US 5377002

I'll search for the specific patent number and check for any litigation or 2026 docket activity.

Let me verify litigation history and the related family members, and check on the second independent claim.

US Patent 5,377,002 — Analyst Summary

Searches performed: Google Patents full text, Justia Patents full text, USPTO assignment records (as surfaced via the patent page), and open-web searches for Federal Circuit / CAFC docket activity for the exact number 5377002. Findings below are grounded in those results; where I could not confirm something, I say so.

1. Bibliographic data (as reported; interpreted literally)

Source: https://patents.google.com/patent/US5377002/en and https://patents.justia.com/patent/5377002

Assignment chain shown in the record (quoted literally):

1. 1992-07-14 — Assigned to TET Techno Investment Trust Settlement (assignors: Auderset, Malin, Nesensohn, Sawatzki, Schmid, Steigmeier).
2. 1994-07-18 — Assigned to Tencor Instruments (assignor: TET Techno Investment Trust Settlement).
3. 2003-08-04 — Assigned to Merlin Technology, Inc. (assignor listed as Digital Control Inc.).

⚠️ Uncertainty flag: The 2003 entry is internally odd — the assignor is "Digital Control Inc.," which does not appear earlier in this patent's chain, and Google Patents expressly disclaims that its assignee list "may be inaccurate." I am reporting it literally and not asserting that Merlin Technology owns this patent as a matter of fact.

Patent family (from the WO/EPO search-report family listing): CH 685650 A5; EP 0524348 A; EP 0525286 A/B; AT 142783 T; DE 59207100 D; JP 6167457 A; JP 7103906 A; US 5377001 A (a sibling granted the same day). Note the family includes US 5,377,001 — a different patent number; do not conflate it with 5,377,002.

2. Abstract (verbatim, condensed)

The apparatus permits non-destructive examination of entire surfaces for defects and contamination, and can detect microscopically small dot-shaped and linear defects and extremely fine macroscopic non-homogeneous areas. An adjustable lens system (5) sits in the optical path between the light source (2) and the objective (9) and produces various intermediate images (31). A first cigar-shaped intermediate image is used for a first scan of the whole surface at a relatively large feed offset; a second dot-shaped intermediate image is used for a second scan of partial areas at a small feed offset. A dark-field stop assembly (18) with adjustable dark-field deflection system (8) lies between the lens system and the objective and projects the beam exactly centered at right angles through the objective onto the surface. Reflected light collected by the objective is projected onto a photodetector. An electronic analysis system (21) breaks the amplified photodetector output into measured values attributable to dot-shaped, linear, and planiform defects, and is connected via a computer unit (22) to peripheral equipment (23, 24, 25) for representing all measured values from a measuring cycle.

3. Plain-language overview of the independent claims

The patent has two independent claims: claim 1 and claim 26. (Claims 2–25 depend from claim 1; claims 27–28 depend from claim 26.)

Independent Claim 1 — the "two-resolution scanning inspector"

A machine for high-sensitivity measurement of three classes of surface defects (dot-shaped/point, linear, and planiform/haze). It combines:

• a light source producing a beam, and a supporting disk holding the object with its surface exposed;
• an objective that passes the beam perpendicularly onto the surface;
• drive means that rotate and translate the disk relative to the beam, so the beam scans the surface along a spiral path;
• a photodetector receiving light reflected by the surface and collected by the objective;
• adjustable lens system means between source and objective that can produce a first intermediate image (for a first scan) and a second intermediate image (for a second scan), with the drive's translation adjustable so the spiral feed offset changes according to which intermediate image is selected — i.e., a fast coarse whole-surface scan and a slower, finer partial-area scan from the same hardware;
• a dark-field stop assembly with an adjustable dark-field deflection system between the lens system and the objective, centering the deflected beam exactly through the objective and at right angles onto the surface (this is what makes the reflected/specular light return and the scattered light be collected — a dark-field geometry);
• a confocal diaphragm before the photodetector;
• vignetting optics placed along the image-forming optical axis between the dark-field stop assembly and the confocal diaphragm, positioned to selectably introduce optical asymmetry into that axis so the apparatus can distinguish between surface defects (e.g., oriented vs. random haze — the specification's rod-shaped vignetting device for polishing defects);
• electronic analysis means plus a computer unit that break the photodetector output into measured values for dot-shaped, linear, and planiform defects; and
• means for determining the beam's effective scanning position at any time, tied to the computer unit, so each measurement is tagged with where it was taken.

In short: a dark-field, spiral-scan wafer/substrate inspector whose single optical train can be switched between a large astigmatic (cigar) spot for fast full-surface survey and a small round spot for high-resolution inspection of a chosen region, with selectable image asymmetry for texture discrimination, and with per-measurement position tracking.

Independent Claim 26 — the same machine, defined by its signal-processing architecture

Claim 26 recites substantially the same physical combination (light source, supporting disk, objective, rotational/translational drive producing a spiral scan, photodetector, switchable intermediate-image lens system with image-dependent feed offset, and an adjustable dark-field stop assembly centering the beam). It then adds the analyzer-electronics architecture as claim elements:

• the electronic analysis means has a signal input connected to the photodetector output, feeding a haze channel comprising a haze filter and a peak suppressor;
• the same signal input feeds a particle channel comprising a subtraction circuit and a peak detector with a digital output;
• the haze-filter output goes to an analog-to-digital converter with its own output;
• the haze filter is connected to a nominal-bandwidth input, and the peak detector to a second input;
• the result: haze (planiform) measurements are available at the A/D converter output, while LPD (dot-shaped) and linear-defect measurements are available at the peak detector's digital output, all loaded and stored in the computer unit.

In short: claim 26 is the "how do you separate haze from particles electronically" claim — a two-channel (slow low-pass "haze" channel with peak suppression + subtraction-fed peak-detecting "particle" channel) front end, with a bandwidth-control input that lets the filter time constant track radius/rotation speed.

4. Claim-drafting anomalies you should be aware of (reported literally, not corrected)

• Claims 2 and 13 each recite "An apparatus in accordance with claim 31" — but this patent has only 28 claims, so no claim 31 exists. This is a defect in the printed patent. I am not auto-correcting it to "claim 1."
• Claims 26 and 28 are the only claims I could retrieve from the second (claim-26) family; the Google Patents rendering truncates mid-claim 26 ("...correspondingly changing the offset of each"), and I could not retrieve the text of claim 27 with confidence. Treat claim 27 as unverified.
• Claim 26 as retrieved via Justia does not appear to recite the confocal diaphragm or the vignetting optics that claim 1 recites, while it does recite analyzer-electronics elements that claim 1 does not. If accurate, the two independent claims are not coextensive — claim 1 is optical/dark-field-vignetting-centric; claim 26 is signal-channel-centric. ⚠️ I flag this as an observation on the retrieved text, not a legal conclusion.
• Text artifacts in the record: "shave" (for "shape") in claim 26; "Aystem" (for "system") in the FIG. 5b description. These appear to be OCR/printing artifacts of the source documents.

5. Federal Circuit / CAFC 2026 docket — search result: none found

I searched for 5377002 in connection with CAFC 2026 dockets and found no Federal Circuit appeal, IPR appeal, or district-court case involving US 5,377,002.

Two reasons to be confident this is a genuine negative rather than a search miss:

1. The patent's listed anticipated expiration is July 14, 2012, and its status is Expired – Lifetime. A patent that expired in 2012 cannot generate a 2026 infringement judgment (2026 appeals would involve unexpired or recently litigated patents).
2. My searches for KLA-Tencor/ADE, Nanometrics, Xitronix, and Verity Instruments litigation surfaced the successor company's other patents and cases, but not 5,377,002 specifically.

⚠️ Caveat: The Federal Circuit's own docket system (and PACER/CM-ECF) is not fully indexed by open web search. I cannot certify the absence of every 2026 filing; I can only report that no public result connects the specific number 5,377,002 to a 2026 CAFC docket. If you need a certified negative, that requires a direct PACER/CAFC CM-ECF docket query, which is outside what I could do here.

6. Prior art and citation context (for completeness)

• The patent's own PRIOR ART discussion cites U.S. Pat. No. 4,314,763 as the closest apparatus (spiral-scan dark-field surface inspector with counter and CRT).
• US 5,377,002 was cited as an "A" (background) reference in the ISR for PCT/IB99/01188 / WO 00/03259 (Philips, Land), at column 9 line 8 – column 13 line 62.
• It is also cited as background in US 7,858,911 (Confocal wafer inspection system and method — "5377002 | December 27, 1994 | Malin et al.").
• Do not confuse with the "537 patent" referenced in the Netlist v. Samsung line of cases — that is a different patent (a '537-numbered patent), not 5,377,002.

7. Bottom line

US 5,377,002 is a 1994 Tencor Instruments–era (originally TET Techno Investment Trust Settlement) Swiss-origin patent by Malin et al. covering a dark-field, spiral-scan surface/wafer inspection apparatus whose distinguishing feature is a switchable intermediate-image lens system enabling one optical train to do both a fast, coarse whole-wafer scan (astigmatic cigar spot, large feed offset) and a high-resolution partial-area scan (dot spot, small feed offset), plus selectable vignetting optics for texture discrimination and per-measurement position registration; the second independent claim (26) adds the two-channel haze/LPD signal-separation electronics. It expired July 14, 2012, and I found no 2026 CAFC docket activity — consistent with an expired patent, though I could not perform a direct CM-ECF docket verification.

I'll research litigation involving US patent 5377002 across multiple sources.

Let me search more targeted litigation sources.

Let me try the litigation-specific databases and patent-troll databases.

Summary Answer

I could not identify any litigation — district court, ITC, PTAB, or CAFC — in which U.S. Patent No. 5,377,002 was asserted or challenged.

Based on the searches performed, there is no record of an infringement suit, declaratory judgment action, IPR/PGR, or appeal involving this specific patent. I want to be explicit that this is a negative finding from limited-search sources, not proof of absence — see "Confidence and Limitations" below.

What the patent is (for identification purposes)

Source: https://patents.google.com/patent/US5377002/en

Cases I expressly screened out (NOT involving 5,377,002)

Searchers often confuse this patent with the Tencor/KLA-Tencor metrology litigation cluster. I found those cases, but none of them is documented as asserting 5,377,002:

• Xitronix Corp. v. KLA-Tencor Corp., W.D. Tex. (Austin), filed 2008-09-24 — KLA-Tencor asserted the '441 and '690 patents (and sought an injunction); jury found the asserted claims invalid and no infringement; judgment 2011-02-02. Follow-on state tort suit: Xitronix Corp. v. KLA-Tencor Corp., No. 03-12-00206-CV (Tex. App. 2014); CAFC jurisdiction decision at 882 F.3d 1075 (Fed. Cir. 2018). Patents at issue: '441 and '690 — not 5,377,002.
• Nanometrics Inc. v. KLA-Tencor Corp., N.D. Cal. (Judge Jeffrey White) — filed 2005; three KLA patents; case stayed pending reexamination. Not 5,377,002.
• ADE Corp. v. KLA-Tencor Corp. — filed 2000-10-11, D. Del.-area docket; ADE '525 patent asserted. Not 5,377,002.
• KLA-Tencor Corp. v. Tokyo Seimitsu Co. / TSK America Inc., N.D. Cal., 2001 — photomask/reticle inspection patent. Not 5,377,002.

Confidence and Limitations

• My searches surfaced mostly false positives: the numeric string "5377002" appears as an Israeli telephone number, a Brazilian corporate registration, and a CourtListener financial-disclosure investment ID (not a case). I excluded these.
• I was unable to complete direct queries of the Unified Patents litigation portal, Docket Alarm, or PACER/CAFC dockets; the searches returned the sources' noise rather than patent-specific hits, and my search budget was exhausted before I could run further targeted queries.
• 5,377,002 expired on 1994-12-27 + 17 years (i.e., 2012-07-14 anticipated expiration), which is consistent with there being no modern assertion campaigns. The 2003 assignment to Merlin Technology, Inc. / Digital Control Inc. is an assignment record, not a litigation record.
• One drafting anomaly worth flagging for any future diligence: original claims 2 and 13 depend from "claim 31," which does not exist in the granted 28-claim set. This is a claim-dependency defect that would have been squarely relevant to any validity or construction dispute — but I found no proceeding raising it.

Bottom line: No known litigation involving US 5,377,002. If your diligence needs to be definitive, I'd recommend running docket-level searches (PACER Docket Search + CourtListener RECAP for "5,377,002" and "5:377002"; Unified Patents Portal and Docket Alarm patent-page lookups) — those are the authoritative sources I could not fully query here, and they should be treated as the controlling check rather than my negative finding.

I'll verify the structured-data finding and check for any AIA trial activity the ODP may not have indexed.

The first pass returned only false positives (other "'002" patents). Let me run tighter queries keyed to the exact patent number and its owner.

Let me run one final confirmation query plus check the sibling patent in the family.

Proceedings overview

Zero AIA trial proceedings — no IPR, PGR, or CBM — have been filed against U.S. Patent No. 5,377,002: total count 0 (active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0). This is consistent with the structured "PTAB proceedings on file" block, which reports that the USPTO Open Data Portal returns no AIA trial proceedings for this patent as of the most recent ingest, and with my independent web searches, which surfaced only false positives (see below). The bottom-line defensive posture is unusual and, if anything, favorable in a different way than a hardened patent: there is no claim canceled, no estoppel to leverage, but also no modern assertion campaign — and, critically, the patent is expired, which is almost certainly why no AIA trial exists. A defendant's strongest position here is temporal (expiration + no PTAB history), not a PTAB kill shot.

Proceedings detail

(none) — no proceedings on file

There is no proceeding number, panel, institution decision, Final Written Decision, settlement, or Federal Circuit appeal to report for this patent. I will not invent one. What follows is why that absence is structurally expected, and what I screened out.

Timing analysis (the decisive point). The AIA trial regime became available on 2012-09-16. The '002 patent's anticipated expiration is 2012-07-14 (Google Patents: https://patents.google.com/patent/[US5377002](/patent/US5377002)/en). In other words, the IPR/PGR window opened roughly two months after the patent's term ended on its anticipated expiration basis. That is the single best explanation for the void:

• IPR — 35 U.S.C. §§ 311–319 became available 2012-09-16. An IPR can technically be filed against an expired patent (the Board has done so where past damages are in play), but with no documented infringement suit and no live enforcement, there was no practical petitioner.
• PGR — limited to patents issuing from applications subject to first-inventor-to-file (effective 2013-03-16), and must be filed within 9 months of grant. The '002 patent issued 1994-12-27; it is categorically ineligible for PGR.
• CBM — limited to patents claiming a "financial product or service" and, for a "technological invention," subject to the § 18(d)(2) carve-out. A dark-field optical wafer-inspection apparatus (claims 1–28, e.g., "cylindrical lens," "confocal diaphragm," "dark-field stop assembly") is not a CBM candidate. The CBM transitional program has also since sunset (2020-09-16).
• § 315(b) one-year bar — an IPR must be filed within one year of service of a complaint alleging infringement. No such complaint is documented (see the Litigation section), so the trigger never started.

False positives I screened out (NOT this patent). Searching "002" + PTAB reliably produces other patents:

• U.S. Patent 7,810,002 — Qualcomm secure-debug IPRs (Giles/Moyer art). Different patent entirely.
• Intellectual Ventures '002 patent — the mobile-interface patent litigated in IV v. Capital One / IV v. Erie Indemnity, raised in CBM2014-00035. Different patent.
• Numeric-string noise — "5377002" appears as an Israeli phone number (mssociety.org.il), a Brazilian corporate registration (JUCEC), and misc. document IDs. Excluded.
• The only substantive web hit referencing US 5,377,002 is a prior-art citation in a WO international search report (WO2000003259A3) citing "US 5377002 A (C. MALIN ET AL), 27 December 1994 … column 9, line 8 – column 13, line 62." That is a citation by a later applicant, not a PTAB challenge.

Confidence/limitation. I could not directly query PTAB E2E / P-TACTS or P-TACTS/PACER docket APIs within this session; searches returned the portals' general content rather than a patent-specific record. My negative finding should therefore be treated as strong-but-not-conclusive and confirmed against the authoritative sources below.

Strategic summary

Claim status: nothing is CANCELED, nothing is SUSTAINED by the Board, and all 28 claims are effectively UNTESTED at the PTAB — because no AIA trial ever reached the merits. Under the "expected expiration" reading, the enforceable life of the patent ran out (2012-07-14) before the trials bar even opened (2012-09-16). Note the claim-set itself is untested but not clean: original claims 2 and 13 depend from "claim 31," which does not exist in the granted 28-claim set (confirmed against the full claim text in the record above). That § 112 ¶ 4 dependency defect is the kind of issue a petitioner would normally press — but § 112 is unavailable in IPR (limited to §§ 102/103 on patents/printed publications), which means it could only have been raised in a PGR (unavailable) or in district court. It was never raised at the PTAB.

Estoppel landscape: § 315(e)(2) is a blank slate. Because no IPR/PGR was instituted and no FWD issued, no petitioner (and no privy) is estopped from raising any § 102/§ 103 ground. There is likewise no § 325(e) estoppel from a PGR. Practically, this means a defendant is not foreclosed from running an IPR today — but the § 315(b) bar is un-ignitable without an infringement complaint, and the new Director institution policy (2025-10-16) centralizes institution with the Director and the proposed rules would bar IPRs against patents that already survived a validity challenge or where a parallel proceeding will decide validity first. An expired patent with no live litigation is a poor candidate even under the old regime.

Pattern signals: none. No single petitioner filed multiple petitions; no patent owner pursued PTAB appeals (there was nothing to appeal — no IPR FWD is an appellate event); and there is no defensive aggregator (Unified Patents, RPX, etc.) in the chain. The 2003 assignment to Merlin Technology, Inc. / Digital Control Inc. in the Google Patents record is an assignment, not a litigation or PTAB event — do not read it as enforcement activity. The absence of any IPR over a 30-year-old, once-commercial Tencor-family patent is itself the signal: patents that are actually asserted attract IPRs; this one left no such footprint.

Recommended next steps

1. Confirm the null result at the authoritative source before relying on it. Run the patent number (5,377,002 and US5377002) directly in:

   • USPTO PTAB E2E / P-TACTS (https://ptab.uspto.gov) — search by patent number across IPR/PGR/CBM.
   • USPTO Patent Trial and Appeal Board Decisions (https://www.uspto.gov/patents/ptab/decisions) and P-TACTS public search.
   • CourtListener / RECAP (https://www.courtlistener.com) for "5,377,002," "5377002," and "5:377002" (case and document text).
     These are the controlling checks; my finding is a strong negative, not proof of absence.

2. For a defendant currently facing a demand letter citing the '002 patent: the decisive facts are (a) anticipated expiration 2012-07-14 (term expired), and (b) no PTAB history and no documented litigation — so there is no FWD to cite and no canceled claims to quote. Do not represent that any claim has been invalidated at the PTAB; it hasn't. Instead, verify the expiration/term status (including any PTA/terminal-disclaimer adjustments) and confirm there are no live family members (US 5,377,001; CH 685650 A5; EP 0524348; EP 0525286; JP 6167457; JP 7103906) still being asserted.

3. If you want a PTAB backstop anyway (e.g., past-damages exposure): an IPR can be filed against an expired patent, but you will face (i) no § 315(b) trigger unless you've been served, (ii) the 2025 Director institution policy and proposed anti-duplicative rules, and (iii) the practical argument that there is no live infringement dispute to resolve. Budget accordingly and expect a discretionary-denial fight rather than a merits win.

4. Preserve the § 112 dependency defect — claims 2 and 13 depending from a non-existent "claim 31" — as a district-court invalidity/indefiniteness theory. It is unavailable in IPR, so it will not appear in any PTAB record; it lives only in litigation.

Bottom line for the record: No IPR. No PGR. No CBM. No institution decision. No FWD. No PTAB appeal. The patent expired on its anticipated date before the AIA trial window opened — that timing, not any Board ruling, is the dominant fact a defendant should build on.

I'll research the assignment chain for US 5,377,002, starting with the recorded transfer events and the entities involved.

Let me dig into the specific assignment records for this patent number and the entities involved.

I've hit the tool-call limit. Here's the assignment-chain reconstruction based on the authoritative patent record I have (Google Patents legal events, which are sourced from USPTO assignment data) plus the search results obtained. I want to be explicit up front: I could not open the USPTO Assignment Center / assignment.uspto.gov record pages themselves, so no verified reel/frame numbers appear below. Everything is flagged accordingly rather than filled in.

Inventors

Six named inventors (source: Google Patents; Justia patent page for 5,377,002):

Employer at time of filing: not determinable from the patent record. The residences cluster tightly in the Rhine valley at the Switzerland/Liechtenstein/Austria border (Mauren and Schaan are Liechtenstein; Batschuns and Thüringen are in Vorarlberg, Austria; Hedingen and Aeugst are in the Zürich/Zug corridor). That geographic clustering is consistent with a single cross-border employer in the Liechtenstein/Swiss instrumentation industry, but the patent and assignment records do not name an employer, so I will not name one. Unclear.

Departure pattern: The 1992-07-14 assignment to "TET TECHNO TRUST INVESTMENT SETTLEMENT" lists all six inventors as assignors on the same day the application was filed (1992-07-14). That is the signature of a standard employee obligation-to-assign, not of post-filing departures. I found no evidence of inventors leaving within 12 months of filing, and no inventor-level corporate data source was accessible — unclear/not present.

Original assignee

TET Techno Investment Trust Settlement (also rendered "TET Techno Trust Investment Settlement" in some citation tables) is named as assignee on the face of the patent (Justia: "Assignee: Tet Techno Trust Investment Settlement") and is the assignee of record for the whole family — e.g., sibling US 5,377,001 ("Apparatus for surface inspection") and unrelated-looking portfolio patents such as US 5,133,635 ("Method and apparatus for holding and conveying platelike substrates," TET Techno Investment Trust Settlement). This indicates TET Techno was a portfolio-holding vehicle (a "Trust Settlement," i.e., a trust-form entity), not an operating manufacturer.

• Primary line of business: patent holding / technology investment (trust). No product literature, no SEC filings, no commercial catalog associates TET Techno with a shipped product.
• Product embodying the claims: the apparatus itself — a dark-field laser spiral-scan wafer surface inspection tool — corresponds to the commercial field later occupied by Tencor's Surfscan line (Tencor Instruments launched Surfscan in 1984 and it remains a KLA product family today; see KLA's fiscal 2025 10-K product table listing "Surfscan® Series"). I cannot attribute a specific TET Techno product.
• Current status: No evidence of current operation. The entity's only visible activity is the 1991–1992 patent filings and the 1994 transfer out to Tencor Instruments. Status after 1994 is not documented in the sources I could reach — treat as unclear/likely wound down. (Do not over-read: this is absence of evidence, not proof of dissolution.)

Assignment timeline

Recorded events, per Google Patents "Legal events" for US 5,377,002 (https://patents.google.com/patent/US5377002/en). Google Patents does not expose the reel/frame in the text I retrieved, and I could not open the Assignment Center record pages, so reel/frame is listed as "not retrieved" throughout — I will not invent one.

• 1992-07-14 (executed) / recorded 1992-07-14 — Reel not retrieved

  • Conveyance: Assignment (initial; "ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Auderset, Heinrich; Malin, Comas; Nesensohn, Thomas; Sawatzki, Harry L.; Schmid, Gert; Steigmeier, Edgar F. (all six inventors)
  • Assignee: TET Techno Investment Trust Settlement
  • Correspondent: not displayed in the retrieved record — not retrieved
  • Context: acquisition at filing — inventors assigned to the trust vehicle on the application's filing date (employment/obligation-to-assign).

• 1994-07-18 (recorded; execution date not displayed) — Reel not retrieved

  • Conveyance: Assignment (reassignment)
  • Assignor: TET Techno Investment Trust Settlement
  • Assignee: Tencor Instruments (Mountain View, CA)
  • Correspondent: not retrieved
  • Context: portfolio acquisition / fire-sale-out of the trust — the patent moved from the holding vehicle to the operating semiconductor-inspection company that manufactured the Surfscan product line. This is the substantive ownership event.

• 2003-08-04 (recorded per Google Patents legal events) — Reel not retrieved

  • Conveyance: Assignment (reassignment)
  • Assignor: Digital Control Inc.
  • Assignee: Merlin Technology, Inc.
  • Correspondent: not retrieved. A similar-looking but not identical record — Plainsite id 7292969, "Patent Assignment from Digital Control Inc. to Merlin Technology Inc.," reel 40330/891, executed 2003-05-01, recorded 2016-11-15, correspondent PRITZKAU PATENT GROUP, LLC, 993 Gapter Road, Boulder, CO — is attached to application 15/352,363 (a Horizontal Directional Drilling locator), not to 5,377,002. The execution/recording dates do not match the 2003-08-04 Google Patents event. I therefore treat the 2003-08-04 event on US 5,377,002 as an unverified anomaly and do not assert the PritzKau correspondent against this patent.
  • Context: unresolved / likely data artifact. Digital Control Inc. and Merlin Technology Inc. are Renton/Kent, WA horizontal-directional-drilling businesses (see Digital Control Inc. v. Radiosection/Charles Machine Works, and the Stanford NPE database entry listing Merlin Technology as a "Product company"). Neither has any connection to wafer surface inspection. Because Tencor merged into KLA-Tencor in April 1997 — before this 2003 event — and there is no record of KLA transferring this patent to a drilling company, the "assignee = Merlin Technology, assignor = Digital Control" entry is almost certainly a mis-merged USPTO/Google Patents assignment record, not a real transfer of 5,377,002. Flag this to any diligence team.

• 2012-07-14 — anticipated expiration (legal-status event, not a transfer). Consistent with a filing date of 1992-07-14 + 20 years.

If a verified ledger is needed, this section must be re-run against https://assignmentcenter.uspto.gov/ keyed to patent 5,377,002. I could not complete that step.

Timeline diagram

timeline
    title Ownership of US 5377002
    1991 : Priority CH 2154/91
    1992 : Filed
         : Six inventors assign to TET Techno
    1994 : Granted
         : TET Techno assigns to Tencor Instruments
    1997 : Tencor merges with KLA
    2003 : Merlin Technology event in records
         : Unverified likely data artifact
    2012 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The direction of travel is the opposite of the NPE pattern. The patent moved from a holding trust to an operating manufacturer, Tencor Instruments (recorded 1994-07-18). No LLC with "IP / Holdings / Licensing" naming appears as a downstream assignee.
2. Known asserter in the chain — NOT PRESENT. Neither Tencor Instruments, KLA-Tencor, Digital Control Inc., nor Merlin Technology appears on the asserter lists named in the prompt (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, etc.). Stanford's NPE database classifies Merlin Technology/ Digital Control as "Product company," and KLA-Tencor is a NASDAQ-listed capital-equipment maker. No match to any high-frequency-plaintiff list.
3. Repeat correspondent across the chain — UNCLEAR. I could not retrieve the correspondent of record for either the 1992 or 1994 recording, and the only correspondent surfaced (PritzKau Patent Group, Boulder, CO) belongs to a different assignment record for a different application. A single would-be appearance is not a finding, and here it isn't even confirmed to be this patent's record.
4. Cascading transfers (<24 months through chained LLCs) — NOT PRESENT. The documented chain is two transfers spread across 1992 → 1994 (and a disputed, likely-spurious 2003 entry). No rapid multi-hop LLC sequence, no shared correspondent address, no common principals.
5. Pre-litigation transfer (assignment within 6 months of a suit) — NOT PRESENT. The prior litigation screen (see Litigation summary) found no suit, ITC action, or PTAB proceeding naming 5,377,002. There is no suit for a transfer to precede.
6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 record was found for TET Techno or Tencor. Tencor's exit was a merger (KLA Instruments + Tencor → KLA-Tencor, April 1997; LA Times 1997-01-15; KLA 10-K), not a liquidation.
7. Privateering — NOT PRESENT. No operating company → NPE transfer asserted on the operating company's behalf. The 1994 transfer runs toward the operating company, not away from it.
8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

Verdict

Insufficient data.

The records I could reach show a benign, operating-company chain with zero NPE signals: TET Techno Investment Trust Settlement (holding trust) → Tencor Instruments (recorded 1994-07-18, operating manufacturer of the Surfscan wafer-inspection line) → KLA-Tencor via the April 1997 merger, with the patent expiring 2012-07-14. Nothing in that chain resembles an NPE campaign, and the earlier litigation screen found no assertion. I nevertheless decline to render "operating-company assertion" or "defensive/non-asserting" because (a) the 2003-08-04 "Merlin Technology, Inc. / Digital Control Inc." legal event on this patent is almost certainly a mis-merged record and cannot be resolved without the Assignment Center, and (b) I could not retrieve the reel/frame or correspondent for any link — the very data the NPE test depends on. Re-run the ledger against the Assignment Center before relying on this for diligence: https://assignmentcenter.uspto.gov/ (search "5377002").

I'll research the prior art cited against US 5,377,002, starting with the patent's citation records.

Let me dig into the citation record specifically.

Let me pull the citation tables from secondary databases.

Scope and source status (read this first)

I could not complete the requested citation-table retrieval. Here is exactly what I have and what I could not get, so you can calibrate:

• The authoritative full text supplied for US 5,377,002 (from https://patents.google.com/patent/US5377002/en) contains the specification, claims, classifications, and "Prior art keywords," but the fetched payload does not include the front-page "References Cited" / "Patent Citations" table. That table is the primary answer to your question.
• My search steps for the citation tables on Espacenet, Justia, FreePatentsOnline, and Patent Center were exhausted before I could retrieve them. The searches returned adjacent documents (search reports belonging to other patents, forward citations, unrelated hits) rather than the '002 reference list.
• Therefore I am not going to enumerate front-page references I cannot verify. Inventing or "recalling" a PTO-1449 list would be the single most damaging error possible in a prior-art analysis. Everything below is labeled by evidentiary status.

A. What the record definitively establishes

The '002 specification expressly cites exactly one prior-art patent. In the "PRIOR ART" section, it states:

"U.S. Pat. No. 4,314,763 relates to an apparatus with the above attributes, in which a beam of light from a light source in the form of a laser is projected via two prisms and an objective perpendicularly upon the surface of an object..."

Source: https://patents.google.com/patent/US5377002/en (Background/Prior Art section)

That is the applicant's own admitted closest prior art. Everything else in the "References Cited" table (if any) is unverified in this session.

Note also the front-page "Prior art keywords" field: light beam / accordance / dark / light / objective — this reflects the concept set (dark-field optics, beam/objective geometry) the '002 is measured against, not a citation list.

B. Reference 1 — U.S. Pat. No. 4,314,763 (admitted prior art)

§ 102 claim mapping (element-by-element against claim 1):

Anticipation conclusion: U.S. 4,314,763 anticipates at most the preamble / generic apparatus class of claim 1 — the features the specification itself concedes as "the above attributes." It does not anticipate claim 1 or any of claims 2–28, because every one of those claims carries at least the adjustable lens-system limitation, and claim 1 additionally requires the dark-field stop assembly, confocal diaphragm, vignetting optics, and the analyzer/computer position-association chain. Realistically this reference is § 103 art, not § 102 art, and it is a § 102 reference only as to unclaimed background matter.

C. Family / co-pending documents — considered and rejected as § 102 art

These come from the family listing carried over from the earlier litigation section and my searches. They are flagged here because they are the documents most often miscited as prior art to the '002:

Because the '002 was filed 1992-07-14 with a 1991-07-20 priority, the § 102 window runs against pre-1991-07-20 disclosures (plus pre-AIA § 102(e) US applications filed before the invention date and granted later). The siblings filed the same day fail that "before" test.

D. Confirmed forward citations — expressly NOT prior art

These cite the '002; they are relevant to the patent's technological impact, not its validity. Listing them to prevent them being mistaken for cited prior art:

• KR 100808702 B1 — "A pixel-based method, apparatus and optical head for defect detection on semiconductor wafer surfaces," which lists: US5377002A (en) * 1991-07-20 | 1994-12-27 | Tet Techno Trust Investment Settlement | Apparatus for surface inspections. Source: https://patents.google.com/patent/KR100808702B1/en
• CN 100347584 C (Electro Scientific Industries; PCT/US2004/002284 → WO 2004/077336, US 10/373,934) — its [56] 参考文献 block lists US5377002 A 1994.12.27 among cited references. Source: https://patentimages.storage.googleapis.com/47/f7/8d/3d91e22997a47e/CN100347584C.pdf

Both are later documents citing the '002, i.e., the reverse direction.

E. Bottom line on § 102 exposure

1. No reference verified in this session anticipates any claim of US 5,377,002. The only confirmed cited patent, U.S. 4,314,763, anticipates nothing beyond the claim 1 preamble.
2. The '002's novelty as issued appears to rest on the combination the specification emphasizes: switchable adjustable lens system (cylinder/sphere) producing two different intermediate images with correspondingly changed feed offset, plus the rotationally symmetrical adjustable dark-field stop assembly, confocal diaphragm, vignetting optics, and the haze/LPD decomposition + polar-to-Cartesian look-up-table position-association electronics. No single one of those is shown in '763.
3. Drafting anomaly to carry forward: original claims 2 and 13 recite dependency on "claim 31," which does not exist in the granted 28-claim set (claim 2 → "in accordance with claim 31"; claim 13 → "in accordance with claim 31"). When you map prior art to claims, map claim 2's lens limitations to claim 1, and claim 13's look-up-table limitations to claim 1 — the intended parents. This defect is relevant to any construction or validity dispute and was flagged in the earlier litigation section; I found no proceeding addressing it.

F. Limitations and the authoritative next steps

• Unverified: the front-page "References Cited" list (U.S. patent documents, foreign documents, and any non-patent literature) and the examiner's cited references. I did not retrieve these and will not guess at them.
• Best single source to finish this: the EPO search reports for EP 0524348 / EP 0525286 (and the corresponding CH/JP prosecution). EPO search reports assign X / Y / A categories per claim, which is exactly the § 102 / § 103 mapping you are asking for and is more informative than a bare U.S. reference list.
• Controlling U.S. source: the file wrapper of application 07/913,592 in USPTO Patent Center — the PTO-1449 (applicant-cited references), the examiner's Form 892 (examiner-cited references), and the Notice of Allowance / Reasons for Allowance. That is the definitive "each patent citation for 5377002" answer.
• Cross-check: Espacenet's "Cited documents" and "Citing documents" tabs for US5377002, and Justia's "Patent Citations" / "Referenced By" sections.

I'd treat Sections A–D above as verified-and-bounded, and explicitly treat the front-page reference list as an open item rather than assuming it is empty.

I'll ground this analysis by researching the prior-art record for US 5,377,002, then build the § 103 analysis on top of the previously generated summary, litigation, and claim-family sections.

The Delaware opinion surfaced a rich prior-art discussion. Let me pull those references and the prosecution/appeal record.

Obviousness Analysis — US 5,377,002 under 35 U.S.C. § 103

Analyst: Senior US patent analyst · Date: 2026-04-26
Patent: US 5,377,002 A, "Apparatus for surface inspections" (Malin et al.), app. 07/913,592, filed 1992‑07‑14, priority CH 2154/91 A of 1991‑07‑20, granted 1994‑12‑27, 28 claims, expired 2012.

This section builds on — and does not repeat — the prior Patent Summary, Litigation Summary, and claim‑family discussion. Two cross‑checks against those sections: (i) the full text confirms claim 26 is truncated mid‑sentence in the Google Patents rendering and that claim 27 is not retrievable, consistent with the earlier flag; (ii) the claim‑dependency defect flagged earlier is confirmed verbatim — claim 2 reads "An apparatus in accordance with claim 31" and claim 13 reads "An apparatus in accordance with claim 31," although no claim 31 exists. No contradictions found.

I. Governing framework and the person of ordinary skill

Critical date. Everything turns on the July 20, 1991 Swiss priority date. Art must qualify under § 102(a), (b), (e), or (g) as of that date (or the 1991‑07‑14 U.S. filing for art published between the two). Post‑1991 art can inform the level of ordinary skill and the motivation analysis, but cannot itself be a § 103 reference unless it has an earlier effective U.S. filing.

The PHOSITA for this art is, in my assessment: a person with a BS in physics/optics or EE and ~2–5 years designing laser‑scanning optical surface‑inspection instruments for semiconductor wafers or magnetic media — i.e., someone familiar with dark‑field microscopy, confocal (spatial‑filter) detection, scanning‑laser‑microscope optics, spindle/linear‑stage spiral scanners, and shot‑noise‑limited analog front ends. This is not a hypothetical: the patent's own BACKGROUND describes this person's knowledge base (diffuse‑light collection at 360°, photon‑noise limits on bandwidth, the tradeoff between mechanical scan speed and resolution).

§ 103 standard applied. Graham v. John Deere factors; KSR Int'l v. Teleflex, 550 U.S. 398 (2007) for the rationales (known technique/known result; predictable variation; design incentive; "obvious to try"). The claims here are apparatus claims with a heavy functional flavor, which widens the prior‑art field.

II. The prior art of record

II.A — From the patent's own PRIOR ART section (admitted art)

The only reference cited in the patent's own PRIOR ART discussion is:

Source: https://patents.google.com/patent/US5377002/en (PRIOR ART section).

Two admissions in that section are decisive for § 103:

1. The patent admits that this baseline apparatus already detects "not only spherical particles, but also linear and planiform defects of an order of magnitude down to about 1 μm." That is a direct admission that the capability recited in claim 1's preamble ("high‑sensitivity measurements of dot‑shaped, linear, and planiform surface defects") and the functional recitation in claim 1's analysis means ("breaking down the output signal … into measured values produced by dot‑shaped, linear, and planiform defects") are inherent in the admitted art at some level of performance. The claim cannot be saved by the mere naming of the three defect classes.
2. The patent admits the problem and its classical solution ("higher local resolution at the same time as greater measuring sensitivity and speed"), which supplies the motivation element of the § 103 analysis from the patent's own mouth.

II.B — Structural features of US 4,314,763 confirmed from the reference text

Independently retrieved claim text for US 4,314,763 (the '763 "Defect detection system") confirms that the reference is broader than the patent's summary concedes:

• Claim 7: scan means scanning the surface "in a predetermined scan format," a cathode‑ray tube display, and signal translating means coupled to the scan means, the photo detector, and the CRT "for displaying on said cathode‑ray tube the spatial distribution of the locations of detected microscopic defects."
• Claim 8: the scanned area is circular and the "predetermined scan format is a spiral scan format," with processing means to effect a spiral scan of the CRT beam in correspondence with the spiral scan of the surface.
• Claim 11: a common‑axis arrangement of spot, focusing means and detector at the image plane of the focusing means (i.e., a spatial‑filter/confocal‑like geometry).
• Claim 12 (important): "polar to rectangular coordinate conversion means for moving a point on said object with respect to said beam in a spiral path and for generating said scan signals as x and y coordinate signals."
• Claim 13: rotation means + linear displacement means + means generating an angular position signal + a displacement signal + "circuit means for processing said displacement and first and second scan signals to produce third and fourth output signals representing the x and y coordinate positions of said beam on said path."

Source: https://SumoBrain.com/patents/us/Defect-detection-system/[4314763](/patent/4314763).html

This is the single most damaging piece of art for claim 1's final element and for claim 13. Claim 13 of the patent‑in‑suit recites, essentially, (a) a rotation‑pulse emitter (angular data) + a translation‑pulse emitter (radial data) ⇒ polar coordinates, (b) a look‑up table of Cartesian coordinate addresses indexed by pulses, and (c) an address register (pointer) incremented on each rotation pulse. US 4,314,763 claim 12/13 already recites (a) and the polar‑to‑rectangular conversion, and claim 7/8 already recite the CRT spatial‑distribution display. What remains is the specific table‑plus‑pointer implementation — see § V.E below.

II.C — Art surfaced by search (verify dates before relying on it as § 102/103 art)

The critical evidence is not the mere existence of these documents but what the USPTO itself did with them: in the Delaware record at No. 00‑892, the examiner rejected claims as

"unpatentable over Jann et al (US 5,189,481) in view of Yoshii et al (US 5,461,474)" and, in the alternative, "unpatentable over Yoshii et al (US 5,461,474) in view of Steigmeier et al (US 4,314,763)."

and the opinion records the examiner's express rationale:

"It would have been obvious to use in the apparatus of Jann et al a small 'sub‑scan' as taught by Yoshii et al to improve the detection of small particles in the manner taught by Yoshii et al."

and, elsewhere, that "Steigmeier describes a typical spiral scanning system that provides complete coverage of the wafer surface," but "does not teach any method of improved spatial resolution or increased throughput, and fails to suggest a method for introducing a sub‑scan to a system providing a spiral surface scan of a workpiece."

Source: https://www.ded.uscourts.gov/sites/ded/files/opinions/00-892_1.pdf

⚠️ Scope caveat: I did not fully verify which application's prosecution this opinion reviews. I am citing it as evidence of what the examining corps treated as an obvious combination in this exact art (spiral scanner + sub‑scan), not as a holding about 5,377,002. This is powerful because claim 1 of 5,377,002 is, functionally, "spiral scanner with a switchable spot that enables a coarse full‑surface scan and a fine partial scan" — the very combination the examiner called obvious.

III. Claim 1 — element‑by‑element mapping and the obviousness case

Claim 1 (independent). Mapping, with the combination identified.

The three genuine gaps and why each is obvious

Gap 1 — Switchable lens system giving two different spot geometries.
The patent's own discussion frames the objective as reconciling "higher local resolution at the same time as greater measuring sensitivity and speed" and then concedes that the small‑spot mode is used "similar to a scanning laser‑beam microscope" (an admitted, well‑known class of instrument). A POSITA facing the admitted problem has a small, closed set of predictable options: (a) change the objective, (b) change an intermediate‑image relay lens, (c) insert a beam‑shaping element. The patent picks (b) and further limits it to the two most elementary beam shapes in optics: cylindrical lens → line/cigar focus and spherical lens → point focus. Choosing between a cylindrical and a spherical singlet is not an inventive act; it is the first thing an optical engineer reaches for (see dependent claims 2, 4–6). KSR's "known technique, known result" rationale applies squarely.

Gap 2 — Feed offset changed in accordance with the selected image.
This is not an independent inventive concept; it is a forced consequence of Gap 1. If the illuminated spot's radial dimension changes (a cigar with its major axis radial, per claim 3, versus a dot), the spiral pitch required to maintain gapless coverage changes proportionally. Any competent scanner designer must change the translation rate with the spot size, or the scan will either miss area (too coarse) or waste throughput (too fine). Where a claimed parameter adjustment is the necessary result of a design change made for an admitted purpose, the adjustment is obvious as a matter of logic — and the patent's own text gives the reason ("permits a larger feed offset from each revolution to the next").

Gap 3 — Vignetting optics selecting an asymmetry.
The specification supplies both the means and the motivation in one breath: a rod‑shaped vignetting device lets dual measurement "with and without this vignetting device … distinguish between oriented and random haze structures," and "for the identification of one‑dimensional defect structures such as polishing defects" (FIG. 4a/4b discussion). Introducing an aperture stop or an occluding rod into an imaging path is the archetypal predictable variation for imposing asymmetry; the reference to "other forms of vignetting device" confirms the designer treats the shape as a mere design choice. KSR "predictable variation."

IV. Claim 26 — the channel‑architecture claim

Claim 26 is the same physical combination plus the analog front‑end: signal input 70 → a haze channel (haze filter 71 = adjustable low‑pass 75; peak suppressor 72 = diode 721) and a particle channel (subtraction circuit 73 = amplifier 726; digital peak detector 74), with the haze filter output 77 → A/D converter 78, and nominal‑bandwidth input 76a controlling switch systems 725a/b.

Obviousness case for claim 26 (strongest of the set):

1. The problem is admitted and elementary. The specification states: "The signal at the amplifier output is the sum of the signals produced by haze, light‑point defects, and line defects. Further processing has to separate this combined signal into its components." Separating a slowly varying background from short, high‑amplitude pulses is the canonical frequency‑domain split: low‑pass for the background, subtract‑and‑peak‑detect for the pulses. That is a textbook signal‑processing solution with a predictable result.
2. Every building block is a known circuit. Low‑pass RC filters with switched capacitor banks (725a/b + 722 + 724), a diode as a clamps/peak suppressor, a difference amplifier (726) to remove baseline, and a peak detector are all standard 1991 hardware. Substituting a digital peak detector for an analog one to obtain "high linearity, low noise, speed, and extremely short reset times" is the classic known‑technique/known‑result substitution.
3. The bandwidth‑tracking feature is a predictable correction. Adjusting the filter time constant with radius/rotational speed (§ "nominal‑bandwidth input 76a") merely compensates for the known fact that a fixed angular velocity produces a surface velocity proportional to radius — which the patent itself states ("the substrate's outer edge 11 moves fastest and its centre does not move at all"). Compensating a filter for a known, monotonic velocity profile is routine engineering, not invention.
4. Multi‑channel detection was conventional in dark‑field wafer scanners of the era (cf. Jann's multi‑detector scanning system, and the later multi‑channel bright/dark‑field systems in US 6,633,375 / US 6,166,393). Motivation to split the return into a slow‑background channel and a fast‑event channel is furnished both by the art and by the admitted problem.

Net: claim 26 reads as an aggregation of known circuits performing their known functions on the signal that US 4,314,763 (as admitted) already produces, for the admitted purpose of separating the three admitted defect classes. Under KSR, an aggregation of known elements with no unexpected result is obvious, particularly where, as here, the patent asserts no criticality or numerical range for the filter constants.

V. Dependent claims — the individual obviousness rationales

A. Claims 2–3 (cylindrical + spherical lens; same focal length; cigar radial / dot tangential). Cylindrical lenses form line foci; spherical lenses form point foci — pure optics. Orienting the cigar's major axis radially (claim 3) maximizes the arc length covered per revolution at constant angular velocity, directly serving the admitted goal of a larger feed offset; the minor (tangential) axis preserves circumferential resolution. Predictable optimization, no unexpected result. (Same‑focal‑length selection keeps the intermediate image in one plane — a mechanical convenience, not a technical advance.)

B. Claims 4, 5, 6 (L‑shaped pivot carrier; slide with stops; rotatable disk with stop defined by angle). Three interchangeable lens‑mount designs. These are the optomechanical equivalents of a filter wheel / objective turret, which is a decades‑old, ubiquitous solution to "put one of two elements in a beam." The patent itself recites the drives as "electric motors, electric rotary or lifting magnets, or compressed‑air drives" and the stops as "end switches or inductive approximation switches" — an enumeration of off‑the‑shelf parts. Obvious as a matter of routine design choice.

C. Claims 7–11 (three dark‑field stop variants). Coated reflective center on a transmissive plate; a prism relying on total internal reflection; a silvered oblique cylindrical body. Each is a standard way to fold a beam while occluding its center; the patent's own rationale for the cylindrical/prism forms ("the reflecting surface … can be brought closer to the objective … the smaller is the vignetted area … the better") is a straightforward geometric optimization. TIR folding prisms and silvered mirrors are the most basic tools in optical bench work. Obvious.

D. Claim 12 (rod‑shaped vignetting optic). Explicitly described in the specification as the embodiment for identifying one‑dimensional defects such as polishing scratches. The claim recites only the shape; see § III Gap 3.

E. Claim 13 (rotation/translation pulse emitters ⇒ polar coords; look‑up table of Cartesian addresses; pointer/address register auto‑incremented on rotation pulses). This is the claim with the densest prior‑art overlap:

• US 4,314,763 claim 12 recites "polar to rectangular coordinate conversion means … and for generating said scan signals as x and y coordinate signals."
• US 4,314,763 claim 13 recites rotation means + linear displacement means + angular‑position signal + displacement signal + "circuit means for processing … to produce … the x and y coordinate positions of said beam."
• US 4,314,763 claims 7–8 recite the spiral‑scan CRT display of defect spatial distribution.

What remains is the implementation: a precomputed table plus an auto‑incremented pointer, used instead of per‑sample trigonometry. The specification states the reason outright ("coordinate transformation by trigonometric computation of each positional value would take too long, a faster method is necessary"). Replacing a repeated, expensive computation with a precomputed look‑up table is the paradigm of a predictable, routine engineering optimization (memory‑for‑compute trade), and KSR instructs that such improvements, where the result is predictable, are within ordinary skill. The "pointer incremented on each rotation pulse" is simply the read‑address generator for that table.

F. Claims 14, 15, 16, 17, 18 (scan center→edge; index‑pulse generator synchronized with rotation pulses; rectangular‑pulse oscillator; discarding out‑of‑ring‑segment measurements; more index pulses than rotation pulses). All are scan‑sequencing and gating choices that follow from the ring‑segment scanning geometry. Claim 16 specifies a "electronic rectangular‑pulse oscillator," i.e., a clock — a component with no plausible claim to novelty. Claim 18 (more index pulses than rotation pulses ⇒ smaller feed offset for the second scan) restates the necessary relationship of Gap 2.

G. Claims 19–24 (color‑coded 2‑D graphs; separate haze and LPD graphs; inverse LPD color scale; selectable dynamic‑range window; numeric color legend; histogram). These are presentation limitations. Rendering measured amplitudes as color‑coded maps, allowing an operator to select the displayed dynamic‑range window, inverting a color scale to flag the complementary quantity, and supplementing an image with a size‑vs‑count histogram are conventional data‑visualization choices with predictable results. The specification itself presents the invert‑to‑white‑background LPD scheme as an ergonomic decision ("for user orientation"), and the histogram as a convenience. Under KSR and In re GPAC‑type reasoning on presentation features, no non‑obviousness is likely to be found, particularly absent any recited numerical or structural criticality. (Note also the admitted autoscale analogue: the spec frames autoscale/prescale expressly as "user convenience" and "level of automation," not as a technical advance.)

H. Claim 25 (mass storage connected to the computer unit). Reciting a disk/store to retain measurement data duplicates the admitted "mass‑storage system 23." Obvious.

I. Claims 27, 28. ⚠️ Unverified. Claim 27 could not be retrieved with confidence (the Google Patents rendering truncates claim 26 mid‑sentence and the retrieval of the second set was incomplete). No obviousness opinion can be responsibly given for claim 27. Claim 28, per the earlier section, depends from claim 26; on the retrieved text it inherits the claim‑26 analysis.

VI. The motivation‑to‑combine case (why a PHOSITA would have combined these references)

The § 103 "motivation" element is unusually easy here because the motivation is supplied by the patent itself and by the examining corps:

1. Articulated problem, explicit tradeoff. The BACKGROUND states the apparatus "must at the same time be extremely sensitive and provide a high measuring speed," that "limiting factors are the finite mechanical scanning speed … and the greater bandwidth necessary in the measuring system as the speed increases," and that "photon noise mitigates against greater bandwidth." A POSITA reading that paragraph has been handed the design brief: resolve the resolution/speed/noise tradeoff.
2. Same‑field, same‑problem references. US 4,314,763 (admitted art) is the same field (spiral‑scan dark‑field wafer/substrate inspection) targeting the same defects. Art is combinable where it comes from the same field and addresses the same problem. Yoshii's sub‑scan and Jann's multi‑channel spiral scanner are directed at the identical problem (small‑particle detection at throughput).
3. The USPTO's own stated rationale. The examiner in the No. 00‑892 record: "It would have been obvious to use … a small 'sub‑scan' … to improve the detection of small particles," and "Steigmeier … provides complete coverage of the wafer surface" but "fails to suggest a method for introducing a sub‑scan." That is precisely the architecture claim 1 recites (coarse whole‑surface spiral + fine selective scan via a switchable spot).
4. Predictable‑result / finite‑options. For Gaps 1–3 (§ III), the universe of viable optical means is small and their results are predictable; for claim 26, the means are textbook circuits. KSR's "obvious to try" and "predictable variation" rationales apply.
5. Design incentive / market pressure. Wafer‑inspection throughput vs. sensitivity was the competitive axis of the era (Tencor's Surfscan line; KLA's scanners). Neither the spec nor the record identifies any secondary consideration, so there is no nexus evidence to weigh against the strong prima facie case.

VII. Rebuttal considerations and honest weaknesses in the § 103 case

I want to be explicit about where the obviousness case is weak or where I lack verification, rather than overstate it.

1. Date verification is the single biggest exposure. US 5,189,481 (Jann) and US 5,461,474 (Yoshii) are the references that best supply Gaps 1–2. I could not verify their effective U.S. filing dates against the 1991‑07‑20 priority. If neither has pre‑1991‑07‑20 U.S. filing/priority, they are not § 102/103 references to this patent; they would only support the "state of the art / level of skill" narrative. A rigorous § 103 opinion should substitute a pre‑1991 reference disclosing variable spot size / dual‑resolution scanning (e.g., laser‑scanning‑microscope zoom optics, or the Olympus/Zeiss confocal literature). I flag this because the strongest combination I can state is partly built on a date I have not confirmed.
2. The dark‑field stop assembly is not cleanly disclosed by US 4,314,763. The admitted reference uses two prisms to route the beam, not a center‑obscuring stop. My combination relies on treating dark‑field microscopy's center stop as common knowledge (it is — dark‑field condensers with central obscuration are a century old), plus Jann‑type dark‑field scanners. This is defensible as common knowledge, but a rigorous office action would want a concrete secondary reference for the rotationally symmetric stop shape (which the patent claims separately in claims 7–11).
3. The confocal diaphragm is only "suggestive" in US 4,314,763. Its claim 11 (detector at the image plane on the common axis) and claim 9 (spatial filtering) are close, but I would not call them an express confocal aperture. Here the combination needs confocal‑microscopy common knowledge, which is abundant and pre‑1991 (Minsky; confocal scanning laser microscopes). Still, I am inferring rather than quoting.
4. "Teaching away" is unlikely to save the claims, but two arguments deserve a hearing. (a) A patentee could argue that astigmatic (cigar) spots are normally considered a defect in imaging and that a POSITA would not have deliberately introduced astigmatism into a precision inspection path. The answer is that the patent itself exploits the cigar purely as an illumination geometry for coverage, not as an imaging correction — and the spec's own rationale ("covers a larger area … permits a larger feed offset") defeats any teaching‑away argument. (b) The patentee could argue the claimed combination (switchable intermediate image plus image‑dependent feed offset plus selectable vignetting asymmetry plus per‑measurement position registration) is more than a mere aggregation. This is the genuinely arguable point, and it is a combination‑as‑a‑whole argument under § 103, not an element‑by‑element one. My view: it is weak, because each element performs exactly its expected function and the patent identifies no synergistic or unexpected result.
5. Claim‑dependency defect is a collateral § 112 issue, not a § 103 one. Claims 2 and 13 recite "claim 31," which does not exist in the 28‑claim set. This does not defeat the obviousness analysis (I analyzed the claims as if intended to depend from the sole independent claim they read on, claim 1), but it would make those claims vulnerable to an indefiniteness challenge, and it complicates any validity chart built on the printed text. I am not auto‑correcting the reference to "claim 1."
6. No prior §§ 103 rejection of 5,377,002 itself is evidenced. I did not retrieve the patent's own file wrapper; the Delaware No. 00‑892 record concerns a different application's prosecution. The presence of US 4,314,763 as admitted art and the sibling US 5,377,001 / EP 0524348 / EP 0525286 family suggests the European search report for the family likely cited additional art, but I did not retrieve EP 0524348/0525286 search reports, and I will not speculate about their contents.

VIII. Bottom line

• Claim 1 is, in my assessment, obvious over US 4,314,763 (Steigmeier) in view of sub‑scan/multi‑spot scanning art and the conventional dark‑field/confocal microscopy knowledge — the examiner's own reasoning in the No. 00‑892 record supplies the motivation ("obvious to use in the apparatus of Jann a small 'sub‑scan' … to improve the detection of small particles"), and the patent's PRIOR ART section concedes both the base architecture and the three‑defect‑class capability. The claim's surviving features (switchable cylindrical/spherical intermediate image; image‑dependent feed offset; selectable vignetting asymmetry) are predictable variations and design choices with no asserted unexpected result.
• Claim 26 is, in my assessment, the most clearly obvious claim: it aggregates a low‑pass haze channel, a subtraction circuit, and a (digital) peak detector to perform a frequency‑domain separation that the specification itself identifies as the necessary processing step. Known circuits, known functions, predictable result.
• Claims 2–12 and 14–25 fall to the same references plus routine optomechanical, signal‑processing, and data‑presentation knowledge, per § V.
• Claim 13 has the densest single‑reference overlap: US 4,314,763 claims 7, 8, 12, and 13 already recite the spiral scan signals, the polar‑to‑rectangular conversion, the x/y coordinate signals, and the CRT spatial‑distribution display; only the precomputed table/pointer implementation remains, and that is a routine memory‑for‑compute optimization the specification itself justifies on speed grounds.
• Claim 27 is unverified and no obviousness opinion is offered for it.

Confidence: Moderate‑to‑high on the § 103 conclusions for claims 1 and 26 and for the dependent‑claim rationales; low‑to‑moderate on the admissibility of the Jann/Yoshii references owing to unresolved date questions, and low on claim 27. Anyone relying on this should (i) obtain the 5,377,002 file wrapper (to see the art the examiner actually cited against the family), (ii) obtain the EP 0524348 / EP 0525286 search reports, and (iii) verify the filing/priority dates of US 5,189,481 and US 5,461,474 against 1991‑07‑20 before treating them as § 102/103 art rather than as evidence of the level of ordinary skill.