US 7848023

I'll search for this specific patent number and check for any litigation activity.

US Patent 7,848,023 — Summary and Litigation Check

Sourcing note

The authoritative source used here is the full patent document from Google Patents (https://patents.google.com/patent/[US7848023](/patent/US7848023)/en), plus the pre-grant publication US2009/0190228 A1. The two searches I ran for litigation activity returned no record of any court proceeding involving this specific patent — the hits were unrelated items that merely contained the string "7848023" (a Brazilian municipal notice, a Czech contract registry entry, Italian hiking coordinates) and unrelated CAFC decisions. I want to be explicit: I found no CAFC 2026 docket entry, ITC investigation, or district court case asserting U.S. 7,848,023. I cannot rule out an un-indexed or recently filed action, but I have no authoritative basis to report one, so I am not going to invent one.

Bibliographic data

Abstract

An exposure head comprising a lens array having lenses arranged in a first direction and in a second direction orthogonal to the first direction, and a light emitting element substrate provided with light emitting elements that emit light to be imaged by the lenses, wherein the relationship 1 < L₁/L₂ is satisfied, where L₁ is the length of the lens in the first direction and L₂ is the length of the lens in the second direction.

Plain-language overview of the independent claims

The patent has three independent claims (1, 12, and 13), all built on the same single dimensional inequality:

Claim 1 — An exposure head. Two components are required: (a) a lens array whose lenses are arranged in both a first direction and a second direction perpendicular to it (i.e., a two-dimensional lens arrangement, not a single line), and (b) a light emitting element substrate carrying light emitting elements whose light is imaged by those lenses. The only additional limitation is that each lens is longer in the first direction than in the second direction (L₁/L₂ > 1). In the specification the first direction corresponds to the main scanning / longitudinal direction and the second to the sub-scanning / width direction. The stated advantage: the lenses can gather more light in the scanning direction while the pitch between lenses in the width direction stays small — miniaturizing the print head while keeping exposure quality at high resolution, and reducing sensitivity to photosensitive-drum speed variation.

Claim 12 — A lens array per se. A standalone claim to the lens array, with no exposure head or light source: lenses arranged in a first direction and a second direction orthogonal to the first, satisfying the same 1 < L₁/L₂ relationship. This is the broadest claim in the sense that it does not require a light emitting element substrate or any apparatus context.

Claim 13 — An image forming apparatus. Combines (a) the claim-1-type exposure head (lens array + light emitting element substrate) with (b) a latent image carrier (e.g., a photosensitive drum) that is exposed by the exposure head to form a latent image, again requiring 1 < L₁/L₂. This is an electrophotographic system-level claim.

Key dependents worth noting (context on claim scope)

Claims 2–11 depend from claim 1 and progressively narrow it: L₁/L₂ < 1.2 (claim 2, framed as reducing astigmatism from mold shrinkage); a lens substrate carrying the lenses (claim 3); a diaphragm between the light emitting element and the lens (claim 4); diaphragm asymmetry 1 < La₁/La₂ (claim 5); matching lens and diaphragm aspect ratios L₁/L₂ = La₁/La₂ (claim 6); identical lens and diaphragm shapes (claim 7); elliptical diaphragm (claim 8); a convex lens surface facing the light emitting element (claim 9); diaphragm positioned on the image-plane side of the apex of that convex surface (claim 10); and free-form surface lenses (claim 11). These dependents, not the bare independent claims, carry most of the practically litigable specificity.

Uncertainty flags

• Litigation: none found. My searches were designed to surface a CAFC 2026 appeal or a district court/ITC action, and returned nothing on point. Treat this as "no evidence located," not as affirmative proof of a clean litigation history.
• Assignee: Seiko Epson is the assignee of record on the face of the document. Given that Seiko Epson has historically sold off large patent tranches (e.g., to IPValue affiliates), I cannot confirm from these searches whether current ownership still rests with Seiko Epson — the Google Patents listing says Seiko Epson as current assignee, and I am reporting that literally.
• Claim count: the pre-grant publication text I retrieved shows 13 total claims (1–13), consistent with the issued patent's independent claims 1, 12, and 13. I did not retrieve a USPTO-verified copy of the issued claim set, so if exact issued-claim wording matters for a legal purpose, pull the granted patent from USPTO PatentCenter rather than relying on this summary.
• Claim 6/paragraph [0135] typography: the source text renders the matching-ratio formula inconsistently (e.g., "L1/L2=la1/La2" and "I<L1/L2<1.2" appear to be OCR artifacts for "1<L1/L2<1.2" and "La1"). Interpret those literally only after checking the official printed patent.

I'll research litigation involving US patent 7848023. Let me search multiple sources.

The first searches returned mostly unrelated results. Let me search more specifically for this patent.

Let me search additional litigation-specific sources.

Summary: Litigation involving US Patent 7,848,023

I found no evidence of any litigation involving US Patent No. 7,848,023 (the "'023 patent").

I searched for the specific patent number across general web search, litigation-oriented queries, and patent databases (including attempts to reach Unified Patents litigation resources, RPX Insight, PACER-style dockets, and CourtListener/Justia). Every query for the exact number "7,848,023" / "7848023" returned only:

• The patent's own prosecution/publication records (Google Patents, Justia Patents) — no litigation section or "Litigation" entry;
• Coincidental matches to unrelated numeric strings (e.g., a Brazilian state gazette act number, a Czech contract ID, a trademark registration number, GPS coordinates, and a French plumbing part code) — none of which refer to the patent.

No district court docket, CAFC appeal, ITC Section 337 investigation, or PTAB proceeding (IPR/PGR) listing US 7,848,023 as an asserted or challenged patent appeared in any result.

Patent context (for identification purposes only)

• Patent: US 7,848,023 B2 — "Lens array, an exposure head and an image forming apparatus"
• Assignee: Seiko Epson Corporation
• Inventors: Ryuta Koizumi, Yujiro Nomura, Takeshi Sowa
• Appl. No.: 12/357,231; Filed: 2009-01-21; Granted: 2010-12-07
• Priority: 2008-01-25 (JP 2008-14497 / 2008-304814)
• Status: Expired – Fee Related (anticipated expiration 2029-01-21)
• A related continuation/priority family member is US 8,274,741 (from application 12/914,408).

Caveats and limitations

1. Absence of evidence is not proof of absence. I could not complete a systematic sweep of paid databases (PACER docket search, LexMachina, Docket Navigator, RPX Insight, Unified Patents) or run a full-text claim-chart search. My conclusion is based on public, freely indexed sources reached via web search.
2. This patent covers a fairly narrow optical/printing component (elongated lens arrays and exposure heads for electrophotographic image forming), and it expired for failure to pay maintenance fees — both factors that reduce the likelihood of it appearing in active assertion campaigns, which is consistent with the nil result.
3. Any litigation would most plausibly have involved Seiko Epson as plaintiff or as an indemnitor/intervenor (as Epson has done in other printer/LCD suits, e.g., the Honeywell '371 and Ninestar matters), but I found no case naming this specific patent.

Recommended verification steps if a definitive answer is required

• PACER Case Locator / Docket Navigator / LexMachina full-text search for "7,848,023."
• Unified Patents litigation portal and RPX Insight (patent-level "Litigation" tab).
• USPTO Patent Trial and Appeal Board (IPR/PGR/CBM) and USITC EDIS for § 337 investigations.
• Search Epson's and any successor's SEC filings (10-K/10-Q "Legal Proceedings") for the number.

Bottom line: Based on the sources I was able to search, no litigation is known involving US Patent No. 7,848,023. If you need a litigation-grade confirmation, the paid docket databases listed above should be checked directly.

I'll verify the structured data against live sources before drafting.

Let me run additional targeted checks for any AIA trial or related assertion activity.

One more targeted sweep to confirm the negative before I draft.

Proceedings overview

Total AIA trial proceedings on US 7,848,023: zero (0). Breakdown: 0 active / 0 claims invalidated by the Board / 0 claims sustained by the Board / 0 settled / 0 institution denials — the structured USPTO ODP "PTAB proceedings on file" block for this patent is empty, and my independent web checks (2026-09-20) surfaced no IPR, PGR, or CBM petition naming this patent, its application number 12/357,231, its publication US2009/0190228A1, or its continuation sibling US 8,274,741.

Bottom-line defensive posture: This is not a "hardened by IPR" patent and not a "claims already canceled" patent — it is an untested-at-the-PTAB patent whose enforceability is more likely to die on the fee clock than on § 103. Google Patents lists the legal status as "Expired – Fee Related" (anticipated expiration 2029-01-21), meaning the patent appears to have lapsed for failure to pay a maintenance fee well before its nominal term. If that status is accurate, the assertion risk is almost entirely retrospective (pre-lapse damages within the 35 U.S.C. § 286 six-year lookback), not prospective. Also note the framing: the patent owner of record is Seiko Epson Corporation, a large operating company with a 30,000+ patent portfolio — this is not a troll patent, and the "demand letter" posture in the template doesn't map onto the facts.

---

Per-proceeding detail

No per-proceeding entries are provided, because there are no proceedings to report. I am explicitly declining to generate ### IPRxxxx-xxxxx — Petitioner v. Patent Owner headers, because doing so would require inventing proceeding numbers, and the instruction not to fabricate proceeding numbers controls.

What I checked and what came back empty:

Important negative caveat: "No proceedings found" is evidence of absence, not proof of it. My searches are keyword-based and can miss a recently filed petition that ODP has not yet indexed, or a proceeding docketed under an odd caption. If this matters for a live matter, confirm directly against PTAB E2E (https://ptacts.uspto.gov/ptabweb/) by patent number, and against USPTO PatentCenter (https://patentcenter.uspto.gov/) for the maintenance-fee / expiration record — the latter is the single most consequential fact here and I could not verify the exact lapse date from the sources retrieved. I am flagging that as an open item, not a conclusion.

---

Strategic summary

Claim status: every claim (1–13) is UNTESTED at the PTAB. Nothing is canceled, nothing has been confirmed patentable in an AIA trial, and no claim has been construed by the Board. Claim 1 (exposure head), claim 12 (lens array per se), and claim 13 (image forming apparatus) all stand exactly as issued on 2010-12-07. The practical significance of "zero IPRs" cuts both ways: it means no petitioner has yet built an institution-worthy record against these claims, but it also means the broadest claims — which as noted in the earlier summary rest on nothing more than the inequality 1 < L1/L2 — have never been tested against the art. A claim whose only limitation beyond "two-dimensional lens array + light emitters" is an aspect-ratio inequality is unusually vulnerable to a § 103 obviousness attack, and it is telling that in roughly 16 years of existence this patent attracted no challenger. The likeliest explanation is commercial rather than legal: the patent is directed at OLED line-head printers, a narrow product niche, and it lapsed for non-payment.

Estoppel landscape: there is none, and that is the point. Because no IPR or PGR was ever instituted, no petitioner is subject to the 35 U.S.C. § 315(e)(2) estoppel bar, and there is no prior Board record that a current defendant must work around or that a patent owner can point to as a validity win. For a defendant today, the full universe of prior art remains available — both in an IPR and in district court — including the JP-A-6-278314 reference the specification itself identifies as the closest known art, and the pre-grant siblings in this family that claim the inverse relationship (US 2009/0189970, app. 12/358,042, claims 1 < L2/L1), which are worth mining for obviousness-type double-patenting and written-description arguments. Two procedural clocks matter: (a) PGR is unavailable — the nine-month post-grant window closed on or about 2011-09-07; and (b) CBM is unavailable — the patent is not a financial-services business-method patent and the CBM program has sunset. So a defendant's PTAB options today are IPR only, subject to the § 315(b) one-year bar running from service of a complaint.

Pattern signals: none established. No petitioner has filed repeatedly against this patent, because no petitioner has filed at all. No PTAB appeal history exists. There is no defensive aggregator (Unified Patents or similar) in the chain — and none would be expected, since the patent's owner is an operating company and the patent appears to be expired, which removes it from the value proposition of a defensive aggregator. The earlier-generated litigation section independently found no district court, ITC, or CAFC activity, which is consistent with the PTAB finding: a patent that was never litigated is very unlikely to have been IPR'd. The one cross-reference worth carrying forward is the family breadth noted earlier — Seiko Epson holds a dense cluster of line-head/exposure-head patents (e.g., US 7,952,601; US 7,889,222; US 8,300,076; US 8,354,185; US 8,405,700), so a defendant facing assertion of any of them should run the PTAB/fee-status check across the whole family rather than on this patent alone.

---

Recommended next steps

1. Verify the expiration record first — it dominates everything else. Pull the maintenance-fee history for US 7,848,023 in PatentCenter (https://patentcenter.uspto.gov/). Google Patents reports "Expired – Fee Related," which if correct means the patent lapsed for non-payment; the earlier-generated summary's "anticipated expiration 2029-01-21" is the nominal term and does not reflect an early lapse. If the patent is expired, prospective infringement is off the table and the sole remaining exposure is past damages inside the § 286 six-year window — which, for a lapse date several years back, may be a closed window entirely. I could not pin the exact lapse date from the sources retrieved, so treat this as the top verification item.

2. If you nonetheless need an invalidity position, note that the PTAB route is unencumbered: file an IPR within one year of service of any complaint (§ 315(b)). There is no prior Board record to distinguish and no estoppel to navigate. Attack the independent claims on the art the applicant itself cited (JP-A-6-278314) in view of conventional elongated/elliptical lens practice, and check the sibling applications in family ID 40898796 for the mirror-image 1 < L2/L1 claim, which is a useful obviousness and double-patenting exhibit.

3. Do not expect an FWD, a panel, or an appeal to cite. If a colleague or an outside vendor hands you an IPR number for this patent, treat it as fabricated until you confirm it in PTAB E2E (https://ptacts.uspto.gov/ptabweb/) — I found none, and the structured ODP data contains none.

4. Confirm the negative before relying on it in a filing. Where the absence of PTAB activity is itself the argument (e.g., in a § 315(a)/§ 325(d) or discretionary-denial pitch, or in a stay analysis), state it as of a verified date and cite the empty ODP record plus a PatentCenter screenshot, not a web search.

Sources consulted for the negative findings: structured USPTO ODP "PTAB proceedings on file" block (canonical, empty); Google Patents record for US7848023B2 (https://patents.google.com/patent/US7848023/en); Google Patents record for US8274741B2 (https://patents.google.com/patent/US8274741); free patent/publication records for US2009/0190228A1 and US2009/0189970; USPTO PTAB E2E and PTAB Decisions portals (https://ptacts.uspto.gov/ptabweb/, https://www.uspto.gov/ptab/decisions); CourtListener (https://www.courtlistener.com) — all checked 2026-09-20.

I'll research the assignment record for this patent.

Let me try to directly access the assignment records and other sources.

I have important context. Let me dig deeper on the specific assignment records and the Seiko Epson divestiture programs.

I hit the tool-call limit while chasing the Assignment Center entry directly. Below is what I could verify, with explicit gaps flagged. I will not fill gaps with invented reel/frame data.

Inventors

All three inventor addresses fall in Epson's Nagano (Suwa/Owa/Shiojiri/Matsumoto) engineering corridor, which is consistent with all three being Seiko Epson employees at filing and with the inventors-to-Seiko-Epson assignment recorded against the application.

Unusual patterns: None detectable. I found no evidence of any inventor departing Epson within 12 months of the 2009-01-21 filing. Worth noting structurally (not as a red flag): the same Epson team filed a sibling application on the very next day, US 12/358,042, published as US 2009/0189970 ("Lens Array, Exposure Head, and Image Forming Apparatus"), which claimed the inverse inequality 1 < L₂/L₁ and added a fourth inventor (Ken Ikuma, Suwa-shi). So Epson filed two mirror-image dimensional claims one day apart. This matters for any later validity/scope analysis, and I flag it because the earlier generated section's claim mapping for the '023 could be confused with the sibling's 13-claim set.

Original assignee

Seiko Epson Corporation (Tokyo/Shinjuku-ku, JP) is the assignee named on the issued patent and the original assignee of record.

• Primary line of business: printers, projectors, watches/timepieces, semiconductor and electronic devices. Epson is a large, currently operating company (not acquired, not dissolved, not in bankruptcy).
• Product embodying the claims: The disclosure is an organic-EL line head (exposure head) with a two-dimensional lens array for an electrophotographic image forming apparatus; Epson publicly described OLED line-head development and OLED display technology in this period. Whether Epson ever shipped a commercial product practicing 1 < L₁/L₂, I cannot confirm — the spec is directed at a component that may not have reached a shipping SKU. Treat "shipped a product" as unconfirmed.
• Current status: Operating. Critically for this task, Epson has been an active patent seller (see NPE signals below), which is the one live risk to the "still owns it" assumption.

Assignment timeline

Finding, stated plainly: I could not retrieve a reel/frame for any assignment on US 7,848,023. My searches of the Assignment Center surface, third-party mirrors (uspto.report, PlainSite, Google Patents legal events) returned only one assignment-type event, and it does not expose a reel/frame in the text I obtained:

• 2009-01-21 (executed) / recorded 2009-01-21 — Reel/Frame: not exposed in the sources I could retrieve
  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Koizumi, Ryuta; Nomura, Yujiro; Sowa, Takeshi (as listed on Google Patents legal events)
  • Assignee: Seiko Epson Corporation
  • Correspondent: not exposed. The prosecution attorney of record on the pre-grant publication US 2009/0190228 A1 is DLA Piper LLP (US), San Diego, CA — but that is prosecution correspondence, not the assignment-recordation correspondent, and I will not conflate the two. For context only, contemporaneous Epson US filings were frequently recorded through Harness, Dickey & Pierce, P.L.C. (correspondent G. Gregory Schivley, later Bryant Edward Wade; deposit account "Epson R&D," Bloomfield Hills, MI) — I found those names on other Epson reels (e.g. 017431/0184; 035691/0967), not on this patent, so I am not attributing them here.
  • Context: internal — inventor-to-corporate assignment incident to filing. Routine.

No post-issuance assignment is recorded. Google Patents' legal-event timeline for US 7,848,023 lists only: filing (2009-01-21), the inventor assignment (2009-01-21), publication (2009-07-30), priority to the continuation 12/914,408 (2010-10-28), grant (2010-12-07). There is no transfer entry after grant. The listing names Seiko Epson Corporation as current assignee.

Caveat I must state: the Assignment Center UI itself was not retrievable through my searches, and Assignment Center indexes some records that Google Patents legal events do not. So "no post-issuance assignment recorded" is my best-evidence finding, not a certified clearance. If this matters for a legal purpose, run the patent number directly at https://assignmentcenter.uspto.gov/ and pull the reel/frame for the 2009 inventor assignment.

Timeline diagram

timeline
    title Ownership of US 7848023
    2008 : Priority JP 2008 14497
    2009 : US application 12357231 filed
         : Inventors assign rights to Seiko Epson
    2010 : Patent US 7848023 B2 issues
         : Continuation 12914408 filed
    2016 : Epson sells motor patents to IP Bridge
    2018 : IPValue affiliate 138 East buys Epson patents
         : No transfer of this patent recorded

Diagram note: the 2016 and 2018 nodes are portfolio-level context, not ownership changes of this patent. I include them because they are the only concrete evidence bearing on whether Epson still owns the '023, and I have deliberately not drawn them as links in this patent's chain.

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No post-issuance assignment to any "IP/Holdings/Licensing/Ventures"-style entity is recorded against this patent. Assignee of record remains Seiko Epson Corporation.

2. Known asserter in the chain — not present for this patent; context present at the portfolio level. No named asserter (Acacia, Marathon, IV, Wi-LAN, Round Rock, etc.) appears as an assignee of this patent. However: IPValue Management affiliates are now active asserters of Epson-origin patents — 138 East 57th Street (IPValue affiliate) asserted a '512 patent against BOE in Case No. 2:23-cv-00515 (E.D. Tex.) and again in an April 2025 suit, per BOE's PTAB filings. IPValue's press release states 138 East acquired "over 2,500 worldwide patents … from Epson in multiple transactions since April 2018," and a separate IPValue affiliate, Lumitek Display Technology Limited, took ~200 OLED patents from Seiko Epson. I found no evidence that US 7,848,023 is in any of those conveyances. Unconfirmed, not asserted.

3. Repeat correspondent across the chain — unclear. There is only one assignment link, so no recurrence is possible to test. I could not retrieve its correspondent. One appearance would in any case not be a finding under the stated rule.

4. Cascading transfers — not present. No chained LLC transfers; no assignments at all after the 2009 inventor assignment.

5. Pre-litigation transfer — not present. I found no infringement suit naming this patent, so there is no 6-month window to test. (The previously generated section likewise found no litigation, and the 2023/2025 suits I located assert other Epson-origin patents.)

6. Bankruptcy fire-sale — not present. Seiko Epson never filed; it divested businesses (exited the LCD display business; OLED assets) voluntarily, which is a licensing/sale program, not a bankruptcy sale.

7. Privateering — unclear, and worth flagging as the only genuine risk vector. Epson has demonstrably transferred large Epson portfolios to licensing vehicles (IPValue/138 East/Lumitek; IP Bridge's Godo Kaisha IP Bridge 1, which bought 200+ motor patents in 2016) that then license or assert downstream. That is the classic privateering fact pattern. What I cannot show is that this particular exposure-head patent went out in one of those packages. The technology class (organic-EL print-head optics) sits near the Lumitek OLED-display portfolio but is not a display patent, so inclusion is speculative.

8. Defensive aggregator (anti-NPE) — not present. No RPX / AST / LOT / Unified / OIN link.

Verdict

Insufficient data.

Justification: the available record shows exactly one assignment — the routine 2009-01-21 inventor-to-Seiko-Epson Assignment of Assignors' Interest (assignors Koizumi, Nomura, Sowa) — with no post-issuance transfer and no litigation found, so there is neither a shell-entity link, a named-asserter link, nor a pre-suit transfer to cite. Seiko Epson Corporation remains the assignee of record, and there is no evidentiary basis to call this patent an NPE asset today. The one live caveat is that Epson has verifiably sold large portfolios to IPValue/138 East/Lumitek and IP Bridge since 2016, so a later, unindexed transfer out of Epson cannot be excluded — but absent a reel/frame tying this patent to one of those conveyances, speculating further would violate the evidence rule.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 7848023) or the mirrored index https://patents.google.com/patent/US7848023/en under "Legal Events → Reassignment."

Two items for the record-keeping file:

• Reel/frame for the 2009 inventor assignment is not in hand — the earlier generated summary likewise did not carry it. Anyone relying on this chain should pull it directly.
• Possible cross-contamination between this patent and sibling app 12/358,042/US 2009/0189970 in claim mapping: the '023 uses 1 < L₁/L₂; the sibling uses 1 < L₂/L₁ (inverse), and the sibling added inventor Ken Ikuma. Do not let the two 13-claim sets bleed into each other.

I'll research the prior art cited against US 7,848,023. Let me pull the patent's citation records from multiple sources.

The first searches returned unrelated material. Let me target the patent's own citation record directly.

I have the core cited-reference list forming. Let me confirm the full citation table and the granted claim set.

Prior-Art Analysis — US 7,848,023 B2 ("Lens array, an exposure head and an image forming apparatus")

Sourcing note and a scope limitation up front

The task's method runs into a hard wall that I need to state plainly: the citation record I could retrieve for US 7,848,023 is short and, in at least one place, truncated. Two things follow:

1. I could not open the granted patent PDF or the USPTO PatentCenter "References Cited" pane directly; the citation table below is reconstructed from the Justia patent record for 7848023 and the patent's own specification, which I treat as the authoritative full text provided in the prompt.
2. The Justia rendering of the "Referenced Cited" list for 7848023 shows one U.S. patent and two Japanese publications, with an ellipsis indicating the visible snippet was cut off. There may be additional rows I did not see. Treat my list as "complete as retrieved," not "certified complete."

I also could not retrieve the disclosure text of the single cited U.S. patent (US 6,980,366). Under my operating rules I will not invent what it teaches or which claim it maps to; I flag that gap explicitly rather than fabricating a claim chart.

---

1. The patent citations that appear on the face of US 7,848,023

Non-patent citations: none retrieved. I cannot confirm whether any exist; the snippet I saw did not reach that part of the field.

Important cross-check — two of these are also cited in the specification body

The full text supplied in the prompt names references B and C by number in the description, which tells us what the applicant itself considered relevant art:

• JP-A-6-278314 is expressly identified in the Background as the known line head, "in which a plurality of substantially circular lenses are aligned in a longitudinal direction, for example, disclosed in FIG. 2 and the like of JP-A-6-278314." This is, by the patent's own characterization, the closest structural starting point.
• JP-A-2005-276849 is cited in the description as the source of the lens-array fabrication method: a mold with lens-shaped recesses pressed onto a glass lens substrate, the gap filled with photocurable resin, cured, and the mold released. This maps onto the specification's discussion of the lens array substrate (glass) and photocurable-resin lenses.

---

2. Reference-by-reference description

A. US 6,980,366 B2 (Ishikawa), issued 2005-12-27

Brief description: ⚠️ Not verified. This is the only U.S. patent in the retrieved citation list. My searches returned its number, inventor and issue date but not its title, assignee, abstract or claims, so I cannot state what it discloses or which element of claim 1 it was cited against. I am explicitly declining to characterize its teaching.
Why it matters: As the sole U.S. citation, it is the reference most likely to have been the examiner's §102/§103 anchor. If this analysis is going to be used for anything, US 6,980,366 must be pulled and charted element-by-element. See the verification checklist in §5.

B. JP 06-278314 A, published October 1994

Brief description: A line head (exposure head) for an electrophotographic apparatus in which a plurality of substantially circular lenses are aligned in the longitudinal direction; the lenses image light from light emitting elements onto a latent image carrier such as a photosensitive drum. The patent characterizes it as having a single row of lenses aligned only in the longitudinal direction.
Relevance to US 7,848,023: It is the direct antecedent of the problem the '023 patent solves. The '023 specification states that because the prior-art lenses are substantially circular, enlarging them to capture more light increases the pitch between lenses in the width direction, enlarging the line head and raising cost.

C. JP 2005-276849 A, published October 2005

Brief description: Per the '023 specification, this discloses a method of forming a lens array by holding a mold formed with lens-shaped recesses in contact with a glass substrate, filling the clearance with a light-curing resin, curing the resin, and releasing the mold. It is a process/lens-array-substrate reference, not an exposure-head reference.
Relevance: Supports the structural teaching that lenses can be formed on a light-transmissive lens array substrate, optionally of glass and optionally of photocurable resin.

---

3. Which claims could each reference anticipate under 35 U.S.C. § 102?

Framing. The application was filed 2009-01-21 on a JP priority of 2008-01-25, so pre-AIA § 102(a)/(b)/(e) governs. A § 102 anticipation requires a single reference disclosing every element of the claim, arranged as claimed. None of the retrieved references is an anticipation reference on its face; they are background/§ 103 material. That is the honest conclusion, and I map it out below.

A threshold claim-set discrepancy you must resolve before relying on any mapping. The previously generated summary in this thread states the patent has 13 claims with independents 1, 12 and 13 (exposure head / lens array / image forming apparatus). The Justia rendering of the granted claims for 7848023 instead shows a 10-claim set, all directed to the exposure head, in which:

• claim 1 recites the lens array (lenses in first and second directions), the light emitting element substrate, and a diaphragm arranged between the light emitting elements and the lenses, with 1 < L1/L2; and
• a standalone "lens array" independent claim and an "image forming apparatus" independent claim do not appear in that listing.

This is a genuine contradiction with the earlier summary, and it changes the anticipation answer materially (a diaphragm limitation gives the claims a great deal more to distinguish). The 13-claim structure is consistent with the pre-grant publication US 2009/0190228 A1; the 10-claim structure with the diaphragm appears to be the issued claim set. I analyze against both, and I flag this for verification against the printed patent.

Claim-by-claim for the issued (10-claim) set, rolled up: claims 1–10 all share the 1 < L1/L2 limitation, and claims 1, 4–9 additionally require the diaphragm. Because not one of the three retrieved face citations (as characterized) discloses a diaphragm with 1 < La1/La2, an elliptical diaphragm, a diaphragm–lens aspect-ratio match, or a diaphragm positioned on the image-plane side of the lens apex, the diaphragm family (claims 4–9) is where the patent's novelty most likely resides — and it is precisely the family the retrieved art cannot reach.

---

4. Most relevant prior art — ranked

1. JP 06-278314 A (Oct. 1994) — the closest single reference. It is the same field, same apparatus type (line head for a photosensitive drum), and the same functional problem (imaging light from emitting elements through a lens row). It is the § 103 starting point; it is not an anticipation reference because of the substantially circular lens shape and the single-row arrangement.
2. US 6,980,366 B2 (Ishikawa, 2005-12-27) — the only U.S. citation; ranking is provisional pending disclosure review.
3. JP 2005-276849 A (Oct. 2005) — method-of-manufacture reference; relevant to the lens-substrate dependent claim, not to the independents.

---

5. What this citation list does not capture, and what to pull next

The three retrieved citations are a thin record for a patent whose entire point is the aspect ratio of microlenses in a two-dimensional array for an LED/organic-EL print head. A complete prior-art picture should add:

• The actual granted claim set and the examiner's cited-art list from USPTO PatentCenter (referenced above) — this resolves the 10-vs-13 claim discrepancy and reveals any citations the public aggregator truncated.
• The full text and claims of US 6,980,366 B2 — mandatory; it is the U.S. anchor.
• The sibling/family applications' citation records, which corroborate the field. The continuation US 7,952,601 (Seiko Epson, appl. 12/358,042, Koizumi/Nomura/Sowa/Ikuma, filed 2009-01-22) lists the same JP 06-278314 and JP 2005-276849, plus US 4,734,734 (Yano), US 5,023,442 (Taniguchi), US 5,363,240 (Miyashita), US 7,432,947 (Yosuke). ⚠️ Those four are not shown in the 7848023 list I retrieved; whether they also appear on 7848023 (hidden by the truncation) is unverified.
• The continuation US 8,274,741 B2 (appl. 12/914,408) — same specification, and its claim set is worth comparing because it recites the inverse inequality 1 < L2/L1 < 1.2 in the sibling US 7,952,601, i.e., the family flipped which direction is the "long" one. That inversion is a strong hint about how the applicant was fencing around the prior art.
• Third-party art in the same space that is not on the face — e.g., the Oki Data / Yamamura microlens-array family. ⚠️ Date caution: the closest of those (US 2009/0067055, appl. 12/230,908, filed 2008-09-08; US 2009/0322847, filed 2009-05-28; US 7,957,067, filed 2009-06-29) post-date the 2008-01-25 priority, so they are not § 102 prior art against claims entitled to that priority. Do not treat them as anticipatory without a priority analysis.

---

6. Explicit uncertainty flags

• US 6,980,366 B2's disclosure is unverified. I did not retrieve its text, so I cannot name the claim(s) it might anticipate. This is the single largest hole in this analysis.
• The citation list may be incomplete — the Justia "Referenced Cited" field was truncated in the retrieved snippet.
• The granted claim set is disputed within this very thread: 10 claims with a diaphragm limitation in claim 1 (Justia, granted) versus 13 claims with independents 1/12/13 and no diaphragm in claim 1 (earlier summary, matching pre-grant US 2009/0190228 A1). § 102 conclusions depend on which is correct.
• Publication days for the two JP publications (I have only "October 1994" and "October 2005") were not independently verified. Both comfortably pre-date the 2008-01-25 priority, so the month-level date suffices for § 102 purposes.
• Bottom line: On the retrieved record, no face citation anticipates any claim of US 7,848,023; the citations function as background and § 103 material. The strongest single reference is JP 06-278314 A, and the patent's defensible novelty sits in the diaphragm limitations (claims 4–9 in the issued set), which none of the retrieved references touches.

I'll ground this analysis in the prior art actually cited on the face of the patent and in the field, then search for their content and for related family members.

I have strong material on the cited references. Let me pull the front-page citation list and one more potentially material reference.

Obviousness analysis — US 7,848,023 B2 under 35 U.S.C. § 103

Note on the governing law: Appl. 12/357,231 was filed 2009‑01‑21 and issued 2010‑12‑07, so the pre‑AIA version of § 103(a) and §§ 102(a)/(b)/(e) apply. All art quoted below must therefore predate the challenged claims' effective date. The patent claims priority to JP 2008‑14497 (2008‑01‑25) and JP 2008‑304814 (2008‑11‑28); assuming the JP priority is perfected, the critical date is 25 January 2008. I flag date problems for each reference, because two of the most tempting references fail on this point.

Note on sourcing / a contradiction to flag: the previously generated "Prior Art section" content was not actually captured in the fetched Google Patents text — the page as fetched omitted the "References Cited" table. I recovered the examiner‑cited list from Justia (https://patents.justia.com/patent/[7848023](/patent/7848023)), which shows only three references: US 6,980,366 B2 (Ishikawa, 2005‑12‑27), JP 06‑278314 (1994‑10), and JP 2005‑276849 (2005‑10). I also could not retrieve the disclosure text of US 6,980,366 in this session, so I will not characterize it; where it matters I say so explicitly rather than guess. Any statement below about JP‑A‑6‑278314's or JP 2005‑276849's content is drawn from how the patent itself characterizes them (which is an admission) plus corroborating later Seiko Epson filings.

---

1. What claim 1 actually requires (the linchpin of the analysis)

Claim 1 requires only:

1. a lens array with lenses "arranged in a first direction and in a second direction orthogonal to the first direction" (i.e., any 2‑D arrangement);
2. a light‑emitting‑element substrate bearing elements whose light is imaged by those lenses; and
3. 1 < L₁/L₂, where L₁ is the lens length in the first direction and L₂ the length in the second direction.

Critically, claim 1 nowhere ties the "first direction" to main scanning, or the "second direction" to the direction in which the lens‑row pitch must be conserved. All of the specification's reasoning for why the elongation points one way — "larger quantities of lights can be incident on the lenses in the first direction while pitches between the lenses in the second direction are suppressed" — is description, not claim language. Figure 1/2's assignment of LGD = main scanning = "first direction" is a disclosed embodiment, not a claim limitation. That asymmetry (broad claim / narrow rationale) is what makes claims 1 and 12 the most vulnerable claims in the patent.

Dependent claims add the only genuinely narrow material: L₁/L₂ < 1.2 (cl. 2), lens substrate (cl. 3), diaphragm between emitter and lens (cl. 4), 1 < La₁/La₂ (cl. 5), matched ratios (cl. 6), identical shapes (cl. 7), elliptical diaphragm (cl. 8), convex incidence surface (cl. 9), diaphragm on the image‑plane side of the apex (cl. 10), free‑form‑surface lenses (cl. 11).

---

2. Level of ordinary skill

A POSITA here is a person with a bachelor's degree in optical engineering, physics, or electrical engineering and roughly 3–4 years' experience in designing microlens arrays and LED/organic‑EL print heads, or equivalent. That person would know, as of January 2008: (i) that irradiance at the image plane scales with lens aperture area; (ii) that in an array, aperture area is bounded by the array pitch in the stacking direction but not (up to the point of overlap) in the row direction; and (iii) that anamorphic / non‑rotationally‑symmetric lens outlines are the standard way to allocate aperture between two axes (e.g., lenses "cut" along the array direction to permit dense packing of lens rows).

---

3. The prior‑art set, with dates

Two further points worth recording for the file:

• The patent's own admissions are usable prior art. The Background admits that to raise resolution one "can be formed by two‑dimensionally arranging a plurality of lenses" with "a plurality of lens rows … arranged in a width direction," and admits that the prior art's problem is that "since the lenses … are substantially circular, pitches between the lenses in the width direction … increase if the lenses are enlarged." Admissions in the specification about the state of the art are § 103 evidence.
• The applicant's own sibling application claims the mirror‑image inequality. Seiko Epson filed, one day apart, US 2009/0189970 (appl. 12/358,042, filed 2009‑01‑22; inventors Koizumi/Nomura/Sowa/Ikuma) claiming 1 < L₂/L₁, with line‑head claims reciting L₂ > L₁ (https://uspto.report/patent/app/20090189970), and US 7,952,601 (issued 2011‑05‑31) describing "lenses LS have an elliptical shape long in the latitudinal direction LTD … L₂ > L₁" (https://patents.google.com/patent/US7952601). This is not § 103 prior art — same inventive entity / same assignee, same‑day filings. But it is powerful evidentiary support that the direction of elongation was a recognized, freely selectable design variable, not an inventive insight: the same team, from the same specification, claimed the elongation pointing the other way, and both directions were treated as yielding "satisfactory exposure." That is the essence of a KSR design choice.

---

4. Grounds of rejection

Ground 1 — Claims 1, 12, 13: JP‑A‑6‑278314 + JP‑A‑2005‑276849 + US 2006/0072199 (and/or Official Notice), optionally with US 6,980,366

JP‑A‑6‑278314 supplies the exposure‑head architecture of claim 1's preamble and of claim 13 in full: a lens array, light‑emitting elements whose light is imaged by the lenses, and a latent‑image carrier exposed to form a latent image. It also supplies the baseline deficiency that motivates everything else: a row of substantially circular lenses, where any attempt to gather more light by enlargement necessarily increases the lens pitch in the row‑stacking direction.
(https://patents.google.com/patent/US7848023/en, ¶[0002]); https://patents.google.com/patent/US20090086328

US 2006/0072199 supplies (a) a molded lens array in an EL/LED electrophotographic exposure train, (b) express recognition that light‑concentration efficiency is a design variable to be optimized by shaping lens elements, and (c) a quantitative framework (ray‑trace, "normalized intensity") for trading shape against collected light. That is the direct motivation to modify the lens footprint of a line head in order to collect more light from a weak emitter. (https://patentimages.storage.googleapis.com/c5/9b/fe/1b15c123972b3f/US20060072199A1.pdf)

JP‑A‑2005‑276849 supplies the enabling, low‑cost route to non‑circular lens footprints on a glass substrate with photo‑curing resin — i.e., the "lens substrate" of claim 3 and, more importantly, proof that the lens outline is a free design variable rather than something fixed at "substantially circular."

The 2‑D lens arrangement is supplied either by the applicant's own admission in the Background (2‑D lens rows are the known answer to higher‑resolution exposure), by JP‑A‑6‑278314's own two‑row staggered architecture as characterized in US 2009/0086328, or by Official Notice that arranging microlenses in a matrix in a line head was routine by 2008 — a fact already embodied in the cited JP‑A‑6‑278314 line head itself and in the applicant's own admission.

Why a POSITA would combine — the motivation is unusually clean here:

1. The field's central design problem was light starvation. Both the patent and its siblings recite that organic‑EL emitters (especially bottom‑emission types) have "smaller light quantities as compared with LEDs." The art (US 2006/0072199) independently documents and quantifies the same problem. Collecting more light per lens was an identified, pressing need, not a hindsight construct.
2. The geometric solution is a one‑dimensional constraint problem. In a 2‑D lens array, the lens‑row pitch (LTD/sub‑scanning) directly sets head width and, in the patent's own words, head cost and drum‑diameter limits; the lens pitch along the row (LGD/main scanning) is less constrained. Increasing aperture area in the unconstrained direction only is the obvious solution: it strictly increases collected light (the emitter is a Lambertian/perfectly‑diffusing source per ¶ on "Lambert's cosine law," so added aperture in any direction adds flux) while leaving the constrained pitch untouched.
3. The luminous source is itself elongated in the row direction, and so is the spot group. The patent's own FIG. 8 element group has longitudinal width W_egg greater than widthwise width W_egt, and the exposure regions/spot groups are arranged consecutively along the main scanning direction. Matching an anamorphic illuminant footprint to a correspondingly anamorphic lens aperture is textbook aperture–etendue matching, and it is precisely the design incentive that would lead a POSITA to make L₁ > L₂.
4. Predictable result, no unexpected effect. The asserted benefit ("larger quantities of lights can be incident on the lenses in the first direction while pitches between the lenses in the second direction are suppressed") is arithmetic, not a discovered phenomenon. KSR applies with full force: "a finite number of identified, predictable solutions."

Claim 12 (lens array per se) is a fortiori obvious: it deletes the light‑emitting‑element substrate and the apparatus context, leaving only the 2‑D array of elongated lenses — the very structure that JP‑A‑2005‑276849 teaches how to make and that the admitted art describes.

Claim 13 (image forming apparatus) adds only a latent‑image carrier — already disclosed by JP‑A‑6‑278314.

Ground 2 — Claim 12, alternative: US 2006/0072199 + JP‑A‑2005‑276849 + JP‑A‑6‑278314

Same combination, differing only in which reference is designated "primary." Useful as a hedge if the Board finds JP‑A‑6‑278314's disclosure thinner than the patent's admission suggests.

Ground 3 — Claim 1, alternative: JP‑A‑6‑278314 + US 6,980,366

I flag this as unverified. If Ishikawa (US 6,980,366, cited by the Examiner) discloses a lens array in which the individual lens elements are non‑rotationally‑symmetric or laterally truncated to control array pitch (a common disclosure in this family of references), then the JP‑A‑6‑278314 + US 6,980,366 pair may reduce claim 1 to little more than substituting a non‑circular element for a circular one, with the only remaining question being the orientation of the elongation — and claim 1 does not claim the orientation. This ground should be evaluated only after the reference is read.

---

5. Dependent claims

The internal structure of the dependent set is itself diagnostic: claims 2 and 11 are the standard corrective steps for the very problem the elongation creates, and claims 4–10 are the standard illumination‑conditioning steps for a line head. There is no dependent claim that supplies a result the POSITA would not have predicted from the claim‑1 geometry.

A cross‑claim coherence argument that cuts against the patent: claim 1's stated purpose is to maximize light in direction 1, while claim 2 caps L₁/L₂ at 1.2 — a value that (for the FIG. 28 example, L₁ = 1.7 mm, L₂ = 1.5 mm → ratio 1.13; FIG. 20/29 examples use 1.66/1.9 and 1.66/2.0 in the sibling) yields only a modest asymmetry. A POSITA reading the specification learns only: elongate mildly in the direction where you need more light, and use free‑form surfaces and a matched aperture to keep aberrations down. Every element of that teaching is either admitted prior art or a conventional optical‑design expedient.

---

6. Why the asserted advantages do not defeat these grounds

• "Miniaturization while enabling good exposure at high resolution." Functionally identical to the recognized objective of the field (US 2006/0072199's explicit light‑concentration optimization; JP‑A‑6‑278314's acknowledged pitch problem). A benefit that is the predictable consequence of the claimed geometry gets no patentable weight.
• "Reduced sensitivity to photosensitive‑drum circumferential‑speed variation." This is asserted in the specification as an outcome of the reduced number/size of lens rows in LTD, not as a structural limitation of any claim. There is no nexus between this asserted advantage and the only claimed structure (the inequality 1 < L₁/L₂, unoriented). Under KSR, an unc claimed benefit cannot rescue an otherwise obvious claim.
• Astigmatism / mold‑release "teach away." Any such argument is self‑defeating: the applicant voluntarily claimed the remedy — a bounded aspect ratio (claim 2) and free‑form surfaces (claim 11). A reference that identifies a problem the applicant then solves with a conventional fix does not teach away; it supplies the motivation.
• No non‑obvious‑range argument is available for claims 1 and 12, because they recite no numerical bound at all (open‑ended > 1) and no orientation relative to scanning, resolution, or pitch.

The priority fork (an important contingency)

If the 25 January 2008 JP priority is not perfected for claim 1 as issued (e.g., if JP 2008‑14497 did not support the fully generalized "lenses arranged in a first direction and in a second direction" plus unoriented 1 < L₁/L₂ combination in the form allowed), the effective date shifts to 21 January 2009, and the following additional art becomes available as § 102(b)/§ 103 material: JP‑A‑2008‑036937 (two‑dimensional lens array imaging plural emitters per lens — the exact claim‑1 architecture) and US 2009/0067055 A1 / EP 1905602 A2 (Oki Data; micro‑lenses truncated along the array direction, with aperture plate and pitch design rules). On that alternative footing, claim 1 becomes very difficult to sustain. Verify the priority‑document support for claim 1 before relying on the 2008 date.

---

7. Rebuttal arguments the applicant/patentee would likely press, and my assessment

1. "No reference teaches or suggests elongating the lens in the first (main‑scanning/longitudinal) direction specifically, as opposed to the width direction." — Strong only for claims whose scope is tied to that orientation. It fails against claims 1 and 12, which are not so tied. It also fails as a matter of policy weight, because the same applicant's contemporaneous sibling application treated the opposite orientation as equally acceptable.
2. "The art teaches circular lenses and would lead away from non‑circular ones because of astigmatism." — Weak. The patent itself cures the astigmatism with a bounded ratio and free‑form surfaces, and the Oki Data art (timing‑caveated) shows that truncating micro‑lens outlines to fit array pitch was conventional.
3. "Unexpected magnitude of light gain." — Weak. The specification supplies no comparative data against a circular lens of equal area; the benefit is asserted geometrically.
4. "Objective indicia." — No evidence of commercial success, copying, licensing, or industry praise is on the record of this document set. The patent expired for failure to pay maintenance fees and, per the previously generated litigation section, no assertion activity was found — factors inconsistent with a strong secondary‑considerations story.

---

8. Bottom line

Most probable outcome: claims 1 and 12 would be held obvious as a matter of design choice over JP‑A‑6‑278314 in view of JP‑A‑2005‑276849 and US 2006/0072199, with the 2‑D lens arrangement supplied by the applicant's own Background admission and/or Official Notice (alternatively, with US 6,980,366 substituted once its disclosure is verified). Claim 13 falls with claim 1 because JP‑A‑6‑278314 already discloses the latent‑image carrier. Claims 2–11 are each obvious over the same base combination in further view of conventional micro‑lens‑array practice and free‑form‑surface design, with the strongest single point being that the applicant's own claims 2 and 11 are the textbook corrections for the aberration the elongation introduces.

Residual risk / where I am not confident: the entire case for claims 1 and 12 turns on how one reference (JP‑A‑6‑278314) is read, on whether US 6,980,366 in fact discloses non‑circular or laterally constrained lens elements, and on whether the examiner's sparse three‑reference citation list reflects a genuinely thin art or merely a narrow search. I could not retrieve the text of US 6,980,366 or a USPTO‑verified copy of the printed claims in this session, and the fetched Google Patents page did not contain a "References Cited" table (I obtained that list from Justia). Before relying on this analysis for any filing or validity opinion, pull: (a) the printed claims from USPTO PatentCenter; (b) JP‑A‑6‑278314 and JP‑A‑2005‑276849 in full with certified translations; (c) US 6,980,366; (d) US 2006/0072199; and (e) the JP priority documents for support of claim 1.