductor operations) is not determinable from the sources reviewed.
• Departure-pattern check: unclear. I found no evidence that any inventor left Sanyo within 12 months of filing, and no evidence of the opposite either. Because the portfolio stayed with Sanyo for ~12.5 years (1997 → 2009), there is no "inventors bailed before the fire-sale" signature here; the divestiture, when it came, was corporate, not inventor-driven.
• The priority chain is JP 9-025652 (1997-02-07), JP 9-099122 (1997-04-16, pub. JPH10282496A) and JP 9-151180 (1997-06-09, pub. JPH10339815A); US filing 1997-09-02; issued 1999-12-07.

Original assignee

SANYO ELECTRIC CO., LTD. (Osaka, Japan) — the entity named on the issued patent.

• Primary line of business: diversified Japanese electronics manufacturer: LCDs and display modules, optical pickups, batteries, semiconductors, home appliances and consumer electronics. The '685 patent sits squarely in Sanyo's LCD backlight / light-guide activity.
• Did they ship a product embodying the claims? The patent is directed to a light-guide plate and side-light surface light source for LCD backlights — a component class Sanyo manufactured in the relevant period. I could not document a specific Sanyo product model embodying the '685 claims in the sources reviewed; treat "shipped an embodying product" as likely but unverified. (Related Sanyo light-guide / surface-light-source patents — e.g., US 6,679,613, "Surface light source device," assigned to Sanyo Electric — show the company was active in exactly this component space.)
• Current status: Sanyo Electric was absorbed into Panasonic. Panasonic's tender for a majority stake completed in the 2008–2009 window, with full integration following; the Sanyo brand was largely wound down, and Sanyo's semiconductor business was sold to ON Semiconductor (2011). Sanyo Electric Co., Ltd. survives as a Panasonic subsidiary, but it is no longer an independent operating vendor of the kind that would assert this patent.
• The 2009 divestiture to Foltora Service LLC (below) therefore coincides with the Panasonic acquisition period — a classic "portfolio monetization during a corporate integration" moment.

Assignment timeline

Chronological list of the five reassignment events recorded against US 5,999,685. Reel/frame and correspondent are marked [NOT RETRIEVED] for each — I will not fabricate them.

1997-09-02 (executed) / recorded ~1997-09-02 — Reel [NOT RETRIEVED]

• Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
• Assignor: Youichiro Goto; Michiaki Sato; Yasuko Teragaki; Katsutoshi Hibino; Kenji Torazawa; Masahiro Higuchi; Takanari Kusafuka
• Assignee: SANYO ELECTRIC CO., LTD.
• Correspondent: [NOT RETRIEVED]
• Context: Original employment assignment — inventors convey to their employer at filing; the pre-divestiture baseline.

2009-09-17 / recorded 2009-09-17 — Reel [NOT RETRIEVED]

• Conveyance: Assignment
• Assignor: SANYO ELECTRIC CO., LTD.
• Assignee: FOLTORA SERVICE LIMITED LIABILITY COMPANY
• Correspondent: [NOT RETRIEVED] — this is the single most important field to capture, since it would confirm whether the recording was handled by IV-side counsel.
• Context: Portfolio divestiture — Sanyo moves the asset out during the Panasonic integration. Because Foltora later merges into Callahan Cellular L.L.C. (an Intellectual Ventures entity, per its Delaware disclosures naming Corporate Parent Invention Investment Fund II, LLC), the Sanyo→Foltora step is best read as the on-ramp into the Intellectual Ventures family, not a Sanyo internal reorg. Flag: the IV linkage of Foltora is an inference from the 2016 merger; verify with reel/frame + correspondent.

2016-01-14 / recorded 2016-01-14 — Reel [NOT RETRIEVED]

• Conveyance: Merger
• Assignor: FOLTORA SERVICE LIMITED LIABILITY COMPANY
• Assignee: CALLAHAN CELLULAR L.L.C.
• Correspondent: [NOT RETRIEVED]
• Context: Internal reorg within the IV family — Foltora is merged into Callahan Cellular. Callahan Cellular is documented in litigation as an IV holding entity (Corporate Parent: Invention Investment Fund II, LLC), i.e., an IV affiliate, not an operating company.

2020-01-28 / recorded 2020-01-28 — Reel [NOT RETRIEVED]

• Conveyance: Assignment
• Assignor: CALLAHAN CELLULAR L.L.C.
• Assignee: INTELLECTUAL VENTURES ASSETS 158 LLC
• Correspondent: [NOT RETRIEVED]
• Context: Transfer-to-asserter staging — asset moved into an IV "Assets" vehicle as part of the broad IV divestiture wave to IPInvestments-affiliated Georgia LLCs (publicly reported as November–December 2019 transfers; recorded here in January 2020).

2020-01-04 / recorded 2020-01-04 — Reel [NOT RETRIEVED]

• Conveyance: Assignment
• Assignor: INTELLECTUAL VENTURES ASSETS 158 LLC
• Assignee: HANGER SOLUTIONS, LLC
• Correspondent: [NOT RETRIEVED]
• Context: Transfer to a known monetization/asserting NPE — Hanger Solutions, a Georgia LLC associated with Atlanta monetization firm IPInvestments Group LLC, received thousands of IV assets in the 2019–2020 divestiture and has since filed infringement actions (e.g., Hanger Solutions, LLC v. Extreme Networks, Inc., D. Del. 1:21-cv-00040). Hanger is reported to be represented in litigation by Stamoulis and Weinblatt LLC (litigation counsel — do not conflate with the assignment correspondent of record).

Data anomaly to flag (not a contradiction of the chain, but a quirk in the source): Google Patents lists the Hanger event at 2020-01-04 and the IV Assets 158 event at 2020-01-28, i.e., the downstream transfer appears earlier than the upstream one. This is almost certainly a recorded-vs-executed date mismatch on one of the two entries (the two events are within the same month and are mutually dependent). The reel/frame pull will resolve which date is which; the substantive chain Callahan → IV Assets 158 → Hanger is not in doubt.

Timeline diagram

timeline
    title Ownership of US 5999685
    1997 : Filed by Sanyo Electric
         : Inventors assign to Sanyo
    1999 : Patent issued
    2009 : Assigned to Foltora Service LLC
    2016 : Merged into Callahan Cellular LLC
    2020 : Moved to IV Assets 158 LLC
         : Divested to Hanger Solutions LLC

NPE / troll-pattern signals

1. Shell-entity transfer — present. The chain leaves an operating manufacturer (SANYO ELECTRIC CO., LTD.) and passes through a naming sequence of holding vehicles: FOLTORA SERVICE LIMITED LIABILITY COMPANY (2009-09-17) → CALLAHAN CELLULAR L.L.C. (2016-01-14) → INTELLECTUAL VENTURES ASSETS 158 LLC (2020-01-28) → HANGER SOLUTIONS, LLC (2020-01-04). The "Assets" and "Solutions" suffixes and the IV-affiliate nature of Callahan Cellular (Corporate Parent: Invention Investment Fund II, LLC, per its Delaware corporate disclosures) satisfy the naming-plus-context test rather than naming alone. Caveat: no products-in-commerce evidence for these entities, and registered-agent addresses were not retrieved — corroborate with the correspondent/address fields.

2. Known asserter in the chain — present. Two links independently match public NPE rosters: Intellectual Ventures (via Callahan Cellular and IV Assets 158) and Hanger Solutions, LLC, which is documented by RPX/Mondaq as an IPInvestments Group monetization entity that acquired thousands of IV assets in late 2019 and has sued (D. Del. 1:21-cv-00040 against Extreme Networks). This is the strongest single signal in the chain.

3. Repeat correspondent across the chain — unclear (not retrievable). Every correspondent field returned [NOT RETRIEVED]. This is the one signal that could materially strengthen or weaken the case, and it is the specific reason to pull the assignment-center records: if the same attorney/firm recorded Foltora, Callahan, IV Assets 158 and Hanger, that is a textbook repeat-player tell. I decline to name any correspondent without the reel/frame.

4. Cascading transfers — present. Callahan Cellular → IV Assets 158 LLC → Hanger Solutions LLC are three consecutive links inside a single month (January 2020), and the broader chain shows 4 hops across 2009/2016/2020. The compressed January 2020 pairing (same Georgia-asserter ecosystem, back-to-back recordings) is the classic "stage then hand to the plaintiff" cascade.

5. Pre-litigation transfer — not present (as to this patent). Consistent with the earlier litigation section, I found no infringement suit naming the '685 patent, so there is no assignment-dated-within-6-months-before-suit event to point to. The 2020 transfer fits Hanger's general campaign window, but this specific patent has not (per available sources) been asserted.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 of Sanyo. Sanyo was acquired (Panasonic), and the 2009 transfer reads as an integration-era portfolio divestiture, not a bankruptcy sale. No Kodak/Nortel/Polaroid-style court-supervised sale appears in the record.

7. Privateering — not present. Privateering requires the operating company to direct an NPE's assertion on its own behalf against competitors. Sanyo/Panasonic is not documented as directing Hanger's or IV's assertion activity, and there is no SEC 10-K/8-K reference to such an arrangement in the sources reviewed. The transfer to IV looks like outright monetization, not agency.

8. Defensive aggregator — not present. The chain terminates at an asserting NPE (Hanger Solutions), not at RPX, AST, LOT Network, Unified Patents, or OIN. This is the inverse of neutralization: the asset ended in a litigation-oriented vehicle.

Verdict

NPE — high confidence.

The chain runs from an operating manufacturer (SANYO ELECTRIC CO., LTD.) through a sequence of holding/monetization vehicles — Foltora Service LLC (2009-09-17), Callahan Cellular L.L.C. (2016-01-14, merger, an IV affiliate with Corporate Parent Invention Investment Fund II, LLC), Intellectual Ventures Assets 158 LLC (2020-01-28) — and terminates at Hanger Solutions, LLC (2020-01-04), a publicly documented IPInvestments Group monetization entity that acquired thousands of IV assets and has filed infringement suits. That is two independent strong signals (known asserter in the chain + shell/asserter staging) plus a compressed three-hop cascade inside January 2020, i.e., well over the "2+ strong signals" threshold. The verdict speaks to ownership posture only: I found no litigation asserting the '685 patent, and the patent expired 2017-09-02, so this is an NPE-held-but-apparently-unasserted asset.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 5,999,685) — capture reel/frame and the correspondent for each of the five events to convert signals #1 and #3 from "present/inferred" and "unclear" to documented findings.

I'll build on the litigation/assignment work already done and focus on the prior-art citations for US 5,999,685. Let me verify the cited references.

I have strong detail on the most technically relevant references. Let me verify the older cited references and the Japanese non-patent citation.

Prior-Art Analysis — US Patent 5,999,685

1. Confirmation of the patent searched (literal ID check)

The fetch of the USPTO‑mirrored record for the exact number 5,999,685 confirms:

This is consistent with the earlier litigation section (Hanger Solutions; expired 2017‑09‑02). No contradiction with the prior section.

Method note: I retrieved the citation table and full text via the Google Patents mirror of the USPTO record, and pulled the specification/claims of the key references. I retrieved detailed text for US 5,600,455, US 5,575,549, US 5,735,590, US 5,854,872 and US 1,707,965, but my tooling hit its step limit before I could retrieve full text for US 2,347,665, US 3,328,570, US 5,408,344, US 5,420,761, or the JP 7‑120752 abstract. Those entries below are grounded on the citation table and titles only and are flagged accordingly. None of the date-sensitive facts were auto-corrected.

2. The claim set being tested (independent claims)

• Claim 1 — resin light guide plate; a first lens portion of recessed/projected lenticular unit lenses formed of the same resin as the plate on the light‑emission surface; triangular prism portions with vertex angle 125°–165°; ridge lines nearly parallel.
• Claim 6 — surface light source = claim‑1 plate + edge light source.
• Claim 8 — claim‑1 plate + a second lens portion (same‑resin triangular prisms, 125°–165°, parallel ridges) on the reflective surface opposite the emission surface.
• Claim 15 — surface light source with both first and second lens portions.
• Claim 16 — groove‑based plate: grooves of approximately equal width on the surface opposite the emission surface, parallel, spacing narrowing with distance from the incidence surface, the nearest‑to‑source groove spacing being 5–15× the groove width.
• Claim 19 — surface light source with the claim‑16 groove plate.
• Dependent: 2–5, 7, 9–14, 17 (N‑gon/isosceles/part‑ellipse groove cross‑section), 18 (wedge), 20 (wedge).

Critical framing: every reference below is an examiner‑cited reference ("*" in the record). Because the patent issued, the examiner did not find that any of them anticipates any claim as allowed. The two limitations that most clearly defeat § 102 identity are (i) the recited vertex angle of 125°–165° and (ii) the "same resin / formed simultaneously with the plate" integrality of the lens portion.

3. Examiner‑cited references (9 patents + 1 NPL)

3.1 U.S. Patent No. 1,707,965 A — Scantlebury, "Illuminated Sign"

• Dates: filed/priority 1928‑11‑03; granted 1929‑04‑02.
• Description: Multi‑plate edge‑lit etched‑glass sign; light entering the polished plate edges is internally reflected; an opaque backing plate 25 prevents rear light penetration. A foundational edge‑lit light‑guide disclosure.
• Potential § 102 (pre‑AIA § 102(b)) mapping: None of claims 1–20. It contains no prismatic lens portion and no grooves. Relevant only as background for the edge‑lit guide + reflective backing concept.

3.2 U.S. Patent No. 2,347,665 A — Bandy, "Internal Reflection Lighting Means"

• Dates: priority 1941‑03‑04; granted 1944‑05‑02.
• Description (from title/record only — full text not retrieved): Edge‑lit panel exploiting total internal reflection. Background art.
• Potential § 102 (§ 102(b)) mapping: None of claims 1–20.

3.3 U.S. Patent No. 3,328,570 A — General Electric, "Illuminated Panel Member"

• Dates: priority 1965‑04‑09; granted 1967‑06‑27.
• Description (from title/record only): Large‑area edge‑illuminated panel. Background art.
• Potential § 102 (§ 102(b)) mapping: None of claims 1–20.

3.4 U.S. Patent No. 5,408,344 A — Ricoh Company, Inc., "Reflection Type Liquid Crystal Display with Selective Reflecting Means"

• Dates: priority 1993‑01‑11; granted 1995‑04‑18.
• Description (from title/record only): Reflective LCD employing a selective reflector. Background for the display‑backlight context.
• Potential § 102 mapping: None of the structural claims; at most generic context for surface‑light‑source claims 6/15/19, without the lens/groove limitations.

3.5 U.S. Patent No. 5,420,761 A — Precision Lamp, Inc., "Flat, Thin, Uniform Thickness Large Area Light Source"

• Dates: priority 1993‑03‑29; granted 1995‑05‑30.
• Description (from title/record only): Edge‑lit flat light source. Background for the light‑source half of claims 6/15/19.
• Potential § 102 mapping: None of claims 1–20 as a whole.

3.6 U.S. Patent No. 5,575,549 A — Enplas Corporation, "Surface Light Source Device"

• Dates: priority 1994‑08‑12; granted 1996‑11‑19. § 102(e) date = its U.S. filing date (1994).
• Description (retrieved): Light‑conducting member with a multiplicity of fine conical or polygonal pyramid‑shaped concave portions on the surface opposite the light‑emitting surface; "top angles and inclined angles … are set within a given range" to obtain a uniform brightness distribution. Also discloses filling/covering the concave portions with diffusing material.
• Closest claims: 8 and 15 (a structured portion on the reflective surface behind the emission surface) and 17 (N‑sided‑polygon / conical cross‑section). However, '549 discloses discrete spaced pyramid/cone depressions, not continuous parallel V‑grooves of approximately equal width with monotonically decreasing spacing, and no 125°–165° triangular‑prism vertex or same‑resin integral lens portion. Not a full § 102 anticipation; strong § 103 art for the reverse‑surface construction. (The exact numeric "top‑angle" range of '549 was not recovered — I will not state it.)

3.7 U.S. Patent No. 5,600,455 A — Enplas (Ishikawa et al.), "Prismatic Member with Coarsened Portions or Triangular Prismatic and Semi‑Circular Prismatic Members…"

• Dates: filed 1994‑08‑31 (Appl. 08/297,956); granted 1997‑02‑04. § 102(e) date = 1994‑08‑31; also § 102(a) as of grant.
• Description (retrieved): A transparent prismatic member (sheet or plate) with a multiplicity of convex portions having triangular cross‑sections, extending parallel to each other, placed on the light‑emitting surface; claim 7 covers two prism members superimposed with their longitudinal (ridge) directions intersecting (≈ perpendicular).
• Closest claims: 1, 6, 8, 10 and 15 — this is the closest structural analog to the "triangular prisms with parallel ridge lines on/behind the plate, two sets crossed" idea.
• Why it does not anticipate: (i) the prismatic member is a separate sheet/plate, not a lens portion "formed of the same resin as that of the light guide plate"; (ii) '455 claim 6 expressly recites an equilateral triangle cross‑section (60° vertex), i.e., outside the claimed 125°–165° and in fact contrary to it. Excellent § 103 art; not § 102.

3.8 U.S. Patent No. 5,735,590 A — Tosoh Corporation (Kashima et al.), "Backlighting Device with a Transparent Sheet Having Straight Ridges"

• Dates: JP priority 1994‑03‑02; U.S. filed 1995‑03‑01 (Appl. 08/396,529); granted 1998‑04‑07. § 102(e) date = 1995‑03‑01 (foreign priority does not carry the § 102(e) date — In re Hilmer).
• Description (retrieved): A light‑conducting plate whose thickness decreases away from the source (wedge); a transparent sheet with a number of prisms/protrusions having straight ridges, parallel, at small (10–1,000 µm) intervals, on the light‑exit side, with ridges oriented perpendicular to the source axis to suppress striping. Discloses an apex angle preferably 70°–150° (more preferably 90°–110° for high‑index material), and cross‑sections that may be triangle, part‑circle, part‑ellipse or wave‑shape.
• Closest claims: 1, 6, 8, 10, 15 (parallel‑ridged prism on the exit surface, crossed/orthogonal arrangement of two sheets) and, because of the wedge plate, 18 and 20 (wedge shape).
• Why it does not anticipate: the ridges are on a separate transparent sheet, not an integral same‑resin lens portion; the light‑extraction features are printed/embossed light‑diffusing elements, not the equal‑width, monotonically‑narrowing grooves of claim 16; and the exemplified apex angles (90°–110°) fall outside 125°–165°, with only the outer bound (150°) overlapping. Strong § 103 art against claims 1/8/15/18/20; not § 102.

3.9 U.S. Patent No. 5,854,872 A — Clio Technologies, Inc. (Tai), "Divergent Angle Rotator System and Method for Collimating Light Beams"

• Dates: filed 1996‑10‑08; granted 1998‑12‑29. § 102(e) date = 1996‑10‑08 (i.e., prior to the '685 priority of 1997‑02‑07 even though the patent published later).
• Description (retrieved): A light pipe/film with elongated microprisms (and microgrooves) having vertices; microprisms on the top surface run in the mean propagation direction, and claim 12 recites bottom‑surface microprisms for directing light out of the top surface.
• Closest claims: 17 (cross‑sectional shape being an N‑sided polygon / part of an ellipse) and 8/15 (microprism structure on the surface opposite the emission surface).
• Why it does not anticipate: its structures serve two‑dimensional collimation (a different function/geometry), and it teaches no 125°–165° triangular‑prism vertex and no same‑resin integral lens portion. § 102(e) reference; § 103 art at most.

3.10 Non‑patent citation — "Abstract of Japanese Patent Publ. No. 7120752, dated May 12, 1995"

• Full citation: Japanese Patent Publication No. 7‑120752 (JP‑A‑H07‑120752), published 1995‑05‑12.
• Description: Content not retrieved; the record cites the published abstract. As a printed publication dated 1995‑05‑12 it falls under pre‑AIA § 102(a)/(b).
• Potential § 102 mapping: Cannot be mapped to specific claims without the reference's content. Flagged as unresolved — this is the one examiner citation I could not substantiate.

4. Mapping summary

5. Additional prior art cited in the '685 specification (not in the examiner table)

Worth adding to any invalidity search, because the '685 disclosure itself characterizes them as prior art:

• U.S. Patent No. 4,729,067 (Ohe) — mat transparent diffusing plate (cited by the '685 background and also listed in the '549 reference list).
• Japanese Utility Model Laid‑Open No. 162201/1980 — side‑light LGP with scattering reflective plate (the '685 background art).
• Japanese Utility Model Laid‑Open No. 107201/1992 — triangular‑prism lenticular lens sheet over an LGP.
• Monthly Display, May 1996, pp. 35–39 — the two crossed lenticular sheets arrangement (relevant § 102(b) printed publication).
• IEICE Technical Report EID98‑79, pp. 67–72 — post‑dates the 1997‑02‑07 priority (1998), so it is not § 102 prior art to '685; treat it only as corroborating the state of the art.

Forward citations (e.g., US 6,075,649; US 6,155,692; US 6,560,455‑family members; the 126/251 “Cited By” entries) are not prior art to '685 and were excluded.

6. Caveats

• The § 102 labels above use pre‑AIA law (the '685 application was filed 1997‑09‑02 and granted 1999‑12‑07), where an issued U.S. patent's § 102(e) date is its U.S. filing date, not its foreign priority date (In re Hilmer).
• I could not retrieve the substantive content of US 2,347,665, US 3,328,570, US 5,408,344, US 5,420,761, or JP 7‑120752; those are stated at the level the record supports and should be verified against full text before any serious invalidity contention.
• No reference here establishes anticipation of any claim as allowed; the 125°–165° vertex angle and the same‑resin integral lens portion are the features that separate the claims from the art. If you want, I can next run the full‑text pull on the five unretrieved references and test the § 103 combinations (e.g., US 5,600,455 or US 5,735,590 in view of US 5,575,549 for claims 8/15, or US 5,575,549 in view of printed‑publication groove art for claim 16).

I'll ground the obviousness analysis in the actual prior-art references of record. Let me pull details on the key citations.

Obviousness Analysis — U.S. Patent 5,999,685 under 35 U.S.C. § 103

Scope note. This builds on the earlier summary and litigation sections (not repeated). Per your instruction I use the Prior Art section of the patent page itself — i.e., the nine U.S. patent citations of record (US 1,707,965; 2,347,665; 3,328,570; 5,408,344; 5,420,761; 5,575,549; 5,600,455; 5,735,590; 5,854,872), the single Non-Patent Citation ("Abstract of Japanese Patent Publ. No. 7120752, dated May 12, 1995"), and the references admitted as prior art in the specification's own Background. Where I could retrieve and verify the substance of a reference, I say so and cite it; where I could not, I say so and do not rely on it for a dispositive disclosure. I did not fabricate any reference content.

Verification status of the references I rely on

Governing law. The application was filed Sept. 2, 1997; pre-AIA §§ 102/103 apply. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) governs the §103 analysis.

1. The person of ordinary skill in the art (POSITA)

A POSITA here would hold a bachelor's degree in optics, physics, electrical or mechanical engineering (or equivalent), with ~2–3 years' experience designing edge-lit (side-light) backlights and light guide plates for LCDs, and would be familiar with: (i) wedge vs. parallel light guides; (ii) back-surface extraction features (printed diffusing dots, molded micro-structures); (iii) prism/lenticular sheets placed on the emission surface; (iv) Snell's-law-based design of prism apex angles; and (v) commercial optical-design tools (the '685 itself used "CODE-V"). This is a mature, crowded, incremental art — exactly the setting in which KSR presumes combinability.

2. Key claim terms (construction, for the §103 analysis)

• "formed of the same resin as that of the light guide plate" — the lens portion is integrally/unitarially molded with the plate, not a separately laminated film. This is the only structural feature distinguishing the claimed plate from a guide + applied prism sheet.
• "lenticular unit lenses … triangular prism portions whose … vertexes have an angle of 125° to 165°" — elongated prism elements with a defined apex angle; per the '685 spec (FIGS. 2B/2C) the apex may be truncated or rounded, so the "angle" is the angle the flanking facets would define.
• "nearly parallel" ridge lines; "nearly perpendicular" for claim 10 (the spec and art treat ~90° ±tolerance).
• Claim 16's "spacing … is 5 to 15 times said groove width" — a numeric ratio measured from the far end of the groove nearest the incidence surface to the near end of the next groove.

3. Claim 1 — element-by-element against the closest art

The two gaps are (a) integral molding in the same resin, and (b) apex angle ≥125°. Neither gap is inventive on this record, for the reasons in §5–6 below.

4. Claim 8 / 15 — second lens portion on the reflective surface

• US 5,600,455, claim 7 & FIG. 10: a first prism member (triangular, "cornered tops") and a second prism member (round tops) superimposed so their longitudinal directions intersect.
• US 5,735,590 (Examples 1–4): two ridged transparent sheets with ridges crossing at ~90° — best shielding and highest luminance (2,200 cd/m² vs. 1,800 cd/m² for one sheet).
• Admitted prior art (Monthly Display, May 1996): two lenticular lens sheets overlapped at right angles over a light guide. The '685 specification concedes this was the conventional way to get high luminance.
• Back-surface prismatic structure: US 5,575,549 (pyramid concave portions on the surface opposite the exit face); JP H7-120752 (fine structure on the bottom surface reflecting light to the top); US 5,854,872 cl. 12 & 25 (bottom-surface microprisms "for directing light out of said top surface").

Claim 8's only delta over the admitted two-sheet art is (i) putting the second lenticular structure on the reflective surface of the plate itself rather than as a second stacked sheet, and (ii) integral same-resin molding. Both are taught by the back-surface-structure references, and both are motivated by the very cost/alignment problem the '685 specification identifies.

5. Claim 16 (and 17–20) — graded-pitch grooves

This is the closest to being squarely met by the examiner's own NPL citation. JP H7-120752 discloses, in a wedge light guide with a side light source: a fine structure on the bottom (reflective) surface that reflects propagated light toward the top surface, a pitch of 0.1–2 mm, and light diffusivity that varies with distance from the light source to obtain uniformity and desired directionality. US 5,575,549 supplies the "vary the back-surface features by area to flatten the luminance profile" motivation and the isosceles/V geometry; US 5,735,590 supplies the cross-section menu (triangle / part of circle / part of ellipse / wave — claim 17); the wedge is admitted and disclosed throughout.

What JP H7-120752 does not expressly state: "approximately equal groove widths" with the spacing (not the width) shrinking, measured far-end→near-end, at 5–15× the groove width. That numeric limitation is the only real hook for claim 16.

6. Motivation to combine (KSR rationales)

1. Same field, same problem, same solution space. All of Enplas '455/'549, Tosoh '590, Clio '872 and JP H7-120752 are edge-lit LCD backlights, addressing luminance uniformity and on-axis brightness. They are a fortiori analogous art.
2. Known technique to improve a similar device in the same way (KSR factor C). Applying a parallel-ridge prism surface to a light guide's exit face to gain directivity was a known technique; applying a micro-prism structure to the guide's back face to redirect/equalize light was a known technique (Enplas '549, JP H7-120752, Clio '872). Doing both on the same plate is the predictable combination.
3. Eliminating parts/alignment cost is an articulated, art-recognized engineering driver. Tosoh expressly criticizes prior configurations for being thicker and "costly" and for requiring extra diffusing sheets; Enplas '549 criticizes the printing/coarse-surface route as impractical. The '685 specification's own stated objects (reduce from two lens sheets to one, or zero; eliminate dot printing) are the same objects already motivating the art.
4. Integral molding is a known manufacturing expedient ready for this improvement (KSR factor D). Molding surface relief directly into a light guide (Enplas '549 cones; JP H7-120752 pitch structure; the '685's own back-surface grooves) was routine. Once the lens is molded into the plate, it is necessarily "formed of the same resin."
5. Apex angle is a result-effective variable, and variation is predictable (KSR factors A/B; "obvious to try"). Tosoh states the optimum apex angle depends on refractive index and emission profile and gives a 70°–150° envelope. Enplas '549 claims a top-angle range for the same class of structure. Snell's law makes the apex/ray-deflection relation analytically predictable, and the '685 obtained its range by running routine optical simulation (CODE-V, 600 nm, PMMA n=1.49). Under KSR, "when there is a design need … and there are a finite number of identified, predictable solutions, a person of ordinary skill has good reason to pursue the known options within his or her technical grasp."
6. The ratio in claim 16 is an "optimization of a result-effective variable." The specification's own FIG. 12 shows improvement across a broad plateau (5×, 10×, 15× all good; only 2.5× and 20× poor). A broad plateau obtained by sweeping a ratio is the paradigm of routine optimization and undercuts a "criticality" argument.

Consolidated grounds

7. Applicant's best counterarguments (and their weight)

1. Teaching away (Graham factor 4). The '685 asserts the art had a "fixed idea that the vertical angle is in the range of 90° to 120°," and Enplas '549 in fact specifies 80°–100° and Tosoh's preferred value for high-index material is 90°–110°. That is real evidence of a directional preference below 125°. Weight: moderate. It is blunted because Tosoh's broad disclosure reaches 150°, so the 125–150 sub-range is not "taught away" — it is disclosed. The teaching-away argument is strongest for the ~150° point of claim 3 (and 12/13) and weakest for the 125–135° end of claim 1.
2. Unexpected results. The FIG. 4/5 simulations show 90°–120° performing worse than flat while 125°–165° improves. Weight: moderate but fragile. The evidence is a single-wavelength (600 nm), single-material (PMMA) simulation generated by the applicant; there is no comparative commercial data, no copying evidence, and no independent verification in the record. Under In re Geisler/KSR, attorney-generated simulation of a monotonic trend rarely establishes unexpectedness for a broad numeric range.
3. Criticality of the 5–15× range (claim 16). The spec reports peak performance at 5×/10×/15× and poor results at 2.5×/20×. Weight: weak. A three-point-wide "good" plateau bounded by smooth degradation is routine optimization, not criticality; and JP H7-120752 already discloses sweeping the back-surface feature pitch/diffusivity with distance.
4. Integral same-resin molding. Applicant would argue the art used separate sheets. Weight: weak, given Enplas '549 and JP H7-120752 mold their extraction structures directly into the guide, and the art expressly motivates eliminating sheets to cut cost/thickness.

8. Claim-by-claim vulnerability summary

Bottom line: On the art of record, the strongest §103 case is claims 1, 4, 5, 6, 7, 8, 10, 14, 15, 17–20, grounded in US 5,600,455 (parallel tri-prism lens on the emission surface) in view of US 5,735,590 (apex-angle envelope 70°–150°, crossed-ridge two-sheet arrangement, cross-section menu, wedge) and further in view of US 5,575,549 and JP H7-120752 (molded back-surface extraction structures varied by distance from the source). Claims 2, 3, 9, 11, 12, 13 and claim 16 present the only colorable nonobviousness positions — the first on teaching-away from the art's preferred ≤120° apex angles, the second on the claimed criticality of the 5–15× width-to-spacing ratio — and both positions rest almost entirely on the applicant's own unverified CODE-V simulations.

9. Caveats and next steps

• Unverified references. I could not retrieve the operative text of US 5,420,761, US 5,408,344, US 1,707,965, US 2,347,665, or US 3,328,570 within my search budget. Each is §102(b) art of record; if any discloses a graded back-surface feature pattern or a prism-apex range, Grounds A/C/D get stronger. Retrieve these before relying on this analysis.
• JP H7-120752 is the single most important reference for claims 16–20 and I have only the published abstract (via J-GLOBAL). The full text/figures should be pulled (JP A 7-120752, 特願平5-264540) to determine whether the "fine structure" is disclosed as grooves of equal width with varying pitch — which would largely dispose of claim 16.
• Do not use the IEICE EID98-79 report in any §103 ground; it post-dates the Feb 7, 1997 priority.
• U.S. Ser. No. 08/742,500 (cross-referenced in the '685) is applicant's own commonly owned application; if it issued, pre-AIA §103(c) common ownership would likely disqualify it as prior art for §103, so I have not relied on it.
• Any invalidity contention is academic as to post-Sept. 2, 2017 conduct (the patent expired then, per the record). The §103 analysis matters only for pre-expiration conduct and for the Family/continuation context noted in the earlier sections.

Explicit uncertainty statement: I have verified the substance only of US 5,600,455, US 5,575,549, US 5,735,590, US 5,854,872, and JP H7-120752 as cited above. For every other reference of record I am relying on the citation itself, not on inspected text, and I have flagged that rather than characterizing the references from memory.