US 6422676

I'll search for this specific patent number across USPTO and CAFC sources.

Let me run additional targeted searches to confirm litigation/docket status.

Summary — US Patent 6,422,676

Important scope note: I searched for the literal number 6422676 (and the formatted 6,422,676). No USPTO reexamination/reissue, no PTAB proceeding, and no CAFC 2026 docket naming this patent appeared in my searches. Searches for "'267 patent" + CAFC/PTAB surfaced a different '267 patent (a video-coding patent in ITC Inv. 337-TA-1380 and related IPRs) and an ITC inkjet investigation (337-TA-723) that asserted other HP patents ('598, '053, '347, '817, '279, '917) — not 6,422,676. I flag these explicitly so they are not conflated with the patent you asked about.

Bibliographic data

Abstract

"A compact thermal ink jet printhead including a printhead substrate, a plurality of side by side columnar arrays of drop generators formed in the printhead substrate, and drive circuits formed in the printhead substrate for energizing each ink drop generator. The printhead substrate has an ink drop generator packing density of at least 10.43 ink drop generators per square millimeter."

Plain-language overview of the independent claims

Claim 1 — compact, high-packing-density printhead (structural). An ink jet printhead built on a substrate having multiple thin-film layers, with four side-by-side columnar arrays of drop generators running along a longitudinal direction. Each column must have at least 100 drop generators at a pitch "P." Integrated drive circuitry must be able to fire each drop generator at ~15 kHz to ~18 kHz. The final limitation is the point of novelty: the substrate must exhibit an ink drop generator packing density of at least 10.43 generators per mm² — i.e., putting a lot of nozzles on a small die.

Claim 5 — same density goal, but reciting how the drive electronics are laid out. Adds that the drive circuits are columnar arrays of FET drive circuits placed adjacent to the corresponding drop-generator columns, and that there are ground busses that overlap the active regions of those FETs. (The overlap is what lets the peripheral circuitry occupy a narrow strip, which is what enables the small die width / high density.) Same ≥100-gen/… / 15–18 kHz framing and same ≥10.43 generators/mm² density requirement.

Claim 28 — method/functional variant of claim 1. Same architecture (four columnar arrays, ≥100 drop generators each at pitch P, density ≥10.43/mm²), but the drive-circuit limitation is phrased functionally as circuits "that energize" each generator at ~15–18 kHz (versus "for energizing" in claim 1). Substantively a close parallel to claim 1.

Dependent-claim themes

• Layout geometry: adjacent columns within a feed-slot pair separated by ≤630 µm (claim 2); the two inner columns separated by ≤800 µm (claim 4); two ink feed slots with columns arranged on either side (claim 3).
• FET/drive-circuit details: on-resistance < (250,000 Ω·µm²)/A where A = FET area (claim 6), gate oxide ≤800 Å (7), gate length <4 µm (8), Ron ≤14 Ω (9) or ≤16 Ω (10), on-resistance sized to compensate for power-trace parasitic resistance (11–15), FET array region width ≤180 µm (16) or ≤250 µm (17).
• Drop sizes / resistors: 12–19 pL (claim 22) or 3–7 pL (claim 23); heater resistor ≥100 Ω (24).
• Die aspect ratio: LS/WS > 3.5 (25); WS ≤~3.29 mm (26) or ~2.95–3.29 mm (27).

Uncertainty / caveats

• The 15 kHz–18 kHz and 10.43 generators/mm² values are as literally recited; I have not independently verified the underlying measurements.
• I found no evidence of litigation, IPR, reexam, or a 2026 CAFC appeal specific to 6,422,676. Given the patent expired (per the record) on 2021-06-19, such activity would be unusual. If you need a definitive docket check, it should be run directly against USPTO Patent Center/PTAB and the CAFC CM/ECF docket (by patent number), since my search covered web-indexed sources only.
• Assignee status shown is the recorded assignment chain (HP Company → HP Development Company, L.P.); the Google Patents notation cautions that listed assignees are not a legal conclusion.

I'll research litigation involving US patent 6422676 across litigation-tracking sources.

Let me search more targeted sources for litigation specifically involving the '676 patent.

Litigation search results for US Patent 6,422,676

Short answer: Based on my searches of patent-litigation sources, I could not identify any litigation (district court, ITC, or PTAB) in which US Patent 6,422,676 was specifically named as an asserted or challenged patent. I want to be explicit that this is a "no confirmed results" answer rather than an assertion that none exists — see the caveats below.

What the patent is

• Patent: US 6,422,676 B1, "Compact ink jet printhead"
• Assignee/owner of record: Hewlett-Packard Company → later Hewlett-Packard Development Company, L.P. (assignment recorded 2005-01-11)
• Inventors: Torgerson, Browning, MacKenzie, Miller, Bakkom, Dodd
• Filing/priority: 2001-06-19; Granted: 2002-07-23
• Legal status (per source): Expired – Lifetime; anticipated expiration 2021-06-19

Source: https://patents.google.com/patent/US6422676/en

What the searches did and did not return

Numeric near-misses that appeared in search results and are not this patent (I did not treat them as matches):

• JP 6422676 B2 — a Japanese patent (Nitto Denko, transparent conductive film) — a different document in a different office.
• "despacho nº 6422676/2021" — a Brazilian court administrative order number.
• "6422676" — a Brazilian company registration number (JUcec-CE filing).
• 6,346,876 (Omega Patents v. CalAmp) — a different US patent number.

Important caveats

1. Litigation databases are not fully indexed by open web search. Docket Navigator, LexisNexis, and PACER do not expose their full case-to-patent mappings to general search engines, so a true negative requires a paid/database query I could not complete here.
2. The patent expired 2021-06-19, so any infringement suit would necessarily have been filed before that date, and the damages window is closed.
3. It is plausible, but not confirmed, that the '676 patent rode along in an HP portfolio assertion (e.g., an HP Inc. inkjet/printhead campaign or a counterclaim), given its subject matter. I found no evidence of this, and I am not asserting it.
4. The Google Patents "Cited By" list shows later patents (e.g., Canon, Lexmark, HP fluid-ejection patents) that cite the '676 as prior art — citation of a patent is not litigation involving it, and I have not treated those as litigation.

Bottom line

No confirmed litigation involving US 6,422,676 was found. If you need a definitive answer, the next step would be a direct query of Docket Navigator or LexisNexis Patent Litigation by patent number, and a PACER/ITC EDIS search, which are the authoritative case-to-patent sources and were not fully reachable via the searches performed here. I'd rather flag this as unverified than list a case I cannot substantiate.

I'll verify the "no PTAB activity" default against live sources before finalizing.

Let me run a couple more targeted checks to be thorough.

Let me do one final check on the patent's assertion/enforcement history and any related IPR family activity.

Proceedings overview

Zero AIA trial proceedings are on file for US 6,422,676. The structured "PTAB proceedings on file" block from the USPTO Open Data Portal reports no AIA trials (IPR/PGR/CBM) against this patent, and my independent web searches returned no IPR, PGR, or CBM docket number, institution decision, or Final Written Decision naming the '676 patent. There are therefore no claims invalidated, none sustained, no settlements, and no institution denials — the defensive posture for a defendant is not "the patent has been hardened by surviving IPRs" and not "claims 1–5 are canceled." The correct read is: the '676 patent's validity has never been tested at the PTAB, and — more importantly — the patent is expired, so the practical exposure is historical-damages-only, not injunctive.

A note on confidence: absence of a proceeding in the ODP extract plus absence of hits in my searches is strong but not the same as an exhaustive PTAB docket scrub. If you are relying on this for a filing or a stay motion, confirm directly in PTAB E2E / Patent Center before you commit, since the ODP ingest can lag and older proceedings (pre-2018) are indexed less completely than recent ones.

---

Per-proceeding detail

None to report. There is no proceeding number to list, and I will not invent one.

For completeness — the fields you asked for, applied to the actual record:

• Type: N/A (no petition filed)
• Filed: N/A
• Status: N/A — no AIA trial on file; the patent's own legal status is "Expired - Lifetime," with an anticipated expiration of 2021-06-19 (20 years from the 2001-06-19 filing date)
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A
• Settlement / termination: N/A
• Appeal: N/A — no FWD to appeal to the Federal Circuit

---

Strategic summary

Claim status: everything is UNTESTED, but the patent is dead. Claims 1–31 of US 6,422,676 (three independent claims — 1, 5, and 28 — plus their dependents) were never challenged in an AIA trial. There is no IPR cancellation, no certificate of correction narrowing claims, and no PTAB record to cite. But the operative fact for a defendant is the expiration date: the patent issued 2002-07-23, was filed 2001-06-19, and per the structured record anticipated expiration was 2021-06-19. The owner of record is Hewlett-Packard Development Company, L.P. (originally Hewlett-Packard Company). HP is not a habitual patent monetizer, and I found no assertion of the '676 patent in the searches I ran — consistent with this being a defensive portfolio patent in HP's thermal-inkjet printhead family rather than a litigation asset. Even so, an expired patent can still be asserted for past damages inside the six-year lookback of 35 U.S.C. § 286, which for the '676 patent means, at the outside, conduct between roughly 2015-06-19 and 2021-06-19. Any demand letter alleging ongoing infringement of this patent, or seeking an injunction, is facially defective.

Estoppel landscape: § 315(e)(2) estoppel is a non-issue because nobody petitioned. There is no petitioner, and therefore no privity chain, whose IPR estoppel could constrain anyone. That cuts both ways for a current defendant: (a) you are not barred from raising any prior-art ground, including grounds a hypothetical petitioner raised or could have raised — the entire § 102/§ 103 universe is available to you in district court or, if you wish, in a fresh petition; but (b) you also get no free ride — there is no FWD to borrow for collateral estoppel, no PTAB invalidity finding to hand the judge, and no IPR-driven stay argument grounded in an existing instituted trial. Note one § 325(d) practical wrinkle if you do petition: the file shows HP itself as the examiner-cited reference US 5,604,519 ("Inkjet printhead architecture for high frequency operation," HP, 1997-02-18) — same assignee. Art previously considered by the examiner, especially same-family HP art, invites a § 325(d) discretionary-denial fight, so any petition should lead with art the examiner never saw.

Pattern signals. No repeat-petitioner pattern (no petitioner at all). No patent-owner PTAB-appeal pattern (no FWD to appeal). No defensive aggregator (no Unified Patents / RPX-style filer) on the docket. The only "family" signal is sibling HP printhead patents that appear as similar documents — US 6,309,053 ("Ink jet printhead having a ground bus that overlaps transistor active regions"), US 6,398,347 and US 6,412,917 (energy-balanced printhead designs), US 6,523,935 ("Narrow ink jet printhead"), and US 6,860,587 ("Ink jet printhead"). Claim 5 of the '676 patent expressly claims the ground-bus-overlaps-active-regions feature that is the subject of US 6,309,053, so if you are mapping prior art against one of these siblings, the same references will likely read on the '676 claims. Also note the international family (WO2002102597A2, EP1397256A2 — withdrawn, CA2451629C, JP4490093B2, KR100816133B1, CN1545452A, AU2001292592B2, RU2279983C2, BR0117071B1, TW541245B, PL199677B1, MY130478A/MY130807A, MXPA03011989A, AR036061A1) — foreign prosecution files can contain examiner-cited art not in the US file, which is a cheap source of § 102/§ 103 ammunition.

---

Recommended next steps

1. If you are a defendant being asserted: the expired-patent defense is your first and cheapest move. Put HP (or whoever is asserting) to proof on the expiration date and on the § 286 damages window. A demand letter citing ongoing infringement or injunctive relief should be met with a request for the specific accused acts and their dates — anything after 2021-06-19 is outside the patent term.
2. If infringement is alleged for the 2015-06-19 → 2021-06-19 window: you have the full prior-art field available with no IPR estoppel. Because there is no FWD, there is no § 315(e)(2) bar and no estoppel from a prior petitioner.
3. If you are contemplating a defensive IPR: understand that you would be the first petitioner on this patent, so General Plastic repeated-petition factors do not apply. But weigh the cost/benefit hard — IPRs on expired patents have reduced practical value, and the Board has historically been reluctant to institute where no forward-looking exclusionary right is at stake. The better forum for an expired patent is district court invalidity or a motion on damages.
4. If no proceeding appears when you check PTAB E2E: say so plainly in your motion papers, and note that the absence of any IPR is a signal this patent was never commercially asserted — it supports a narrative of a non-practiced, never-litigated, expired portfolio patent, which is relevant to willfulness, enhanced damages, and exceptional-case exposure.
5. Verify before filing: confirm the no-proceedings status directly at USPTO PTAB E2E and check CourtListener for any district-court or Federal Circuit activity involving US 6,422,676 that my searches may have missed. My search coverage for pre-2018 PTAB filings and for non-indexed district-court assertions is not exhaustive, and I would rather flag that gap than overstate the record.

I'll research this patent's ownership chain. Let me start with the primary sources and cross-reference litigation/NPE databases.

Let me verify the USPTO assignment records and check for any post-issuance transfers or litigation.

I hit the tool budget before I could open the live Assignment Center directly, so the timeline below is reconstructed from the recorded assignment events mirrored in the patent's own legal-events table (which reproduces USPTO reel/frame data) plus the family/continuation records surfaced in the searches. I flag explicitly where I could not retrieve a data field (correspondent names) rather than guess.

Inventors

Unusual-pattern check: none. These are career HP thermal‑inkjet engineers. No evidence any inventor departed within 12 months of filing — Torgerson reappears as an HP inventor on later filings in the same program (e.g., HP's US 2005/0231562, "Fluid ejection device"). The staggered Dodd execution date is a routine administrative artifact, not a precursor to a fire-sale.

Original assignee

Hewlett-Packard Company (front page), Colorado corporation; later Hewlett-Packard Development Company, L.P. (current assignee of record).

• Primary line of business: computers, printers, and imaging hardware — HP was the co-founder of commercial thermal inkjet printing.
• Product embodying the claims: Yes. The claimed subject matter (four side‑by‑side columnar drop‑generator arrays, dual dual‑edge ink feed slots, ≥10.43 drop generators/mm², 15–18 kHz firing, single‑pass monochrome at 1200 dpi) describes the high‑density printhead architecture HP shipped in its thermal inkjet pens/cartridges. The specification expressly frames it against "a typical Hewlett‑Packard ink jet printhead."
• Current status: Operating. HP's corporate split (2015) separated HP Inc. (PCs/printers) from Hewlett Packard Enterprise; the patent sits with Hewlett-Packard Development Company, L.P., HP's IP‑holding operating subsidiary. The patent itself expired 2021‑06‑19 (anticipated expiration; status "Expired – Lifetime"). HP has never filed Chapter 7/11.

Assignment timeline

Two recorded assignments (AS events). The later "FPAY" events are maintenance‑fee payments, not conveyances.

• 2001‑06‑12 to 2001‑06‑14 (executed) / recorded 2001‑08‑20 — Reel 012097/0964

  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Torgerson, Joseph M.; Browning, Robert N. K.; MacKenzie, Mark H.; Miller, Michael D.; Bakkom, Angela W.; Dodd, Simon
  • Assignee: Hewlett-Packard Company (Colorado)
  • Correspondent: not retrievable from the available record. The mirrored legal‑events entry does not carry the attorney/firm of record, and I did not reach the Assignment Center to pull it. I will not invent it.
  • Context: Original inventor‑to‑employer assignment (standard HP corporate assignment practice).

• 2005‑01‑11 (executed / effective) / recorded 2005‑01‑11 — Reel 015583/0106

  • Conveyance: Assignment
  • Assignor: Hewlett-Packard Company
  • Assignee: Hewlett-Packard Development Company, L.P. (Texas)
  • Correspondent: not retrievable from the available record.
  • Context: Internal corporate reorganization — HP's transfer of its patent estate into its wholly‑owned IP‑holding subsidiary (the same reel 015583/0106 pattern appears across the sibling HP inkjet patents, e.g. US 6,523,935 and AU 2001292592B2).

Cross‑reference (not an assignment on this patent): US 6,422,676 appears as a cited prior‑art reference in Lexmark International's prosecution of application 10/956,939 ("Power and ground buss layout for reduced substrate size," David G. King, filed 2004‑09‑30, correspondence: Lexmark International IP Law Dept., 740 W. New Circle Rd., Lexington KY), later issued as US 7,195,341 / US 7,344,227. That is competitive citation, not assertion.

Timeline diagram

timeline
    title Ownership of US 6422676
    2001 : Inventors assign to Hewlett-Packard Company
         : Recorded at reel 012097 frame 0964
    2002 : Patent issues on 23 Jul
    2005 : HP assigns IP to HP Development Co LP
         : Recorded at reel 015583 frame 0106
    2006 : Maintenance fee paid
    2010 : Maintenance fee paid
    2013 : Maintenance fee paid
    2021 : Patent expires on 19 Jun

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only post‑issuance conveyance (reel 015583/0106, 2005‑01‑11) runs from Hewlett-Packard Company to Hewlett-Packard Development Company, L.P., a wholly‑owned HP IP subsidiary that is not a licensing‑only shell. No "IP/Holdings/Ventures" licensing vehicle appears anywhere in the chain.
2. Known asserter in the chain — not present. Current assignee of record is HP Development Company, L.P.; it matches no entry on the Acacia / Marathon / IV / IPNav / Wi‑LAN‑Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock lists. No RPX or Unified Patents high‑frequency‑plaintiff hit surfaced.
3. Repeat correspondent across the chain — not assessable (and no recurrence). With only two links and the correspondent field not retrievable from the mirrored record, there is no basis to find a recurring NPE filing attorney. I record this as not assessable, not as a negative finding.
4. Cascading transfers — not present. Two recorded assignments over roughly four years, both to HP entities. No chained LLCs within 24 months.
5. Pre-litigation transfer — not present. No infringement action naming US 6,422,676 was located (the only hit was the Lexmark IDS citation above). There is no litigation to back‑date a transfer against.
6. Bankruptcy fire-sale — not present. HP has not filed Chapter 7/11; no judicial sale of this asset.
7. Privateering — not present. No SEC‑filing or Patent Progress/EFF evidence of HP transferring this patent to an NPE to assert against competitors; the 2005 record is an intra‑corporate reorganization, and HP's later continuation filings (e.g., US 2005/0231562, Torgerson) show HP retained and continued the program.
8. Defensive aggregator — not present. The chain terminates at HP's own operating IP subsidiary, not at RPX, AST, LOT, or OIN.

Verdict

Operating-company assertion. The ownership chain is a complete, two‑link record ending with a product‑shipping operating company: inventors → Hewlett-Packard Company (reel 012097/0964, recorded 2001‑08‑20) → Hewlett-Packard Development Company, L.P. (reel 015583/0106, 2005‑01‑11), with no NPE, shell LLC, asserter, or defensive aggregator anywhere. HP's printer products embody the claimed high‑density printhead architecture, and the patent ran to its natural 2021‑06‑19 expiry in HP's hands. Caveat for precision: no infringement suit naming this patent was located, so if the taxonomy requires an actual filed assertion to earn this label, the more literal description is "operating-company‑owned, no NPE activity, assertion unconfirmed" — but nothing in the record supports any NPE characterization.

Verification: USPTO Assignment Center — https://assignment.uspto.gov/patent/index.html (search 6422676); mirror of the two recorded assignments and reel/frame numbers: https://patents.google.com/patent/US6422676/en (Legal Events).

I'll research the prior-art citations for US 6,422,676 and verify each reference's details.

Let me verify the cited reference details and check the family prosecution history/ISR.

Let me check the family's international/EP search reports to see which references were applied against the claims.

Prior-Art Analysis — US 6,422,676 B1 ("Compact ink jet printhead")

Scope and method. I searched the USPTO/Vitible-looking sources and the Google Patents record for the literal number 6,422,676 (and 6422676). Consistent with the previously generated sections, the only reference listed in the patent's own "Patent Citations" field is US 5,604,519 A. The same record also carries a "Family Cites Families" list of 20 references that were cited during prosecution of the family applications (WO 02/102597, EP 1397256, JP 4490093, KR 100816133, etc.) based on the same priority document. I analyze the direct citation in detail and the family-level citations in a second table, clearly separating the two — because only the first is literally a "patent citation for 6422676."

Legal framework note. US 6,422,676 was filed 2001-06-19 (pre-AIA, since the first-inventor-to-file provisions took effect 2013-03-16). Pre-AIA §102(b) art = patented/published more than one year before 2001-06-19 (i.e., before 2000-06-19). §102(a) = published before the invention. §102(e) = U.S. patent/application publication whose effective U.S. filing date predates the '676 filing. Anticipation requires a single reference disclosing every limitation of a claim; where a reference shows several but not all limitations, it is §103 art, not §102 art. I flag both.

---

PART A — The direct examiner citation (the one item under "Patent Citations")

Brief description. US 5,604,519 discloses an ink-delivery/architecture scheme for a thermal ink-jet print cartridge aimed at raising firing frequency. It recites an ink reservoir; a substrate with a plurality of individual firing chambers, each with a firing element; an ink channel (primary channel → secondary channel → separate inlet passage per chamber) enabling high-frequency refill; firing chambers grouped into "primitives" in which only one chamber in a primitive fires at a time; first circuit means on the substrate connected to the firing elements; and second circuit means on the cartridge for transmitting firing signals at >9 kHz. Dependent claims step the frequency up to >10 kHz, >11 kHz, and >12 kHz (claims 7–9), and the method claims recite firing at >8 kHz (claims 10, 13–16). The specification explains that increasing resolution requires nozzles/heater resistors closer together, which in turn requires higher firing frequency.

Why the examiner cited it. It is the closest HP "high-frequency printhead architecture" art bearing on the 15–18 kHz drive-frequency limitation that appears in every independent claim (1, 5, 28), and on the on-substrate drive circuitry / primitive-select architecture.

§102 analysis — which claims could it potentially anticipate?

• Statutory basis: Qualifies under §102(b) (granted 1997-02-18, well over one year before 2001-06-19), with a §102(e)/continuity fallback.
• Claim 1: Discloses (a) a thin-film-patterned substrate with firing chambers/elements and (b) a substrate-borne drive circuitry layer, but it does not disclose (c) four side-by-side columnar arrays of ≥100 drop generators at pitch P, nor (e) the ≥10.43 generators/mm² packing density. Its frequency disclosure reaches only >12 kHz, not the claimed 15–18 kHz. ⇒ Not anticipatory of claim 1; relevant §103 art for the drive/frequency element.
• Claim 5: Discloses substrate-borne "first circuit means" but does not disclose columnar arrays of FET drive circuits with ground busses overlapping the FET active regions, nor the ≥10.43/mm² density. ⇒ Not anticipatory; §103 art for the on-substrate driver limitation.
• Claim 28: Same shortfall as claim 1 (four columnar arrays ≥100 / density ≥10.43 not shown; frequency range not met). ⇒ Not anticipatory; §103 art.
• Claims 11–15 (parasitic-resistance compensation by FET sizing): US 5,604,519's primitive-select/address architecture is background to the drive scheme but does not disclose compensating FET on-resistance for power-trace parasitic resistance. ⇒ No §102 anticipation; at most §103 context.
• Net: No claim of the '676 is anticipated by this reference; the examiner appears to have cited it as the primary §103 teaching of on-substrate, high-frequency drive circuitry.

---

PART B — References cited in the family applications ("Family Cites Families," 20 items)

These are the references the '676's foreign/EP/JP counterparts placed of record; they were not listed as the '676's own examiner citations. Treat their §102 attributions as indicative of the family-level prior-art landscape, not as the U.S. examiner's formal rejections.

---

Bottom line

• The only true "patent citation" for US 6,422,676 is US 5,604,519 A (Keefe et al., Hewlett-Packard). It is §102(b) art (granted 1997-02-18; priority back to 1992-04-02) and is the closest reference on the high-frequency (≥9–12 kHz) on-substrate drive element, but it does not disclose the '676's distinguishing limitations — four side-by-side columnar arrays of ≥100 drop generators and a packing density ≥10.43 generators/mm² — nor the claimed 15–18 kHz range. It therefore does not anticipate any of independent claims 1, 5, or 28; it is best characterized as §103 art.
• None of the 20 family-cited references appears to anticipate a complete independent claim. They are relevant piecemeal: US 5,144,341 (driver layout → claim 5), US 5,317,346 (compound feed slot → claim 3/20), US 6,234,598 (ground returns → claims 5/11), US 6,137,502 (drop sizes → claims 22/23), US 6,111,291 (MOS/FET characteristics → claims 6–10), and US 5,469,199 / US 4,719,477 / US 5,774,148 (thin-film, multi-array, and printhead-IC background).

Caveats / things I could not verify

1. Two statutory-date traps in the family list. Several entries are shown with grant dates that fall after the '676 filing (US 6,259,463 — 2001-07-10; JP 3359253 B2 — 2002-12-24; EP 1,563,998 B8 — 2010-10-20; US 6,439,696 — 2002-08-27). For these, only the earlier A-publication or the §102(e) U.S. filing date can supply prior-art effect; I could not retrieve those underlying dates with confidence, so I flagged rather than asserted them.
2. This list is the family citation set, not the '676's own file-wrapper citations. The U.S. "Patent Citations (1)" field for '676 contains only US 5,604,519. Other references (and any non-patent literature) may appear in the '676 U.S. file history but are not exposed by the indexed Google Patents record.
3. I could not retrieve the Box C "Documents Considered Relevant" of the PCT (PCT/US2001/028151) or EP 1,397,256 search reports, so I cannot state which references the ISA/EPO applied against which counterpart claims. That is the authoritative source if you need a per-claim §102 mapping at the family level.
4. Anticipation is a high bar. My §102 commentary above reflects the fact that no single listed reference teaches the combination of (i) four columnar arrays of ≥100 drop generators, (ii) 15–18 kHz firing, and (iii) ≥10.43 generators/mm². If you need a formal invalidity opinion, the file history (USPTO Patent Center for App. 09/883,398) and the WO/EP search reports should be pulled directly.

Sources: https://patents.google.com/patent/US6422676/en ; https://patents.google.com/patent/US5604519 ; https://patentimages.storage.googleapis.com/81/2c/b3/2da8b01c181c19/US5604519.pdf ; https://patents.google.com/patent/WO2002102597A3

I'll ground this analysis in the prior-art references listed on the patent page. Let me pull details on the most probative ones.

Interesting — my search surfaced PTAB petitions that quote claim language matching the '676. Let me pin down what those proceedings are and get details on the related "similar documents" references.

I now have strong grounding. Let me verify a couple of specifics on the sibling "narrow ink jet printhead" reference (column spacing and nozzle counts).

Obviousness Analysis — U.S. Patent No. 6,422,676 ("Compact Ink Jet Printhead") under 35 U.S.C. § 103

0. Scope, sources, and a flag on the earlier sections

This analysis uses the prior-art content of the patent page itself, which comes in three buckets that must not be conflated:

Two corrections/cross-references to the earlier-generated sections:

1. The earlier litigation summary stated no PTAB or CAFC activity could be found. That remains true for the '676 itself, but my searches for this task surfaced directly relevant proceedings involving the same HP printhead family as prior art: IPR2022-01414, and Slingshot Printing LLC v. Canon U.S.A., Inc., No. 24-1956 (Fed. Cir. Apr. 23, 2026) (nonprecedential), affirming that claims 1–5 and 8 of U.S. Patent No. 7,195,341 are unpatentable over U.S. 6,412,917 (Torgerson) + U.S. 7,240,997 (Bruce). https://www.cafc.uscourts.gov/opinions-orders/24-1956.OPINION.4-23-2026_2681303.pdf — the "Torgerson" reference there is an HP sibling of the '676 (same specification, same reference numerals 11, 56, 61, 81, 85, 181). This is the '676 family as art, not the '676 as a litigated patent. It is nonetheless highly probative of how a fact-finder treats this exact disclosure.
2. The earlier sections noted an ITC action asserting HP's '598 patent. That is US 6,234,598, "Shared multiple terminal ground returns for an inkjet printhead," which appears in the '676's own Family Cites Families list — a real cross-link, not a coincidence.

---

1. Legal framing (pre-AIA controls)

The '676 was filed 2001-06-19 with no priority claim, so pre-AIA § 102/§ 103 apply. Obviousness is judged under Graham v. John Deere and KSR Int'l v. Teleflex: scope/content of the art, differences, PHOSITA level, and objective evidence. Critically, pre-AIA 35 U.S.C. § 103(c)(1) removes from § 103 consideration art that qualifies only under § 102(e)/(f)/(g) where the art and the claimed invention were commonly owned at the time of invention. This matters enormously here (§ 5).

PHOSITA: a person with a B.S. in electrical/mechanical engineering, materials science or physics plus ~3–5 years designing thermal ink-jet printheads, including thin-film resistor/metallization process integration and printhead drive-circuit layout.

Prior-art date map (invention date ≈ mid-2001 at the latest):

• 102(b) (>1 yr before filing): US 5,604,519; US 5,469,199; US 5,317,346; US 4,719,477; US 5,144,341; US 5,774,148; US 6,011,112; JP H0618197B2; JP H06171084A; JP H08162543A.
• 102(a)/(e): US 6,137,502 (issued 2000-10-24, Lexmark); US 6,111,291 (2000-08-29); US 6,190,000; US 6,234,598 (2001-05-22); JP 2001-015719 (2001-01-19).
• 102(e) only, likely HP-common-owned → § 103(c) risk: US 6,309,053 (filed 2000-07-24, granted 2001-10-30); US 6,412,917 (granted 2002-07-02); US 6,398,347; US 6,523,935 (granted 2003-02-25); US 6,488,363 (filed 2001-12-17 — after the '676 filing date, so likely not § 102(e) art at all).

---

2. The single most important observation: the "10.43/mm²" limitation is a quotient, not structure

Claim 1 (and claims 5 and 28) end with: "said printhead substrate having an ink drop generator packing density of at least 10.43 ink drop generators per square millimeter."

That number is arithmetic on elements the claim (and the art) already recites. The specification states the four-column implementation has 100–104 drop generators per column on a die of LS ≈ 11.65 mm × WS ≈ 3.29 mm. 4 × 100 = 400; 11.65 × 3.29 = 38.33 mm²; 400 ÷ 38.33 = 10.43/mm² exactly. The upper figure, 12.10/mm², is the same quotient at 464 generators.

Consequences for § 103:

• The density term adds no structural weight beyond the recited array count, pitch and die footprint. It is an intended/calculated result. A reference disclosing the same four-column ≥100-nozzle array on a die of comparable footprint inherently meets it (In re Best; Continental Can inherency principles require that the result necessarily follow — here it does, as a matter of division).
• Framed as a design target, "at least 10.43 generators/mm²" is an optimization of a result-effective variable. Both inputs are classic design goals in the field: more drop generators (throughput/resolution) and smaller die (cost). KSR holds such a target obvious absent unexpected results; and the '676 specification itself frames the problem purely as a need for "an ink jet printhead that is compact and has a large number of ink drop generators" — the applicant's own stated problem statement supplies the motivation.

---

3. Ground 1 — Claims 1–4 and 28–31 over US 5,604,519 in view of the multi-column/narrow-printhead art

Primary reference — US 5,604,519 (Keefe et al., HP, 1997-02-18). As issued, it discloses:

• a printhead substrate bearing multiple firing chambers with a firing element in each (thin-film structure);
• first and second chamber arrays with chambers "spaced so as to provide greater than 400 dots per inch printing";
• grouping chambers into primitives with one firing element per primitive at a time;
• first circuit means on the substrate connected to the firing elements, controlled by primitive-select and address signals — i.e., integrated drive/selection circuitry on the die;
• second circuit means transmitting firing signals "at a frequency greater than 8/9/10/11/12 kHz" (claims 2–5, 7–9, 13–16).
  https://patents.google.com/patent/US5604519

Secondary references supply the only differences:

• US 5,469,199 (HP, "Wide inkjet printhead," 1995-11-21) and US 5,317,346 (HP, "Compound ink feed slot," 1994-05-31) / US 4,719,477 (HP, 1988-01-12) — thin-film substructure, barrier layer, orifice plate, and slot-fed ink feed with drop generators on one or both sides of a slot; these are the '676's own incorporated-by-reference foundations.
• US 6,011,112 (Lexmark, issued 2000-01-25 — 102(b)) and US 5,774,148 (Lexmark, 1998-06-30) — multiple nozzle columns/arrays on a single printhead die with integrated drive circuitry.
• US 5,144,341 (Xerox, 1992-09-01, "Thermal ink jet drivers device design/layout") — the specific idea of integrating the FET driver array alongside the heater-resistor columns on the printhead substrate, and of laying out driver fingers to conserve die area.

Motivation to combine: all references are in the identical field (thermal ink-jet printheads); the keefe '519 reference already states that closer nozzle spacing/higher resolution creates "space problems" and restricts barrier solutions — the same problem the '676 addresses. Combining a 400+ dpi multi-array architecture with slot feed (5,469,199/5,317,346) and integrated FET drive layout (5,144,341) to reach four columns of ≥100 nozzles at 1/300–1/600 in. pitch is a predictable use of prior-art elements with no change in their respective principles of operation (KSR). The 15–18 kHz firing range is a routine optimization above the 8–12 kHz expressly disclosed in the '519 reference, and is a direct function of chamber refill geometry (which the '519 reference is about).

Claim 1 vs. 28: substantively parallel ("for energizing" vs. "that energize"); the same combination disposes of both. Claims 2–4/29–31 (column pairs ≤630 µm; two slots with columns on either side; inboard columns ≤800 µm) are matters of placing the columns adjacent to their feed edges and packing them, i.e., routine geometry once four slot-fed columns are chosen — the 630/800 µm figures being tied to the barrier/feed channel dimensions of the same HP designs.

---

4. Ground 2 — Claims 5–27: add the ground-bus-overlap and FET-efficiency art

Claim 5's only addition over claim 1 is structural: "columnar arrays of FET drive circuits … respectively adjacent said columnar arrays of drop generators, and ground busses that overlap active regions of said FET drive circuits."

US 6,309,053 B1 (HP, filed 2000-07-24; granted 2001-10-30) — "Ink jet printhead having a ground bus that overlaps transistor active regions." Claim 1 recites exactly: a columnar array of ink drop generators; a columnar array of FET circuits with active regions of drain/source/gate; power traces including a ground bus generally extending along the longitudinal extent of the FET array and partially overlying said active regions. Claims 2–3 add the varying/increasing bus width; claim 6 adds FET circuits configured to compensate for variation in parasitic resistance of the power traces.
https://patents.justia.com/patent/[6309053](/patent/6309053)

Motivation to combine (this is the strongest motivation in the record): the '053 reference states the result the '676 claims as its point of novelty: the overlap "allows the ground bus and FET circuit arrays to occupy narrower regions which in turn allows for a narrower, and thus less costly, thin film substructure." The '676's own specification says the same thing verbatim. A narrower peripheral FET/logic strip at a fixed nozzle count raises nozzle packing density — which is precisely claim 1/5/28's final limitation. This is an explicit teaching of the problem (die size drives cost) and the solution (overlap), directed at the same structure. Under KSR, an express teaching that a known modification achieves the claimed benefit makes the combination obvious; and the Federal Circuit's 2026 Slingshot v. Canon affirmance shows a fact-finder accepting essentially this rationale for this family's art (Torgerson + a second-metal ground/power conductor overlapping the transistor/logic area).

Remaining dependent claims map to art already in the page's lists:

---

5. The decisive legal caveat — pre-AIA § 103(c) may disqualify the best references

The HP "Similar Documents" (US 6,309,053; US 6,412,917; US 6,398,347; US 6,523,935; US 6,488,363) are the substantively most on-point art on the page — they are sibling HP applications whose specifications contain the '676's very numbers (≥100 nozzles/column, 1/300 in. pitch, 250/180 µm FET widths, ≥100 Ω resistors, the Ron < 250,000/A formula, ground-bus overlap, parasitic-resistance compensation). But:

• They were filed/issued in 2000–2003. None is 102(b) or 102(a) art against the '676. At best they are § 102(e) art (filed before 2001-06-19).
• Pre-AIA § 103(c)(1) provides that subject matter that qualifies as prior art only under § 102(e)/(f)/(g) shall not preclude patentability under § 103 where the art and the invention were commonly owned at the time the invention was made. These are all HP (Hewlett-Packard Company / HP Development Company) filings — the same assignee as the '676.
• Two further wrinkles: (i) where the inventive entity is identical (US 6,523,935 lists the same six inventors as the '676 — Torgerson, Browning, MacKenzie, Miller, Bakkom, Dodd), the reference is not "by another," so § 102(e) does not apply at all; (ii) US 6,488,363 was filed 2001-12-17, after the '676 filing date, so it is not § 102(e) art either.

Bottom line: a challenger wanting to invalidate the '676 must lead with the 102(b)-dated art (Ground 1: US 5,604,519 + US 5,469,199/US 5,317,346 + US 5,144,341 + US 6,011,112/US 5,774,148), and may at most argue the HP siblings as evidence of the level of ordinary skill or as non-prior-art context. Conversely, a patent owner defending the '676's validity must confront the fact that its own specification is, word for word, the same document that later courts/PTAB treat as sufficient art to invalidate a related patent. If HP's siblings were to be counted, claim 5 and claims 6–27 would be very difficult to sustain; if § 103(c) excludes them, the case is closer but still favors obviousness because of the '519 + '053 combination and the arithmetic nature of the density term.

---

6. Secondary considerations

Nothing in the record supplied supports objective indicia for the '676 specifically:

• No unexpected results are asserted for the 10.43/mm² figure; it is a computed quotient.
• No long-felt need / failure of others evidence appears; the record shows the opposite — the whole HP family (the '676, '053, '917, '347, '935, '363) implements the same architecture essentially contemporaneously, which is evidence of simultaneous, routine engineering, not of a breakthrough.
• No copying is alleged anywhere in the record. (Treat the later Lexmark/Canon patents in the "Cited By" list — US 7,195,341, US 7,240,997, US 7,784,914 — as mere citation evidence, not as prior art and not as copying evidence.)

---

7. Probable outcome, claim by claim

Overall: claims 1, 5 and 28 — the three independent claims — are the most vulnerable, because their point of novelty is a densely packed four-column, slot-fed, FET-integrated printhead, every element of which is separately known, with an express motivation (die cost) explicitly stated in the prior art and in the applicant's own background. The dependent claims add only numeric design targets that appear in the same HP family's disclosures and in third-party art.

---

8. Explicit uncertainties

1. I could not verify, from the page alone, the filing dates and inventive entities of each HP sibling application; those dates drive the § 102(e)/§ 103(c) analysis in § 5 and should be confirmed in Patent Center before relying on any conclusion there.
2. The PTAB/CAFC record I cite (IPR2022-01414; Fed. Cir. No. 24-1956) concerns US 7,195,341, which uses a Torgerson-family reference as art. I found no proceeding naming US 6,422,676 itself as an asserted or challenged patent — consistent with the earlier litigation summary, and consistent with the '676 having expired 2021-06-19.
3. Sections III/IV of the '676's own recitations (e.g., the exact 630 µm and 800 µm column spacings, the 10.43–12.10/mm² band) are treated here literally as recited; I have not independently re-measured them.