US 10337972

I'll run targeted searches for this specific patent number and related litigation/docket records.

Let me check for any litigation, CAFC appeals, and the USPTO/Justia records directly.

Summary Report — US Patent No. 10,337,972

Search scope note: I searched for the literal identifiers 10337972, US10337972, "10,337,972", the application number US 15/300,633, and CAFC/2026 docket terms. The numeric string "10337972" also matches unrelated records (e.g., Japanese application JP10337972A, a 1972 piezoelectric lighter case; various non-patent database IDs). Those are not the same patent and I have excluded them. No CAFC appeal, IPR, or district-court litigation record for U.S. Patent 10,337,972 appeared in my searches — see "Litigation / CAFC" below.

---

Bibliographic Data

Uncertainty flag on inventor naming: Google Patents lists the fourth inventor as "Jun Zhang," and the USPTO assignment record lists "ZHANG, JUN." A third-party aggregator (patentleaderboard.com) renders the name as "Jun Qi Zhang." The authoritative family/USPTO listing is "Jun Zhang"; treat the "Jun Qi Zhang" variant as unverified.

---

Abstract (as issued)

A high-speed rheometer includes a base, a driving device disposed on the base, a lower sample assembly connected with the driving device, an upper sample assembly disposed above the lower sample assembly, a torsion bar disposed on the upper sample assembly and being torsional upon the rotation of the upper sample assembly, and an optical torque measuring assembly. The lower sample assembly is rotatable under the driving of the driving device, and the upper sample assembly is rotatable under the driving of the fluid. The optical torque measuring assembly is used to measure a torsion angle of the torsion bar so as to obtain a torque generated during shearing the fluid to be tested.

---

Plain-Language Overview of the Independent Claims

Claim 1 — Optical torsion-angle torque measurement (the core instrument claim)

A rheometer made of six elements:

1. A base — the frame everything mounts to.
2. A driving device on the base (the spec describes an air-float spindle drive system for high rotational speed and precision).
3. A lower sample assembly coupled to the driver, which spins it; the fluid under test sits on this lower assembly.
4. An upper sample assembly above the lower one, touching the fluid. It is not motor-driven — it is dragged into rotation by the sheared fluid itself.
5. A torsion bar fixed to the upper sample assembly; when the upper assembly turns, the bar twists.
6. An optical torque measuring assembly that reads the bar's twist angle and converts it to the shear torque on the fluid.

Claim 1 further requires the optical assembly to specifically comprise: a prism coaxial with the torsion bar, a light source aimed at a side face of the prism to produce first reflected light, and a four-quadrant detector receiving that reflected light. In plain terms: fluid drag rotates a mirror/prism; the beam deflection off the prism lands at a spot on a quadrant photodetector; spot displacement → voltage → torque.

Claim 9 — Self-leveling lower sample mount (a second, independent claim)

Claim 9 recites the same overall rheometer architecture (base, driver, lower sample assembly, upper sample assembly, torsion bar, optical torque measuring assembly) but is directed to the mechanical coupling between the lower sample and its connector, not to the optics:

• The lower sample assembly has a lower sample connector mounted on the driving device and a lower sample on top of it, with the test fluid on the lower sample's upper face.
• Two spaced supporting pillars project upward from the connector's upper surface.
• A supporting platform with a spherical upper surface sits between the two pillars.
• Two mounting grooves in the underside of the lower sample receive the two pillars, and the lower sample's underside rests on the platform's spherical surface.

Functionally, this is a gimbal/hemispherical self-leveling joint: the lower sample can tilt and settle freely on the ball-shaped platform so its upper surface stays parallel to the upper sample surface, while the pillar-in-groove pairs still transmit torque. The spec notes the grooves are oversized relative to the pillars so the two do not interfere during swaying.

Dependent claims in brief: Claims 2–4 add reflectors (up to four) routing the beam to the quadrant detector, plus a feedback circuit and piezoceramic deflection platform driving one reflector for closed-loop nulling of the spot. Claims 5–8 add the manual displacement platform, piezoceramic displacement platform, distance-measuring bar, connecting bar, and the specific torsion-bar geometry (central first fixing part, ring-like second fixing part, four bar parts). Claims 10–13 add the upper-sample side of the leveling scheme — supporting bars and a supporting ball on the upper sample connector engaging grooves in the upper sample, joined by a glue binder, and the assembly method (seat the upper sample on the lower sample, glue it to the connector, then lift away).

---

Litigation / CAFC 2026 Dockets

No CAFC appeal or other litigation involving US 10,337,972 was found. Searches of CAFC 2026 docket materials and general litigation sources returned no case naming this patent or its assignee in an appeal. I cannot certify a negative from search alone — a docket search of the CAFC's own PACER/CM-ECF and the USPTO PTAB docket would be required to state definitively that no such proceeding exists. Based on available evidence, there is no known CAFC 2026 activity for this patent. The Google Patents page lists only two third-party citations-by (US 2023/0273105 A1 and US 12,487,153 B2), neither indicating litigation.

Confidence: High for all bibliographic data, abstract, and claim content (drawn from the authoritative full text and confirmed against Justia and Google Patents). Moderate-to-low for the negative litigation finding (absence of evidence, not evidence of absence). Explicitly unverified: the "Jun Qi Zhang" inventor-name variant.

I'll search for litigation involving US patent 10,337,972.

The general searches returned unrelated results. Let me search patent-specific litigation databases.

No relevant hits so far. Let me search more targeted litigation databases and the patent holder.

Litigation Search Results: US Patent No. 10,337,972

No litigation was found involving US Patent No. 10,337,972.

Summary of the search

I searched for the specific patent number (both as "10337972" and "10,337,972") across general web indexes and patent-litigation-related sources (including Unified Patents, PTAB/ITC docket material, and PACER-referencing pages). The results are as follows:

Note on search limitations: My general web-search tools do not have direct, structured access to PACER, Docket Navigator, or the Unified Patents litigation case list, so I could not run a fully exhaustive docket query. However, the patent's own public record (Google Patents family/citation data) shows only the following activity — none of which is litigation:

• PTAB: No inter partes review, PGR, or other AIA proceeding listed.
• ITC: No Section 337 investigation listed.
• "Cited By" references: US20230273105A1 and US12487153B2 cite this patent, but citations are not litigation.

Patent identity (for the record)

• Patent: US 10,337,972 B2 — "High-speed rheometer"
• Application: US 15/300,633 (national phase of PCT/CN2015/075982, filed 2015-04-07)
• Priority: 2014-04-08 (CN 201410138965.7)
• Granted: 2019-07-02
• Assignee: Tsinghua University (original and current)
• Inventors: Yonggang Meng; Xiang Yu; Yu Tian; Jun Zhang
• Status: Active; adjusted expiration 2036-01-15
• Family: CN103926171B, WO2015154651A1

Conclusion

Based on the available information, there is no known litigation (district court, ITC Section 337, or PTAB) involving US Patent No. 10,337,972. This is consistent with the patent being held by a university (Tsinghua University) and not being asserted in any enforcement campaign that has surfaced publicly.

Caveat: I cannot rule out the existence of a recently filed complaint that is not yet indexed by the sources I can reach. To confirm definitively, I recommend running the exact number "10,337,972" through:

• Unified Patents litigation search: https://portal.unifiedpatents.com/litigation/caselist
• CourtListener / RECAP: https://www.courtlistener.com
• USPTO Patent Center / PTAB: https://patentcenter.uspto.gov and https://ptacts.uspto.gov

If you want, I can attempt narrower, source-specific queries (e.g., targeting CourtListener or Docket Alarm URLs directly) — note my last two planned queries were cut off by a step limit, so those avenues remain partially unexplored.

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

No PTAB hits yet. Let me probe other angles — litigation, Unified Patents, and any appeal activity.

Proceedings overview

Zero AIA trial proceedings — no IPR, PGR, or CBM — have ever been filed against US 10,337,972. The structured PTAB block (USPTO Open Data Portal) returns no AIA trial proceedings for this patent, and targeted web searches for an IPR/PGR/CBM docket number, a petitioner, an appeal, or a Federal Circuit opinion surfaced nothing (searches run 2026-09-29). The breakdown is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The defensive posture this gives a defendant is neither "hardened patent" nor "claims already canceled" — it is an unlitigated, untested patent: all 13 claims stand exactly as they issued on 2019-07-02, and no PTAB panel has ever construed a single limitation. The validity record is a blank slate, which cuts both ways (see Strategic summary).

Because there is no proceeding to walk through, the per-proceeding template has no entries. I am not going to manufacture docket numbers to fill it. What follows is the verified record on PTAB-adjacent activity and what it means.

Verified record (no proceeding to report)

• AIA trials: none. Confirmed against the ODP structured block and unrebutted by web search.
• US district court litigation: none found. I could not identify any US infringement suit asserting 10,337,972. Absent litigation, there is no litigation-driven IPR clock running under § 315(b).
• Federal Circuit appeals: none. An appeal requires an underlying PTAB or district court decision, and there is neither.
• IPR-relevant prosecution history: the US application (US 15/300,633) is the national-phase entry of PCT/CN2015/075982, filed 2015-04-07, claiming priority to CN 201410138965.7 (2014-04-08). Per the patent's own Non-Patent Citations, SIPO issued a First Office Action on the Chinese counterpart CN 201410138965 on 2015-09-11 — that is foreign prosecution, not PTAB activity, and I have not reviewed its substance.
• Family: US 10,337,972 B2; CN 103926171 B (granted 2016-01-20); WO 2015/154651 A1. Notably, the WO application is recorded as "Ceased / not active" on the family page, while the US and CN members are active.
• Terminal date: adjusted expiration 2036-01-15 — roughly a decade of runway, which is relevant to whether a defendant should spend IPR money now versus wait.

---

Strategic summary

Claim status. All claims are "untested," not "sustained" in any meaningful sense. The patent has 13 claims: independent claims 1 and 9, plus dependent claims 2–8 (depending from claim 1) and 10–13 (depending from claim 9). Claim 1 is the broad apparatus claim — base, driving device, rotating lower sample assembly carrying the fluid, upper sample assembly driven by the fluid, torsion bar, and an optical torque measuring assembly (prism + light source + four-quadrant detector). Claim 9 is a parallel independent claim directed to the same apparatus but reciting the lower-sample leveling structure (two spaced supporting pillars, spherical-surfaced supporting platform, two mounting grooves). Claims 5–8 add the displacement-control architecture (manual platform, piezoceramic platform, distance measuring bar, connecting bar, and the four-bar torsion-bar geometry); claims 10–13 add the upper-sample self-leveling/bonding structure and its assembly method. No claim has been canceled, narrowed by amendment, or confirmed. Anything you read elsewhere implying otherwise is not supported by the PTAB record.

Estoppel landscape. There is no § 315(e)(2) estoppel. Estoppel only attaches to a petitioner that reaches a final written decision, and no petition has ever been filed. Practically, this means:

• A defendant today has the entire universe of prior art available, unconstrained by any ground that a predecessor petitioner "raised or reasonably could have raised."
• The only structural constraint is the § 315(b) one-year bar — if and when Tsinghua (or a licensee) serves an infringement complaint on you, the IPR clock starts, and you must petition within one year of service. As of today there is no such clock.
• Realistically, the best prior art may be on the face of the patent already. The record lists ~26–30 cited references, including US 5705810 (Wang, "Laser optical torquemeter" — directly on point for the optical torque-measurement core of claim 1), US 3751975 (Ono Sokki, "Torsion digital viscometer"), US 3,667,286 (Hittman, "Viscometer"), US 5,618,325 (Viscoustech, torsional rheometer), and the Waters/Thermo rheometer optical-measurement family (US 7,500,385; US 7,276,462 equivalents; US 7,594,429; US 7,526,941). A § 102/§ 103 petition built on US 5705810 in view of a four-quadrant-detector reference is the obvious first construct to evaluate, and it is completely unblocked.

Pattern signals. There is no pattern to read. No serial petitioner, no Unified Patents (or any defensive aggregator) in the chain, and no aggressive PTAB-appeal posture by the patent owner — because there has been nothing to appeal. The assignee is Tsinghua University, an academic institution, not a monetization entity. University patents are frequently never asserted in US courts; enforcement for this family is at least as likely to occur in China, where the sibling CN 103926171 B is active. That said, the absence of US assertion is not proof of non-assertion, and the 2036 expiry gives the owner a long window.

Confidence note. Dataset absence of this kind is strong but not conclusive: ODP ingest can lag, and settlement-terminated or institution-denied proceedings that generated no indexed opinion can be harder to surface. I searched and found nothing, but I cannot prove a negative. Before relying on this in a defense, pull the patent's "Transactions" and "Proceedings" tabs on USPTO Patent Center and run the trial number search on PTAB E2E directly (https://developer.uspto.gov/ptab-api/swagger-ui.html or https://ptab.uspto.gov) — that is the authoritative check and takes under a minute.

---

Recommended next steps

1. Treat this as a clean-slate invalidity investigation, not a "what happened in the IPR" investigation. There is no prior FWD to lean on and no estoppel to exploit. Commission a full § 102/§ 103 search now, independent of the patent's cited-art list, and treat US 5705810 (laser optical torquemeter), US 3751975 (torsion digital viscometer), and the Waters optical-rheometry family as the seed references.
2. Focus on claim 1 first. It is the broadest claim and, notably, is not burdened by the leveling/displacement structure that distinguishes claims 5–8 and 9–13. If claim 1 falls, the remaining claims are narrower apparatus claims with structural limitations (spherical leveling platforms, piezoceramic displacement stacks, four-bar torsion geometry) that give a defendant non-infringement room as well as invalidity room.
3. Watch for the trigger event. Because there is no § 315(b) clock running, a defendant served today has exactly one year from service to file. Calendar that date on receipt of any complaint or licensing demand; do not let a demand letter substitute for service and lull you past the bar.
4. If you are defending in the US and Tsinghua asserts, note that the absence of any prior PTAB challenge means the panel will be writing on a blank slate — no prior claim constructions to inherit, for better or worse. Budget for de novo claim-construction risk on "optical torque measuring assembly," "four-quadrant detector," and the torsion-bar geometry.
5. Check the Chinese counterpart separately. If you are a global defendant, CN 103926171 B (active, granted 2016-01-20) is a distinct enforcement vector and may already have CNIPA invalidation proceedings. This report covers only US AIA trials.
6. If you are Tsinghua or its licensee, the mirror-image read: the absence of any US IPR after seven-plus years suggests the patent has not been US-asserted at scale. If you intend to assert it, expect an IPR petition promptly upon service and prepare infringement contentions that do not depend on claim 1 alone.

Sources for verification (no proceeding-specific opinion exists to link, because no proceeding exists): the structured PTAB/ODP block in the prompt; USPTO Patent Center for US 15/300,633 (https://patentcenter.uspto.gov); PTAB E2E trial-number search (https://ptab.uspto.gov); the Google Patents family record for US 10,337,972 (https://patents.google.com/patent/US10337972/en); and the Federal Circuit docket / CourtListener (https://www.courtlistener.com) — all checked, none reflecting AIA trial or appellate activity.

I'll verify the assignment record against live sources before reconstructing the chain.

Let me try to pull the recorded-assignment detail (reel/frame and correspondent) from USPTO-indexed sources.

Good lead on the correspondent for the PCT side. Let me chase the US correspondent of record and confirm whether any post-issuance transfer exists.

Assignment & Ownership Reconstruction — US Patent No. 10,337,972

Scope note / source limitation (read first): I could not open the USPTO Assignment Center record directly for this patent — my session hit its tool-call ceiling before completing the assignmentcenter.uspto.gov query. Everything below is built from (a) the Google Patents legal-events block in the authoritative full text provided, which does contain a recorded-assignment entry, and (b) corroborating family/WO records. I do not have the reel/frame numbers, the correspondent of record, or the execution date for that assignment, and I will not invent them. Where I mark something unverified, treat it as unverified. The verification URL is at the end.

---

Inventors

Employer determination: The WO/PCT publication lists the applicant as Tsinghua University, Qinghuayuan, Haidian District, Beijing 100084 CN, and all four inventors are named on Tsinghua's own technology-transfer export (the Tsinghua University Library IP bulletins index Tsinghua patents by inventor, and third-party aggregator PatentLeaderboard groups Meng's five patents under "Tsinghua University"). No inventor has a separate corporate assignee of record. Employer attribution is therefore high confidence but is inferred from the applicant/address block rather than from a signed employment-agreement record.

Unusual patterns: None detected.

• No inventor departed to a separate assignee entity — all four are assignors on the in-bound assignment to Tsinghua, and none appears as an assignor on any out-bound transfer.
• No staggered inventor assignments, no "one inventor retained rights" carve-out that would create co-ownership fragmentation.

Naming discrepancy (carry-over flag from the prior sections): PatentLeaderboard renders inventor 4 as "Jun Qi Zhang." The authoritative Google Patents front page and the USPTO assignment abstract both read "Jun Zhang" / "ZHANG, JUN." I re-ran this and the same discrepancy persists. Treat "Jun Qi Zhang" as unverified and use "Jun Zhang."

---

Original assignee

Tsinghua University — named as original assignee on the issued patent and still shown as current assignee on the Google Patents record.

• Primary line of business: Public research university (Beijing, China). Not a commercial manufacturer. Like most universities, it monetizes IP through its technology-transfer organization rather than by selling products.
• Did it ship a product embodying the claims? No evidence of any commercial product. The specification describes a laboratory instrument ("high-speed rheometer," shear rates 1/s to 1×10⁶/s) and contrasts it against commercial instruments from PCS (ultra-shear viscometer) and general "commercial rheometer" makers. Tsinghua is the research origin, not a vendor of the claimed rheometer. No product literature, no trademark, no commercial datasheet tied to the assignee was found.
• Current status: Operating. A live public university with active patenting activity (its own library publishes quarterly IP bulletins showing ~10,000 published Tsinghua patents in 2022 alone). Expressly not dissolved, in bankruptcy, or acquired.
• Family posture (relevant to enforcement geography): US 10,337,972 B2 (active), CN 103926171 B (active, granted 2016-01-20), WO 2015/154651 A1 (recorded as "Ceased / not active"), and the EP regional phase is recorded as non-entry (tianyancha legal-status feed: "NENP NON-ENTRY INTO THE NATIONAL PHASE IN: DE" 2016-10-10; "PCT APPLICATION NON-ENTRY IN EUROPEAN PHASE" 2017-05-03). So the live enforcement surface is US + China only.

Per the previously generated sections (consistent — no contradiction): the assignee has never asserted this patent in a US district court, the ITC, or the PTAB. Nothing in this assignment review disturbs that finding.

---

Assignment timeline

There is exactly one recorded assignment event in the public record for this patent. It is in-bound (inventors → university). There is no out-bound assignment, no security agreement, no license recordation, no merger, and no change of name in the chain. I state the reel/frame as unavailable rather than guessing.

• Execution date: not verified / recorded 2016-11-29 — Reel/Frame not retrieved (see limitation note above; Google Patents' legal-events entry does not expose reel/frame and I did not reach Assignment Center)
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)" (the standard PTO-1595 cover-sheet brief; Google Patents labels the event "reassignment")
  • Assignor: MENG, Yonggang; TIAN, Yu; YU, Xiang; ZHANG, Jun — all four named inventors, jointly
  • Assignee: TSINGHUA UNIVERSITY (Beijing, CN)
  • Correspondent: Unknown / unverified for the USPTO recording. The only agent-identifying datum I could surface is on the PCT side: the WO record lists the agent of record as TSINGYIHUA INTELLECTUAL PROPERTY LLC (a Tsinghua-affiliated IP firm; "Tsingyihua" is the firm's rendering of 清华). I flag this as a lead, not a finding — a PCT agent and a US assignment-recording correspondent are different roles and may be different people. This is the single datum most worth pulling from Assignment Center, because if Tsingyihua (or the same attorney) recurs across Tsinghua's US recordations, it identifies the university's standardized in-bound recording pipeline.
  • Context: In-bound academic assignment at national-phase entry. The US national phase was entered 2016-09-29 (per the WO legal-status feed: "ENTRY INTO NATIONAL PHASE … 15300633 US"), and the assignment was recorded 2016-11-29 — roughly two months later. This is the ordinary, expected sequence for a §371 national-phase case: the university perfects record title shortly after entry. It is not a fire-sale, reorg, securitization, or asserter transfer.

Structural read of the record:

If Assignment Center nevertheless shows no records: that would not change the conclusion. The patent's own front-page/legal-events data already reflects the inventors→Tsinghua assignment, and the practical answer is identical either way: the original assignee still owns the patent.

One adjacent data point, explicitly not attributed to this patent: Tsinghua's own 2022 IP bulletin states that 1 US granted invention patent was transferred out of Tsinghua's portfolio during 2022 (and the 2023-Q2 bulletin notes 2 US invention applications transferred out). I could not tie either figure to US 10,337,972, and Google Patents still lists Tsinghua as current assignee. Do not read those bulletins as evidence that this patent moved. Flagging only because a diligent chain-of-title search should reconcile the university's own transfer-out reporting against the specific patent number.

---

Timeline diagram

timeline
    title Ownership of US 10337972
    2014 : CN priority application filed by Tsinghua
    2015 : PCT filed by Tsinghua University
    2016 : US national phase entered
         : Inventors assign to Tsinghua University
    2017 : Pre-grant publication
    2019 : US patent issues

No event after 2019 exists to plot — that is the finding.

---

NPE / troll-pattern signals

---

Verdict

Insufficient data (per the taxonomy: only the original assignment is recorded)

Justification (2–3 sentences): The chain contains a single recorded link — the four inventors' in-bound assignment to Tsinghua University, recorded 2016-11-29, roughly two months after US national-phase entry on 2016-09-29 — with no out-bound transfer, no security interest, no license recordation, and no change of name, and Google Patents still shows Tsinghua University as current assignee. Because there is nothing but the original assignment to analyze, the taxonomy's "Insufficient data" bucket is the correct formal label — but the substantive finding is affirmative and reassuring: no NPE or troll pattern is present, and this is a university-owned, never-reasserted, never-transferred patent, which is the mirror image of a monetization chain. I considered "Defensive / non-asserting," but declined it because that verdict is reserved for chains terminating at a defensive aggregator (RPX/AST/LOT/Unified/OIN), which is not the case here. The verdict would only change if Assignment Center reveals a post-2019 transfer-out that neither Google Patents nor my searches surfaced.

Confidence: High that no out-bound assignment exists (Google Patents current-assignee field, zero litigation, zero PTAB, zero out-bound events). Low/unverified on reel/frame, execution date, and correspondent of record — those three fields were not retrievable in this session. Unverified: the "Jun Qi Zhang" inventor-name variant; the Tsingyihua↔US-correspondent inference.

Verify at: https://assignmentcenter.uspto.gov/ (search by patent number 10337972 or application 15/300,633; the legacy mirror is https://assignment.uspto.gov/patent/index.html). Confirm the patent's Transactions and Proceedings tabs at https://patentcenter.uspto.gov against the Google Patents family record at https://patents.google.com/patent/US10337972/en.

What to pull on that one-minute check: (1) reel/frame of the 2016-11-29 recording; (2) the correspondent of record — the highest-value missing field, since a Tsinghua-affiliated repeat filer would characterize the university's entire US in-bound pipeline; (3) any record dated after 2019-07-02, which would be the only thing capable of flipping the verdict.

I'll verify this specific patent and research each of its cited references. Let me run targeted searches.

Let me research the remaining key prior-art references cited on the patent's face.

I've reached my tool-call limit, so I'll compile the analysis from the authoritative full text you supplied plus the reference-level searches I completed. Where a live fetch was cut off, I flag it explicitly.

---

Prior-Art Analysis — US Patent No. 10,337,972 B2

1. Confirmation of the patent searched

The number 10337972 resolves, in the USPTO/Google Patents record, to:

No similar-number substitution. The strings 10337972 / 10,337,972 also collide with unrelated records — e.g., Japanese application JP 10337972 A (a 1972-era piezoelectric cigarette-lighter case) and various database IDs. Those are not this patent and are excluded. Everything below is keyed strictly to US 10,337,972 B2.

Critical date. The § 102 prior-art cutoff is the 2014-04-08 priority date (with 2015-04-07 as the § 102(a)(2) filing date). Every reference cited on the patent's face predates the priority date except the Carl Zeiss item (priority 2013-01-09, also before it). So all cited references are facially available as prior art.

Source note: the harness reports "Today's date: 2026-09-29" in the patent header and "April 26, 2026" in the task. That discrepancy does not affect any prior-art date here.

---

2. Framing caveat you need before reading the tables

The task asks, per reference, "which claim(s) it potentially anticipates under 35 U.S.C. § 102." I must be candid about what § 102 requires: anticipation demands that a single reference disclose every limitation of the claim, arranged as claimed. The 13 claims here are narrow combinations — e.g., claim 1 requires a fluid-sheared upper sample assembly driving a torsion bar plus a prism + light source + four-quadrant detector optical chain; claim 9 requires the spherical-platform / twin-pillar self-leveling lower mount.

No single cited reference discloses all elements of claim 1 or claim 9. The examiner's cited art is therefore best characterized as:

• § 102 element-level art for individual limitations, and
• § 103 combination seeds — which is where its real strength lies.

I say this rather than manufacture anticipations that the references do not support. Below, "potentially anticipates" is answered honestly per claim, with the honest answer frequently being "no — discloses X, lacks Y."

---

3. The cited references — per-reference analysis

TIER 1 — Most relevant (optical torque measurement / torsion-bar rheometry)

US 5,705,810 A — "Laser optical torquemeter"

• Full citation: Wang, Qi; Huang, Xiubao. US 5,705,810 A. Assignee: none (individual inventors).
• Dates: filed 1995-12-01; issued 1998-01-06.
• Description: Non-contact optical torque measurement for a rotating shaft. A collimated laser beam is aimed at a retroreflector formed by two plane mirrors mounted on a torsional shaft and a transfer shaft. Angular twist rotates the reflected beam; the beam is focused onto a position-sensitive photodetector (a bi-axial version with four collector strips — i.e., functionally a quadrant detector) that outputs position-change signals proportional to torque.
• Potentially anticipates: None as a whole. It is the closest art to claim 1's optical core — it discloses the light-source→reflective-element→position/quadrant-detector→torque conversion — but it lacks the entire rheometer: no base-mounted driving device, no lower sample assembly, no fluid-sheared upper sample assembly, and it uses a retroreflector of plane mirrors, not a prism. Strong § 103 art against claim 1 when combined with a parallel-plate rheometer reference; not § 102 art.

US 3,751,975 A — "Torsion digital viscometer"

• Full citation: Ono Sokki Co., Ltd. US 3,751,975 A.
• Dates: filed 1970-09-01; issued 1973-08-14.
• Description: Rotary viscometer in which a rotor turns in a viscous sample; the resulting torque twists a torsion bar. Torque is sensed non-contact, magnetically (toothed cylinders on the bar cooperate with magnetic pickups; phase difference is detected digitally) to avoid friction-coupling errors. Expressly seeks higher accuracy by removing mechanical contact with the torsion bar.
• Potentially anticipates: None as a whole. It discloses the "rotating lower element → torsion bar → non-contact torque readout" architecture that claim 1 generalizes, plus the accuracy rationale the '972 spec itself recites. But the sensing is magnetic, not optical — no prism, no light source, no four-quadrant detector. Relevant § 103 art / background art; not § 102 art for claim 1.

US 5,610,325 A — "Torsional rheometer for granular materials, slurries and gas-solid mixtures and related methods"

• Full citation: Rajagopal, C.; Rajagopal, K.R.; Yalamanchili, R.C. US 5,610,325 A. Assignee: Viscoustech, Inc.; DOE-funded.
• Dates: filed 1995-06-05; issued 1997-03-11.
• Description: Parallel-plate torsional rheometer: a stationary plate and a rotatable plate receive the specimen between them; a motor rotates the rotatable plate and a piezoelectric transducer reads the torque/axial forces applied to the stationary plate, converted to rheological properties by a computer. Roughened plate surfaces reduce slip; oscillatory or 360° modes.
• Potentially anticipates: None as a whole. This is the closest cited art to the mechanical architecture of claim 1 — a parallel-plate rotary rheometer in which one plate is driven and the opposed plate sees torque through the sheared specimen — and to claim 9's "lower sample / upper sample" arrangement. But the transducer is piezoelectric, not optical, so it cannot anticipate the prism/quadrant-detector limitation. High-value § 103 seed for the "rheometer + non-contact optical torque sensor" combination.

US 7,500,385 B2 — "System for in-situ optical measurement and sample heating during rheometric measurements"

• Full citation: Waters Investments Limited (family includes US 2007/0193343 A1; US 7,594,429 B2; US 7,526,941 B2).
• Dates: filed 2005-11-23; issued 2009-03-10.
• Description: Rheometer with an integrated optical measurement path permitting in-situ optical interrogation of the sample during rheometric testing (plus sample heating).
• Potentially anticipates: None. Discloses the concept of optical measurement inside a rheometer, but that optical channel is for sample characterization (spectroscopic/interferometric), not for reading a torsion-bar twist angle, and there is no prism/quadrant detector. Background/§ 103 art only. (Live fetch of this family was cut off by the step limit; description rests on the patent's front-page citation and general knowledge of the Waters optical-rheometry family — flagged as not independently re-verified today.)

US 7,594,429 B2 / US 2007/0193343 A1 — "System and method for improved optical measurements during rheometric measurements"

• Full citation: Liberatore, Matthew (US 2007/0193343 A1); and Waters Investments Limited (US 7,594,429 B2, issued 2009-09-29). Same family as US 7,500,385.
• Dates: filed 2005-11-23; published 2007-08-23 / issued 2009-09-29.
• Description: Improvements to optical (in-situ) measurement during rheometry — beam handling/alignment for better optical signal in a rheometer environment.
• Potentially anticipates: None. Optical-measurement-in-rheometer art, not torsion-bar/torque optical sensing. § 103 context only. (Not independently re-fetched — step limit.)

---

TIER 2 — Rheometers / rotating viscometers (structure, torque sensing, gap control)

US 3,667,286 A — "Viscometer"

• Full citation: Hittman Associates, Inc. US 3,667,286 A.
• Dates: filed 1970-04-21; issued 1972-06-06.
• Description: Three-element concentric viscometer: an unknown-fluid chamber rotated one way, a known Newtonian-fluid chamber rotated the other, and a floating passive torque-sensing plate brought to standstill; viscosity read from the balancing speed. Aimed at low shear rates (blood viscometry); fluid-bearing, low static friction.
• Potentially anticipates: None. Different operating principle (null-balance, opposing rotation), no torsion bar, no optical detection.

US 4,726,220 A — "Method of and apparatus for measuring rheological characteristics of substances"

• Full citation: Contraves AG. US 4,726,220 A.
• Dates: filed 1985-12-13; issued 1988-02-23.
• Description: Method/apparatus for measuring rheological characteristics — a rotating/oscillating measurement system typical of the Contraves (later Anton Paar) rheometer line.
• Potentially anticipates: None on the record as cited. General rotation-rheometer background. (Description limited to the title as cited; not independently re-fetched.)

US 5,228,331 A — "Viscometer"

• Full citation: Yamaichi Electric Co., Inc. US 5,228,331 A.
• Dates: filed 1990-11-29; issued 1993-07-20.
• Description: Rotational viscometer.
• Potentially anticipates: None. Background rotation-viscometer art.

US 6,776,028 B1 — "Induction sensor viscometer"

• Full citation: OFI Testing Equipment, Inc. US 6,776,028 B1.
• Dates: filed 2003-04-29; issued 2004-08-17.
• Description: Viscometer using an induction (non-contact electromagnetic) sensor for the torque/deflection readout of the rotating element (a "non-contact sensor instead of mechanical coupling" theme paralleling the '972 objective).
• Potentially anticipates: None. Non-contact sensing, but inductive, not optical. Relevant § 103 art on the motivation to replace mechanical torque sensors with non-contact sensing.

US 6,978,662 B2 — "Rheometer"

• Full citation: Thermo Electron (Karlsruhe) GmbH. US 6,978,662 B2.
• Dates: filed 2002-12-21; issued 2005-12-27.
• Description: Rheometer apparatus (Thermo/Haake line) with improved torque/drag-cup measurement and gap control — the "commercial traditional rheometer" the '972 background disparages for limited sensor accuracy and range.
• Potentially anticipates: None. It is background art that the '972 patent itself distinguishes.

US 7,275,419 B2 — "Rotation rheometer or viscosimeter"

• Full citation: Anton Paar GmbH. US 7,275,419 B2.
• Dates: filed 2004-05-24; issued 2007-10-02.
• Description: Rotation rheometer/viscometer with torque measurement of the rotating sample holder.
• Potentially anticipates: None. Background rotation-rheometer art. (Not independently re-fetched — step limit; description from front-page citation + general knowledge.)

US 2012/0111097 A1 / US 9,267,871 B2 — "Rheometer or viscometer"

• Full citation: Sierro, Philippe (US 2012/0111097 A1); Thermo Electron (Karlsruhe) GmbH (US 9,267,871 B2).
• Dates: Sierro filed 2010-11-10, published 2012-05-10; US 9,267,871 issued 2016-02-23 (priority 2010-11-10).
• Description: Rheometer/viscometer with improved torque measurement and sample handling.
• Potentially anticipates: None. Background art; relevant only as evidence of the conventional torque-sensor design the '972 patent improves upon.

US 2013/0245968 A1 — "Viscosity testing system and method of using the same"

• Full citation: Vibrac, LLC. US 2013/0245968 A1.
• Dates: filed 2012-03-13; published 2013-09-19.
• Description: Viscosity testing system (Vibrac line — small-sample rotational viscometry).
• Potentially anticipates: None. Background viscometer art.

US 4,343,190 A — "Moving die rheometer, method of testing materials therewith, and die for use therein"

• Full citation: Monsanto Company. US 4,343,190 A.
• Dates: filed 1980-06-02; issued 1982-08-10.
• Description: Moving-die (oscillating) rheometer for rubber/cure testing — rotor oscillates in a sealed die cavity; torque measured by a torque-sensing transducer.
• Potentially anticipates: None. Different geometry (moving-die vs. parallel-disc), no optical measurement.

US 4,202,204 A — "Apparatus and method for measuring changes in conditions in coagulating liquids"

• Full citation: Dr. E. Fresenius Chem.-Pharm. Industrie KG. US 4,202,204 A.
• Dates: filed 1977-09-13; issued 1980-05-13.
• Description: Rheological measurement of coagulating liquids (clotting) — a torsion-sensing viscometer for a changing sample.
• Potentially anticipates: None. Niche viscometer background.

US 8,763,447 B2 / US 2011/0252871 A1 — "Ultraviolet curable resin property measuring apparatus"

• Full citation: Jasco Corporation (US 8,763,447 B2); Jasco International Co., Ltd. (US 2011/0252871 A1).
• Dates: filed 2010-04-14; published 2011-10-20; US 8,763,447 issued 2014-07-01.
• Description: Apparatus measuring properties (including rheological/cure behavior) of UV-curing resins.
• Potentially anticipates: None. Niche cure-measurement background.

US 2015/0345937 A1 — "Test body for determining rotation errors of a rotating apparatus"

• Full citation: Carl Zeiss Industrielle Messtechnik GmbH. US 2015/0345937 A1.
• Dates: priority 2013-01-09; published 2015-12-03.
• Description: Test body/detection of rotational errors of a rotating apparatus (metrology, not rheometry).
• Potentially anticipates: None. Tangential metrology art; cited presumably for optical rotation-error detection.

US 2011/0252871 / US 7,526,941 / US 2007/0295055 — "Rheometer torque calibration fixture"

• Full citation: Waters Investments Limited (US 7,526,941 B2); Doe, Nigel (US 2007/0295055 A1).
• Dates: filed 2006-06-22; published 2007-12-27; US 7,526,941 issued 2009-05-05.
• Description: A fixture for calibrating a rheometer's torque channel.
• Potentially anticipates: None. Calibration accessory art.

---

TIER 3 — Chinese-language and other citations (mostly optical/indirect torque sensing)

---

Non-patent citations (for completeness)

• SIPO First Office Action, CN App. 201410138965, dated 2015-09-11 — foreign prosecution of the priority application (not prior art; noted because it is the only "prosecution" event in the record).
• Wang et al., "An approach to relationship between over-stress behavior and forming mechanism of viscous debris flow surges," Chinese J. Geological Hazard & Control, vol. 11, No. 3, Sept. 2000, pp. 53–57 — printed publication; rheology-of-debris-flow subject matter, not apparatus.
• WIPO English translation of the ISR/WO for PCT/CN2015/075982, dated 2015-07-17 — the PCT search report (procedural, not prior art).

---

4. Bottom line on § 102

The cited art's real value is § 103, not § 102. The strongest obviousness construct on the face of the patent is US 5,610,325 (or US 3,751,975) for the parallel-plate/torsion-bar rheometer, in view of US 5,705,810 (laser optical torquemeter) — and optionally US 3,751,975 for the express motivation to make the torque readout non-contact to remove friction-induced error. CN 101692033 A (FBG optical torque sensing) and CN 101750175 A (torsion-rod deformation torque sensor) are secondary references for the "optical readout of a torsion bar" sub-limitation. Note that the '972 patent's own background expressly acknowledges the prior-art problems (limited sensor accuracy/range, motor-driven gap control) — that admission narrows the distance between it and these references and will help a § 103 case.

---

5. Verification / confidence notes

• High confidence: The identity of US 10,337,972 B2, its claim set, priority date, and the "References Cited" list — all taken verbatim from the authoritative full text in your prompt and corroborated by Google Patents and FreePatentsOnline hits.
• Medium confidence: Descriptions of US 7,275,419, US 7,500,385, and US 2007/0193343 — the live fetches for these were cut off by my step limit; their bibliographic data (assignee, dates, title) is from the patent's front page, and the functional descriptions rest on the cited titles plus general knowledge. Re-verify against USPTO Patent Center (https://patentcenter.uspto.gov) if you rely on them in a filing.
• Unverified flag (carried forward): the "Jun Qi Zhang" vs. "Jun Zhang" inventor-name variant — immaterial to prior art but noted for consistency with the earlier sections.
• No anticipation asserted where none exists. I have not manufactured § 102 hits; where the answer is "no," it is stated as such.

If you'd like, I can next build a formal claim-chart-style § 103 matrix mapping each claim-1 limitation to the US 5,610,325 + US 5,705,810 + US 3,751,975 combination, or run the same analysis against the sibling CN 103926171 B.

I'll ground the prior-art references before building the § 103 combinations. Running targeted lookups on the cited references now.

Obviousness Analysis — U.S. Patent No. 10,337,972 ("High-speed rheometer") under 35 U.S.C. § 103

Analysis date: April 26, 2026 · Sources: the Prior Art section of the US 10,337,972 record (Patent Citations, Non-Patent Citations, family data) plus targeted live verification of the key references run 2026-04-26.

AIA applies. Priority date is 2014-04-08 (CN 201410138965.7); the US filing is 2015-04-07. Because the effective filing date post-dates 16 March 2013, AIA § 102(a)(1)/(a)(2) and AIA § 103 govern. Every reference in the Citations list published before 2014-04-08 and is therefore available as printed publication art, except US 2015/0345937 A1 (published 2015-12-03), which is usable only as § 102(a)(2) art if its earlier 2013-01-09 filing is a US/ PCT application publication — its own priority column shows 2013-01-09, so treat it as a § 102(a)(2) candidate only.

---

0. Corrections to the previously generated sections (flagged per instructions)

Before the analysis, three contradictions between earlier sections and the verified record:

1. "US 7,276,462 equivalents" does not exist in the record. The prior generated "Recommended next steps" cites a "Waters/Thermo rheometer optical-measurement family (US 7,500,385; US 7,276,462 equivalents; US 7,594,429; US 7,526,941)." There is no US 7,276,462 in the Citations list. The near-miss is US 7,275,419 B2, Anton Paar, "Rotation rheometer of viscosimeter" — a different assignee and a different technical thrust (measuring-system identification and gap/lift control, not optical measurement). Do not rely on "US 7,276,462."

2. The Waters family is not optical torque art. The earlier strategic summary listed the Waters references among art "on the face of the patent" relevant to "the optical torque-measurement core of claim 1." Verified content contradicts that characterization: US 7,500,385 and US 2007/0193343 / US 7,594,429 use a laser and detector to measure the fluid's optical properties (birefringence, dichroism, scattering), and use the laser/encoder to track plate angular position — they do not transduce torque from beam deflection. Their value to a § 103 attack is (a) parallel-plate rheometer architecture with one driven plate and a sample between plates, and (b) proof that laser/detector instrumentation was a routine rheometer retrofit. They are not anticipatory of claim 1's optical torque readout. This materially narrows the earlier section's claim.

3. Reference-count drift. The record lists both "Citations (26)" and "Patent Citations (30)" blocks with overlapping but non-identical membership. Element mapping below uses the union, with the verified subset flagged.

---

1. Legal framework and the person having ordinary skill in the art (PHOSITA)

Framework. Obviousness is assessed under Graham v. John Deere, 383 U.S. 1 (1966): scope and content of the prior art, differences between the prior art and the claims, level of ordinary skill, and secondary considerations. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), the prior art need not contain an express teaching, suggestion, or motivation; predictable combinations, simple substitutions of known elements, and the use of known techniques to improve similar devices are all obvious. The MPEP 2143 rationales (A)–(G) are cited below by letter.

PHOSITA (my construction): a person with a bachelor's degree in mechanical engineering, instrumentation, or applied physics, plus 3–5 years of experience in rotational rheometry/viscometry and precision optical metrology, including familiarity with torsion-bar torque transducers, parallel-plate measuring geometries, air-bearing spindles, piezo nanopositioners, and non-contact displacement sensors. This is a mechanical-instrumentation artisan, not a software or pure-optics specialist — relevant because the claim set is dominantly mechanical.

---

2. Claim-limitation inventory (what must be accounted for)

---

3. Combination 1 — Primary attack on claims 1 and 2

Primary reference: US 3,751,975 (Ono Sokki Co., "Torsion digital viscometer," 1973-08-14) — of record.
Secondary reference: US 5,705,810 (Wang, "Laser optical torquemeter," 1998-01-06) — of record.
Tertiary: US 7,500,385 / US 2007/0193343 (Waters/Liberatore) — of record — for the parallel-plate, driven-plate-with-sample-between-plates geometry.

3.1 What US 3,751,975 discloses (verified)

Verified text (freepatentsonline.com/3751975.html; patents.google.com/patent/US3751975):

• A rotary viscometer with tank 1, rotor/propeller 2 on shaft 3, motor 24, and a torsion bar 6 connecting the motor side to the rotor side.
• The express object is to eliminate "an error caused by the mechanical friction between the torque detector and the rotary portion" and to obtain "greater accuracy." Torque "is detected by a rotary member without making contact mechanically to the torsion bar."
• FIG. 5 embodiment: the tank (sample holder) is driven in rotation and the rotor is driven by the viscous fluid, with the torsion bar fixed at its other end — i.e., the fluid itself drives the second member, which twists the torsion bar. That is the exact kinematics of claim 1's "upper sample assembly rotatable under the driving of the fluid" and "torsion bar … torsional upon the rotation of the upper sample assembly."

This single reference therefore supplies claim 1's base, driving device, fluid-carrying driven member, fluid-driven second member, and torsion bar. Its gap vis-à-vis claim 1 is (i) the measuring geometry is a rotor-in-tank rather than parallel plates, and (ii) the torque readout is a magnetic-pickup phase-difference scheme that is non-contact but not optical.

3.2 What US 5,705,810 (Wang) discloses (verified)

Verified text (patents.justia.com/patent/5705810; patents.google.com/patent/US5705810):

• An angular-twist torque transducer: a torsional shaft S4 and transfer shaft S5, with plane mirrors M1, M2 mounted respectively on the torsionally stressed shaft and the reference shaft forming a retroreflector.
• A laser emitter (1) directs a collimated beam at the mirrors "essentially perpendicular to the shaft"; torque-induced angular twist rotates the reflected beam "through an angle that is twice the angular displacement between mirrors."
• "The reflected beam S′ is focused … onto a position sensitive photo-detector 6 on which the deflected angle is measured as a position change on the detector surface."
• Critically, the detector is described as a bi-axial position sensing photoelectric detector whose incident current "is divided into four components, which are discharged over individual collector strips." A four-collector position photodetector is, structurally and functionally, a four-quadrant detector — a point the Examiner can press in claim construction.
• Wang expressly claims a "wide speed and temperature range," non-contact operation with no bearings, no lubrication, and no electronics on the rotating part, and states the transducer "can be used for different sensing members to cover different torque ranges."

3.3 Mapping

Parallel-plate architecture (for the "fluid disposed on the lower sample assembly / upper assembly contacting fluid" recitation): US 7,500,385 (verified, patents.justia.com/patent/7500385) discloses "an upper rotatable plate 12 and a lower fixed plate 14 … a fluid sample 13 in contact with plate 12 during plate rotation or oscillation," plus a laser and detector for in-situ optical probing. Anton Paar's US 7,275,419 and the Anton Paar disclosure in US 2013/0298645 confirm that either the upper (Searle) or the lower (Couette) measuring part may be the driven one — so inverting Waters' driven-upper-plate geometry to a driven-lower-plate with the sample carried on the lower plate is a conventional, predictable orientation choice.

3.4 Why the PHOSITA would combine — motivation

• Rationale (C)/(D): use a known technique to improve a similar device in the same way. US 3,751,975 states the problem (friction-induced error from a mechanically coupled torque detector) and states the goal (greater accuracy, non-contact detection). US 5,705,810 supplies the same kind of device (an angular-twist torque transducer) solved the same way (non-contact optical beam deflection), and expressly touts wide speed range and no bearings — precisely the attributes needed for a high-speed rheometer. Applying Wang's readout to Ono Sokki's torsion-bar viscometer is a same-field, same-problem, same-solution substitution.
• Rationale (B): simple substitution of a known element for another to obtain predictable results. The magnetic phase-difference pickups of US 3,751,975 are replaced by an optical beam-deflection transducer. Both measure the same physical quantity (angular twist of a torsion bar); both are non-contact; the substitution changes only the transducer, not the measurement principle or the instrument's mathematical model.
• Rationale (A): combining known elements with predictable results. A torsion bar (known) + laser/position-detector readout (known) + parallel-plate sample cell (known) yields nothing more than the expected ability to measure shear torque optically. No new function emerges; the combination is a concatenation.
• Rationale (F): design incentives/market forces. The patent's own Background recites the field's demand for wider shear-rate range (>10⁴ s⁻¹ toward 10⁶–10⁷ s⁻¹) and notes that the commercial PCS ultra-shear viscometer handles high rates but not low ones. That articulated market need — a single instrument spanning low and high shear rates — is the incentive to graft a high-bandwidth, high-resolution optical transducer onto an existing rotary rheometer. Demand-driven motivation of this kind is exactly what KSR endorses.
• Rationale (E) as to the "prism"/"four-quadrant detector" specifics: substituting a solid right-angle prism for Wang's pair of 90° plane mirrors is one of a finite, small, and predictable set of ways to implement a 90° deflection at the top of a rotating shaft (mirror pair, right-angle prism, roof/Pentax prism, retroreflector). Prisms are standard where rigidity and alignment stability are desired — a strong design incentive on a high-speed rotating spindle, where a two-mirror assembly with an adjustable 90° included angle is exactly the thing that drifts. Likewise, quadrant photodiodes were, by 2014, a standard commercial part for "where did the spot land?" measurement (optical tweezers, AFM optical levers, laser-beam position sensing, optical-disk servos). Using a known detector in its known role for its known benefit is Rationale (A)/(B).

3.5 Anticipated patent-owner rebuttals and why they fail

• "Wang is a shaft torquemeter, not a rheometer; different field." Rationale (F) answers this: the fields are the same discipline (precision torque metrology on a rotating member), and Ono Sokki already put a torsion bar in a viscometer. Wang does not teach away — it explicitly says the same transducer "can be used for different sensing members to cover different torque ranges," i.e., it is inviting re-application.
• "Wang's detector necessarily receives the beam through a window only ~every 20° of rotation; that's a duty-cycle limitation." That is a limitation of Wang's specific retroreflector-on-a-rotating-shaft geometry, not of the optical principle. In the claimed rheometer the reflecting prism rotates only by the small torsional wind-up angle at the top of the bar, not through full revolutions, so the duty-cycle issue does not arise. To the extent the owner argues a teaching away, the argument also cuts against the claim, because the same contrast shows the combination requires no more than relocating a known reflective element to the non-rotating (torsionally-deflecting only) end of a known torsion bar.
• "Four-quadrant detector ≠ Wang's PSD." Claim construction will govern. Markedly, Wang's detector is a four-collector device; and even if construed narrowly as a monolithic quad-cell, the reference to a four-quadrant detector is a wholesale substitution of a commercially standard position sensor for an equally standard position sensor, i.e., Rationale (B) at minimum, and likely § 102 in view of Wang alone as to that element.

Claim 2 (≥1 reflector generating second reflected light) is a fortiori obvious: Wang's beam already undergoes multiple reflections (M1 → M2 → back), and folding a beam path with simple plane mirrors to fit an instrument envelope is routine — the patent itself justifies the reflector solely by the desire to avoid "a dimension of the base too long in a certain direction," i.e., a stated packaging rationale with no unexpected effect. Rationale (A)/(C).

Confidence for Combination 1 (claims 1–2): high that a prima facie § 103 rejection can be made; moderate on the "prism" substitution being accepted without a secondary reference showing a prism, purely because "prism" invites a narrow construction.

---

4. Combination 2 — Claims 3 and 4 (closed-loop nulling with a piezo-driven reflector)

4.1 Approach

Claim 3 requires (i) a predetermined position on the quadrant detector, (ii) detection of the voltage difference before and after prism rotation, (iii) a feedback circuit producing an amplified signal, and (iv) a piezoceramic deflection platform on the reflector that deflects the reflector so the spot returns to the predetermined position. Claim 4 fixes the count at four reflectors with the piezo on the first.

4.2 Art mapping and motivation

• Elements (i)–(ii) are inherent in a quadrant detector. The entire purpose of a four-quadrant detector is to output difference signals (e.g., (A+C)−(B+D)) proportional to spot displacement from the center. A "predetermined position" that yields zero difference signal is simply the quad cell's null. No separate reference is needed.
• Element (iii)–(iv): servo-nulling is a classic, well-known instrument technique, and piezo mirror deflection is a well-known actuator. Null-balance measurement (servoing a deflection back to a null and reading the servo effort as the measurand) is the operating principle behind null-balance recorders, servo accelerometers, and force-balance transducers. Four-quadrant-detector-plus-feedback-to-a-beam-steering-element is likewise the standard architecture for beam stabilization and for servo track-following in optical drives, and piezo/electrostrictive mirror tilting is the standard solid-state beam-steering actuator.
• "Obvious to try" (Rationale E) with a strong design incentive. The specification itself supplies the problem that the nulling solves: at high rotating speed the shearing torque grows, the prism twist grows, and "the light spot … tends to move out of the detecting surface of the four-quadrant detector." Triangulating the spot back onto the detector is the recognized remedy, and it is the very remedy that converts a limited-linear-range position sensor into a wide-dynamic-range one — the asserted "several orders of magnitude" gain claimed in the patent. The specification's asserted benefit (broadened dynamic range) is precisely the predictable consequence an artisan would expect from servo-nulling, not an unexpected result.
• Claim 4 (four reflectors; piezo on the first): Rationale (A)/(E). The number of folding mirrors in a compact optical train and the location of a single steering actuator along that train are design choices within a finite set, made for the recognized reasons of envelope fit and of actuator placement (upstream actuator requires smaller angular throw for the same spot correction, and is easier to mount off the moving stage).

4.3 The one soft spot, and the candidate that fills it

The specific combination of a piezoceramic mirror-deflection platform inside an optical torque readout is the most "inventive-looking" structural recitation in claims 3–4. Two candidate sources to develop:

• CN 102830041 A (Jinan Meiyilin, "Stepping scanning torque measuring device in magnetic-levitation bearing support structure," of record) — the title itself suggests a scanning/stepped-deflection torque measurement device on a magnetic-bearing (i.e., non-contact, high-speed) support. If its specification discloses a driven deflector that scans the beam across a detector and a control loop that repositions the beam, it becomes the explicit teaching for claim 3's feedback-plus-deflector. I could not retrieve its text (search returned no body content), so this is a to-be-verified lead, not an established disclosure. Treat as the single highest-value verification task for claims 3–4.
• US 2013/0298645 A1 (Anton Paar) — not of record; found in search (freepatentsonline.com/y2013/0298645.html). Verified: it discloses piezoactuators 11, 12 controlled by a voltage generator to correct gap height, plus position and normal-force sensors. That is art for claim 5's piezoceramic displacement platform, not for claim 3's deflection platform. Useful as corroboration that piezo actuation was an accepted instrument-control building block in this exact field at the relevant time.

Confidence for Combination 2 (claims 3–4): moderate. Claims 3–4 are the most defensible of claims 1–8 if the owner can persuasively argue that the pointing of piezo beam-steering to a rheometer's torque readout was non-obvious. But the motivation supplied by the specification's own dynamic-range problem, plus the ubiquity of quad-cell servo-nulling, makes a strong case; the rejection is materially stronger if CN 102830041 A or an equivalent beam-stabilization reference is added.

---

5. Combination 3 — Claims 5–8 (gap metrology and torsion-bar geometry)

5.1 Art mapping

5.2 Motivation

Rationale (D) + express prior-art recognition of the problem. The Anton Paar references do not merely hint at gap control — they quantify the need: US 6,499,336 and US 7,275,419 both derive that "a +1% height measurement error produces a 1% reduction in viscosity," that the gap "has to be determined with an accuracy better than 10 µm or 20 µm," and that thermal drift, stand stiffness, and stand deformation cause the gap to "vary by a few 0.1 mm." Anton Paar's answer is a non-contact position sensor feeding a control loop that drives a gap-adjusting actuator — functionally the same closed loop that claims 5 recites, with an eddy-current sensor as the expressly enumerated sensor option. A PHOSITA seeking to extend a rheometer to high shear rates (thin gaps, high speeds, thermal loading) has every incentive to adopt this proven scheme.

The coarse/fine two-stage split (manual platform for large travel and sample exchange; piezo platform for nanoscale gap trimming) is textbook instrument design: the manual stage addresses the "long displacement route … low resolution … to replace the fluid to be tested" function the patent itself describes; the piezo stage addresses resolution. Splitting travel and resolution across two stages is a well-known engineering expedient, and the specification's own Background criticizes motor-driven gap control for noise and low accuracy and for being unable to reach "relatively smaller gaps" — an explicit problem statement that motivates replacing a motor with a solid-state actuator and a non-contact sensor.

Claims 6–8 (circular first fixing part; rectangular-outer/circular-inner ring second fixing part; four bar parts; connecting holes at four corners) are geometric optimizations of a flexure element. With no asserted and no apparent unexpected result — the specification simply says the elongated bar "tends to deform upon an external force," i.e., it behaves as a spring, exactly as expected — these limitations do not independently patentably distinguish. Rationale (A). Confidence: high that claims 5–8 are prima facie obvious when Anton Paar US 7,275,419 is combined with a non-contact-sensor gap-control teaching (ideally one that names eddy-current sensing, such as US 6,499,336 / DE 100 47 793 — which, note, are not of record and should be searched and added).

---

6. Combination 4 — Claims 9–13 (self-leveling sample mounts and the glue-up method)

6.1 What is claimed and why it is functionally familiar

Claim 9's lower-sample mount is a gimbal/self-aligning spherical seat: two spaced pillars on the connector engaging two oversized grooves in the sample, with a spherical-surfaced platform between them that lets the sample tilt freely until its top face is parallel to the upper sample's face, while the pillar-groove pairs continue to transmit torque. Claims 10–13 mirror this on the upper side (two supporting bars + a spherical supporting ball + connecting grooves + receiving groove), fixed with glue, plus the assembly method: seat the upper sample on the lower sample so parallelism is established by gravity, bring the connector into contact, bond, then lift.

• Spherical-seat self-alignment is ancient and ubiquitous — self-aligning spherical roller bearings, spherical washers under bolt heads, ball-and-socket mounts, and self-leveling instrument feet all perform the identical function: permit angular compliance while transmitting load. The asserted benefit (the sample "may rotate and sway freely … and the lower sample 31 and the upper sample 41 may be maintained to be parallel") is the definition of that class of joint, i.e., a predictable result (Rationale A).
• Two pillars in oversized grooves = an Oldham-type coupling. A spaced-pin-and-oversized-slot coupling is the canonical way to transmit torque between two members that are permitted relative angular/translational misalignment, and the specification's observation that the oversized groove and pillar "may not interfere with each other" during sway is exactly the reason an Oldham/universal coupling uses clearance. This gives a structural analogy argument that does not depend on a single anticipatory reference.
• Claim 11's method is a "transfer the reference plane" fixturing technique. Establish the desired geometric relationship against a reference (the lower plate), then lock the part to its carrier while so fixtured, then remove the reference. This is standard practice in optical and precision-fixture assembly (aligning a mirror or a lapping plate to a reference surface before bonding). The method recites no step that departs from that practice, and the binder is plain glue (claim 13).

6.2 Weakness of the record here, and what to search

This is the weakest-supported cluster in the of-record citations. The listed references that plausibly touch sample-mount alignment/die geometry are:

• US 4,343,190 (Monsanto, "Moving die rheometer … and die for use therein," of record) — a biconical-rotor, sealed-die rheometer; die geometry and die-mounting alignment are its subject matter. Content not verified in this run. A die that self-seats or self-aligns would be the best available primary reference for claims 9–13 and should be pulled in full.
• US 7,526,941 / US 2007/0295055 (Waters, "Rheometer torque calibration fixture," of record) — mounting a fixture/plate on a rheometer with reproducible geometry. Partly verified (abstract-level only via the Waters assignee listing); the fixture-mounting disclosure should be checked for alignment features.
• US 2012/0111097 A1 / US 9,267,871 (Sierro, Thermo Electron, "Rheometer or viscometer," of record) and US 6,978,662 (Thermo Electron Karlsruhe, of record) — commercial rotational rheometers, which necessarily address plate mounting and parallelism. Content not verified.
• Outside the record: the broad corpus of self-aligning/kinematic mount and Oldham coupling art, which supplies the motivation and the structural analogue even without a rheometer-specific reference.

Confidence for claims 9–13: low-to-moderate. I would not presently assert a clean prima facie case on claims 9–13 from the of-record citations alone, because I have not verified a reference that discloses the specific combination of (a) two spaced torque-transmitting pillars, (b) an interposed spherical platform, and (c) oversized receiving grooves arranged to permit automatic leveling — and the closest mechanical analogue (a gimbal seat plus an Oldham coupling) has to be assembled from two different arts, which invites a motivation-to-combine challenge even under KSR. The strongest available non-record argument is Rationale (F): parallel-plate rheometry requires plate parallelism, gap control is known to dominate accuracy (Anton Paar's own ±1%-gap→±1%-viscosity derivation), and self-aligning joints are the classic mechanical answer — but a rejection on this record would likely need one more reference.

---

7. Consolidated assessment

---

8. Secondary considerations (Graham factor 4)

Nothing in the record supports a nexus-bearing secondary consideration:

• Unexpected results: none asserted for any structural feature; every asserted benefit in the specification (wider dynamic range from nulling, higher gap accuracy from piezo + eddy sensor, plate parallelism from self-leveling) is the expected consequence of the known mechanism relied upon.
• Long-felt but unmet need: the specification asserts field-wide need (existing rotation rheometers capped near 5,000 rpm / 10⁴ s⁻¹; the PCS ultra-shear viscometer covering only 10⁶–10⁷ s⁻¹). This is genuine, but long-felt need cuts for the challenger when the solution is a known element applied in a known way — it supplies the Rationale (D)/(F) motivation rather than rebutting it.
• Commercial success / copying / industry praise: no evidence in the record. Note the patent's own statement that commercial ultra-shear instruments already exist, which undercuts any "first to market" narrative for the optical-readout concept.
• Teaching away: none identified. US 5,705,810 affirmatively promotes re-use of its transducer with different sensing members and different torque ranges; US 3,751,975 promotes non-contact torque detection; the Anton Paar family promotes non-contact gap sensors closed-loop to a gap actuator. The references point toward, not away from, the claimed combinations.

---

9. Verification gaps and confidence

Confidence in the overall conclusion: high that claims 1–2 and 5–8 are obvious on the of-record art; moderate for 3–4; low-to-moderate for 9–13. I would not represent claims 9–13 as clearly obvious on this record without the additional searching below.

Reference contents I verified in this run (full or substantial text): US 5,705,810; US 3,751,975; US 7,500,385; US 2007/0193343 / US 7,594,429; US 7,275,419 (abstract/description level). Plus two non-record references located incidentally: US 6,499,336 / DE 100 47 793 B4 (Anton Paar, non-contact incl. eddy-current gap sensing with closed-loop gap control) and US 2013/0298645 A1 (Anton Paar, piezo gap-height actuators).

Reference contents I could NOT verify and did not rely on: CN 102830041 A (stepped/scanning torque measuring device on magnetic-levitation support — highest-value gap, directly implicated in claim 3); CN 101750175 A (single-end-deformation torque rod sensor — implicated in claim 5); US 4,343,190 (Monsanto moving die rheometer — implicated in claims 9–13); US 7,526,941 / US 2007/0295055 (Waters torque calibration fixture — implicated in claims 9–13); US 6,978,662; US 9,267,871 / US 2012/0111097; US 6,776,028 (OFI induction sensor viscometer); US 3,667,286 (Hittman viscometer); US 4,726,220 (Contraves); US 5,220,331 (Yamaichi); US 5,610,325 (Viscoustech); US 4,202,204 (Fresenius); CN 102706800 A; CN 101923033 A; CN 101692033 A; CN 102112861 A; CN 101592581 A; CN 101308076 A; US 2013/0245968 A1; US 2011/0252871 A1; US 2015/0345937 A1. Statements above about those references are drawn from their titles and this record only and are expressly flagged as unverified.

Recommended priority actions for a § 103 petition or invalidity contention:

1. Pull CN 102830041 A in full (with a certified translation) — it is of record and is the likeliest express teaching for claim 3's deflector-and-feedback nulling.
2. Add US 6,499,336 / DE 100 47 793 B4 (Anton Paar) to the gap-control combination for claims 5–8; it names the eddy-current sensor expressly, which the of-record Anton Paar reference (US 7,275,419) does not.
3. Pull US 4,343,190 and US 7,526,941 / US 2007/0295055 in full for claims 9–13, and run an independent search on self-aligning spherical seats + clearance-slot torque couplings in rheometer sample mounts, plus kinematic/self-leveling plate mounts — the record as it stands does not close claims 9–13.
4. Consider whether US 5,705,810 alone supports a § 102 anticipation of the "four-quadrant detector" element under a broad construction (its detector divides incident current into four collector strips), leaving only the "prism" phrase as the § 103 gap — this reduces the claim-1 case to a single-substitution argument.