US 10190900

I'll verify the patent against live sources before summarizing.

Let me check for litigation and any CAFC/Federal Circuit activity.

Verification note

I ran live searches against patent databases and public docket/news sources. The patent records I retrieved (Google Patents, Justia, Onscope) confirm the bibliographic data below and are consistent with the authoritative full text you supplied. Caveat: I do not have direct query access to PACER, the CAFC docket system, or USPTO PatentCenter in this session; my searches returned no litigation, IPR, or Federal Circuit appeal involving US 10,190,900. That is a negative result from web/aggregator sources, not a certified docket clearance.

---

US 10,190,900 B2 — Bibliographic Summary

Abstract (verbatim): "A method for measuring fluid flow within a fluid flow pipe includes partitioning a fluid flow within a pipeline with a nozzle bank, wherein a predetermined number of stepped nozzles is open; measuring the fluid flow in at least one partitioned fluid stream with at least one mass flow device; and calculating a total fluid flow within the pipeline. The nozzle bank includes a plurality of nozzles, each nozzle having a single step at about 1 throat diameter from an inlet plane of the nozzle, wherein the single step is an increase corresponding to about 10% of the throat diameter and has a length of about throat diameter/2."

---

Independent Claim — Plain-Language Overview

Claim 1 (the sole independent claim) is a method claim with three steps plus a structural limitation on the nozzles:

1. Partition the pipeline's fluid flow using a nozzle bank made of multiple stepped nozzles, where a predetermined number of those nozzles is open (the rest can be closed).
2. Measure the flow in at least one of the resulting partitioned streams using at least one mass flow device.
3. Calculate the total pipeline flow from that measurement.

The closing "wherein" clause defines the stepped nozzle geometry that must be in the bank: each nozzle has a single step located at about 1 throat diameter from the nozzle inlet plane; the step is an increase of about 10% of throat diameter; and the step has a length of about throat diameter / 2.

Dependent claims:

• Claim 2 — total flow = (number of open stepped nozzles) × (measured flow rate of at least one open nozzle).
• Claim 3 — nozzle bank has 2 to 100 stepped nozzles.
• Claim 4 — the predetermined number of open nozzles is 1 to 10.
• Claim 5 — mass flow device is a gravimetric meter prover.
• Claim 6 — mass flow device is a mass flow meter.
• Claim 7 — mass flow device is a Coriolis mass flow meter.

Technical gist: The stepped throat induces an oblique shock that closes off the sub-sonic boundary-layer mass flow, so once the nozzle is choked (downstream pressure ~12–18% below upstream; throat/inlet ratio ≈0.528), mass flow becomes essentially invariant to pressure ratio. The nozzle bank thereby acts as a flow partitioner, not a meter, letting a smaller measured stream be scaled up to total pipeline flow without thermodynamic/fluid-dynamic flow-rate equations.

---

Litigation Check (as requested)

• No CAFC 2026 appeal or district court case asserting US 10,190,900 was found in my searches.
• The CPA-related litigation that does surface involves different rights, not this patent: RJ Machine Co. v. Canada Pipeline Accessories Co. (W.D. Tex. 1:13-cv-00579 and 1:15-cv-00505), concerning the expired Laws Patent US 5,341,848 and the "50E" trademark/trade dress (TTAB 92057120); and Canada Pipeline Accessories Co. v. Canalta Controls (S.D.W. Va.). Do not attribute those cases to the '900 patent.
• Related CPA-family patents appear in the "similar documents" list (e.g., US 2016/0097411 A1 and US 2016/0334249 A1, "Flow Conditioner…"), but these are distinct inventions.

Uncertainty disclosure: I cannot affirm with high confidence that no litigation or post-grant proceeding exists, since I lack direct PACER/PTAB/CAFC docket access here. I also note the specification contains a typo ("Canada Pipieline Accessories (CPA) no step") in the FIG. 4 discussion — reported as-is, not corrected.

I'll search multiple litigation-focused sources for this specific patent number.

Let me search dedicated litigation databases and court sources.

Let me check patent-specific litigation databases and Google Patents' litigation tab.

Litigation Search: US Patent No. 10,190,900

Result: No litigation involving US Patent No. 10,190,900 was found.

I searched multiple litigation-focused sources — including the Unified Patents litigation portal (portal.unifiedpatents.com/litigation/caselist), general patent-litigation web sources, and docket aggregators (UniCourt, CourtListener, Stanford NPE database) — and could not identify any case in which US 10,190,900 ("Method of fluid flow measurement using nozzle bank") has been asserted or challenged in litigation, the PTAB, or the ITC.

Case table (relevant results limited to the patent owner):

Important distinction — what I did find (and why it does not count):

The patent owner, Canada Pipeline Accessories Co. Ltd. ("CPA"), appears in litigation, but that litigation concerns entirely different IP (flow conditioners, trademarks, and an antitrust/declaratory-judgment dispute) — not the '900 patent:

• RJ Machine Company, Inc. v. Canada Pipeline Accessories Co. Ltd., No. 1:15-cv-00505 (W.D. Tex., filed 2015). This was an antitrust/declaratory-judgment action regarding the "50E" flow-conditioner design and CPA's trademark rights (i.e., U.S. Patent No. 5,341,848 (the "Laws Patent," expired 2011) and design/trademark registrations — not US 10,190,900). CPA prevailed on the trademark issues. Sources: Law360 case page; CourtListener opinion; Stanford NPE database.
• TTAB Cancellation No. 92057120 (RJ Machine v. CPA) and related "CPA 50E"/"50E" trademark proceedings — again, trademark matter, not the '900 patent.

None of these involve US 10,190,900, and the '900 patent has a different priority/filing lineage (filed 2018-04-02; priority to provisional 62/481,871 filed 2017-04-05).

Caveats and confidence:

• I could not complete exhaustive citation-level verification before reaching search limits, so this should be treated as a good-faith "none found" rather than a certified negative.
• US 10,190,900 issued relatively recently (2019-01-29) and has recorded maintenance-fee payments through the 8th year (2026-01-29) per the patent's legal-events record, so it remains in force; the absence of litigation may simply reflect that CPA has not needed to (or chosen not to) assert it.
• Litigation databases can lag, and not all district-court filings name patents in easily searchable form. If you need a definitive answer, the authoritative check would be a direct search of PACER / CourtListener (RECAP) docket text and the USPTO PTAB and ITC EDIS systems for "10,190,900" / "10190900," which I was unable to run to completion here.

If you'd like, I can attempt a focused follow-up against a specific docket database (e.g., a particular PACER district or the PTAB caselist) to close the remaining verification gap.

I'll verify the "no proceedings on file" baseline against live sources before writing this up.

AIA Trial Proceedings — US 10,190,900 B2 ("Method of Fluid Flow Measurement Using Nozzle Bank")

Source of record: the structured "PTAB proceedings on file" block supplied in this prompt (USPTO Open Data Portal ingest) returns no AIA trial proceedings for US 10,190,900. That is the canonical list. Web corroboration checks run against PTAB/litigation aggregators and general sources this session returned no IPR, PGR, CBM, or Director Review record naming this patent or its owner in a PTAB challenge.

---

Proceedings overview

Zero AIA trial proceedings on file. There is no active IPR, no claims-invalidated proceeding, no claims-sustained FWD, no settlement, and no institution denial to report — 0 / 0 / 0 / 0 / 0. Bottom line for a defendant: the patent is intact and unhardened-and-untested simultaneously — claims 1–7 stand exactly as granted on 2019-01-29, but because no panel has ever construed or tested them, an IPR-based defense remains fully available and unencumbered by § 315(e) estoppel. There is no prior PTAB record to work from, and no prior PTAB record working against you.

---

Proceedings

None. There are no proceedings to list, so the per-proceeding template is not populated — populating it would require fabricating numbers, and the instruction is explicit not to invent proceeding numbers.

What I checked and did not find (negative results, recorded for the file):

• USPTO Open Data Portal AIA-trial feed for the patent — no hits (per the structured block).
• General web queries for IPR / PTAB / petition paired with "10,190,900" and with "Canada Pipeline Accessories" — no petition, institution decision, FWD, or Director Review surfaced.
• The only PTAB/TTAB-adjacent record involving the patent owner is TTAB Cancellation No. 92057120 (RJ Machine Co. v. Canada Pipeline Accessories Co., "50E" trademark/trade dress). That is a trademark proceeding before the TTAB — not an AIA trial, not about this patent, and it does not belong in an IPR/PGR/CBM table. Source: https://ttabvue.uspto.gov/ttabvue/ttabvue-92057120-CAN-9.pdf

---

Strategic summary

Claim status of US 10,190,900. Every claim is UNTESTED, not sustained and not canceled. All seven claims — independent claim 1 plus dependents 2–7 — stand as granted. No claim of this patent has ever been construed by the Board, no claim has been canceled, and no FWD exists to cite. Contrast this with the situation the prompt's framing anticipates ("the patent has survived two IPRs and is hardened"): there is no survivorship here, because there has been no attack. The patent's enforceability rests entirely on the original ex parte examination, which drew only 5 U.S. patent documents and 3 non-patent citations — the weakest form of validity testing.

Estoppel landscape. No § 315(e)(2) estoppel attaches to anyone. Because no IPR/PGR has been instituted and no FWD has issued against this patent, there is no petitioner, RPI, or privy carrying a bar on § 102/§ 103 grounds that were raised or reasonably could have been raised. For a defendant currently facing assertion, the entire prior-art universe is available — including, notably, art the applicant's own family published and the inventors' own earlier work. Specifically and worth flagging as a live opportunity: Studzinski et al. (1995), "Nova's Gravimetric Meter Prover and Sonic Nozzle Facility" is a reference cited on the face of this very patent, and the associated NOVA nozzle literature describes "an intentional step change in the diameter at the throat. This step prevents the possibility of downstream disturbances propagating upstream through the boundary layer to the nozzle throat." (NOVA inter-laboratory flow measurement program, 1996; https://nfogm.no/wp-content/uploads/2019/02/1996-03-North-American-Inter-Laboratory-Flow-Measurement-Testing-Program-Karnik-NOVA.pdf). No estoppel or § 325(d)-adjacent cloud stands between a petitioner and that ground. Caveat: § 325(d) discretion is a risk given the examiner already had that reference, not a bar — and under Advanced Bionics the Board would weigh whether the same art/arguments were previously presented.

Pattern signals. No pattern exists to detect:

• Repeat petitioner: none — no petitioner at all.
• Aggressive PTAB appellant: not applicable — there is no FWD for the patent owner to appeal.
• Defensive aggregator: no RPX, Unified Patents, AST, LOT, or OIN involvement in the chain. Ownership remains with Canada Pipeline Accessories Company Ltd. (Calgary), the original operating assignee, on a single 2018-04-27 confirmatory assignment (Reel 045653 / Frame 0261), with the 8th-year small-entity maintenance fee paid 2026-01-29 and anticipated expiration 2038-04-02.
• Practical read: CPA is an operating manufacturer, not a patent-monetisation entity, and its enforcement energy has historically gone to trademark/trade dress and antitrust (RJ Machine, W.D. Tex. 1:15-cv-00505; Canalta Controls, N.D. W. Va. 3:12-cv-08448), not patent assertion. That is consistent with — and likely explains — the absence of IPRs: IPRs follow assertion (35 U.S.C. § 315(b)'s one-year bar runs from service of an infringement complaint), and this patent has not been asserted in any litigation I could find.

---

Recommended next steps

If you are a defendant and no claims have been invalidated: There is no FWD to link to and no disposition to quote — I will not manufacture one. The operative fact is the inverse of the invalidated-claims scenario: all seven claims are live and available to be asserted against you. Your invalidity posture is strong on paper, because the patent has never been Board-tested, and the cited-art record is thin. Concrete steps:

1. Front-load the IPR decision. File within your § 315(b) window (one year from service of any complaint). No § 315(e)(2) estoppel exists against any earlier challenger, and no earlier PTAB record gives the patent owner a "the Board already blessed this" narrative.
2. Lead with the on-face references you already have. Studzinski (1995) for the stepped-throat sonic nozzle + gravimetric prover combination, and Meier (1978; AIAA 1980) for the abrupt-enlargement/oblique-shock physics. Then reach for ISO 9300:2005 and ASME MFC-7M — both named by the patent itself in the FIG. 4 discussion as the comparative nozzles — plus the parallel-sonic-nozzle-bank literature (Choi et al., Flow Meas. Instrum. 1997/1999) for the "measure one, multiply by the number open" arithmetic of claim 2.
3. Attack the numerical limitations as non-critical design choices. Claim 1's "about 1 throat diameter / about 10% / about throat diameter ÷ 2" recite approximate dimensions with, on the face of the specification, no comparative data across step geometries. FIG. 4 compares the stepped nozzle only against non-stepped nozzles. That supports a KSR / In re Applied Materials argument that the recited ranges are optimization of a result-effective variable.
4. Consider § 112(b) for indefiniteness/relative terms — the specification defines "about" as "±1%, ±2%, ±5%, ±10%, ±20%," a range so wide that "about 10% of the throat diameter" arguably spans 8%–12%, and the throat-diameter formula as rendered is malformed. Also consider written-description/enablement, since the specification asserts accuracy gains without criticality data.
5. Docket hygiene: run a certified search of PTAB E2E (https://ptab.uspto.gov), CourtListener/RECAP (https://www.courtlistener.com), and USPTO PatentCenter for application 15/942,653 to confirm the file wrapper contains no PTO-892/1449 items beyond those on the printed face, and to capture any pending ex parte reexamination (a reexam would not have appeared in the AIA-trial ODP feed).

If active proceedings were pending: none are — so there are no trial-stage milestones (no institution-decision deadline, no oral hearing, no 1-year FWD due date) to calendar.

If no PTAB activity exists — state it plainly: it does not. The absence is itself a signal, and it cuts in the defendant's favour here: well-asserted patents eventually attract IPRs, and this patent's clean PTAB record is most plausibly explained by the absence of any infringement assertion, not by the patent having survived an attack. Note also the procedural environment you would be filing into (2025–2026): the Office has centralized institution decisions in the Director, elevated discretionary denial, and applied a "settled expectations" factor that materially disfavours petitions against patents in force six or more years (this patent issued 2019-01-29 and is ~7.5 years old as of 2026-09-29). Factors that have overcome that presumption include non-commercialization — not available here, CPA sells hardware — and related-family circumstances. Plan for a real risk of discretionary denial notwithstanding strong merits, and consider a Sotera-style stipulation to blunt Fintiv.

---

Verification and confidence

• High confidence in the count: zero. The canonical structured block states the ODP API returns no AIA trial proceedings for this patent, and my independent web checks produced no contrary hit. Per the operating instruction in the "PTAB proceedings on file" block, the default is "no PTAB activity on file."
• Disclosed limitation: I do not have direct query access to PTAB E2E, PACER, or the CAFC docket in this session, and my searches returned no litigation or appeal involving US 10,190,900 (consistent with the prior sections). This is a good-faith negative from the ODP feed plus web sources, not a certified docket clearance.
• No fabrication: I have not assigned any proceeding number, judge panel, institution date, or FWD disposition, because no proceeding exists. The only real adjudicative record touching this patent owner is the unrelated TTAB cancellation, which I have labelled as such and excluded from the AIA-trial table.
• Verbatim-status note: the canonical status string from the structured data is, in substance, "no AIA trial proceedings on file." I have reproduced that as the status rather than paraphrasing it into a proceeding-level field.

I'll research the assignment record for US 10,190,900 before reconstructing the chain.

Let me verify the assignment record details and check for any subsequent transfers or security interests.

Let me try to pin down the assignment correspondent of record and check for any security interests or later transfers.

I could not run a direct query against the USPTO Assignment Center API in this session (the Assignment Center's search UI at assignmentcenter.uspto.gov is not retrievable as structured text here, and my query for the reel/frame string returned no indexed hits). Accordingly, the reel/frame and event dates below are taken from the authoritative full text of US 10,190,900 supplied to me — specifically its "Legal Events" record (the AS / Assignment entry). Where a field is not in that record, I say so rather than guess. Treat the Correspondent field as unretrieved, not as absent.

---

Inventors

Unusual-pattern check — no red flag. The tell the prompt asks about (all inventors leaving the original assignee within 12 months of filing, portending a fire-sale) is not present. The opposite is true: all three inventors assigned to the company three weeks after filing (filed 2018-04-02; assignment executed 2018-04-27) and the inventor-principal (Daniel Sawchuk) remains CEO of the assignee as of company records current to the 2020s. There is no inventor-separation event, no subsequent inventor-to-third-party conveyance, and no record of any inventor re-acquiring or later re-transferring the asset.

---

Original assignee

Entity named on the issued patent: Canada Pipeline Accessories Company Ltd. (US 10,190,900 face / assignment record). Note the naming variant: Google Patents lists the "Current Assignee" as Canada Pipeline Accessories Co Ltd, while the original assignee and the assignment record read Canada Pipeline Accessories Company Ltd. I read these literally and do not treat them as the same legal name — but I found no recorded change-of-name assignment bridging them. The most likely explanation is assignee-name normalization in the aggregator database rather than a recorded corporate act. The company also appears in filings as "Canada Pipeline Accessories, Co. Ltd." The "CPACL" form is documented in the RJ Machine record as CPA's earlier self-designation.

• Primary line of business (operating company, not a holding vehicle): Calgary, Alberta manufacturer/supplier of in-line pipeline flow-measurement hardware — perforated-plate flow conditioners (CPA 50E, 55E, 65E, TBR, TBRL), static mixers, measurement rings, and flow nozzles / venturi nozzles, including restriction and sonic nozzles, plus metering consultation and CFD services. Founded ~1997; small entity per the USPTO fee-status event (ENTITY STATUS SET TO SMALL, 2018-04-27); headcount on the order of 7–10; estimated revenue ~$6.0M (third-party estimate, treat as unverified).
• Does it ship a product embodying the claims? Partially / unclear. CPA's own product literature confirms it manufactures "restriction/sonic nozzles" custom-sized per application, with the throat bore sized to a target pipe velocity "for the meter that is being protected." That is squarely in the field of claim 1, and it is the closest thing to a commercial embodiment of the '900 claims. However, CPA states it does not stock nozzles ("each one is custom designed for the specific application"), and I found no product literature or marketing that describes a multi-nozzle "nozzle bank" used as a flow partitioner with a downstream mass-flow device per claims 1–7. So: the assignee is a genuine operating manufacturer in the relevant technology, but a commercial embodiment of the claimed nozzle-bank measurement method is not documented. I do not assert one.
• Current status: Operating. No bankruptcy, no dissolution, no acquisition found. Corroborated by (a) the patent's own maintenance-fee events (4th year paid 2022-01-31; 8th year paid 2026-01-29, i.e., timely, current); (b) continued US trademark filings by the same entity through 2024 (e.g., serial 98737366 "450S," filed 2024-09-06); (c) current product literature dated 2024.

Assertion posture (context, not evidence of an NPE): CPA is an assertive operating company — but its documented enforcement has been over trademarks/trade dress and the expired "Laws Patent" (US 5,341,848), not over US 10,190,900. In RJ Machine Co. v. Canada Pipeline Accessories Co., Nos. 1:13-cv-00579 and 1:15-cv-00505 (W.D. Tex.), CPA was accused of attempted monopolization based on its 50E trade dress/trademark enforcement and its license to the Laws Patent — i.e., CPA appears in that record as a licensee-enforcer, which is an unusual posture and worth flagging: it means CPA's patent-assertion history runs partly through in-licensed rights, not solely its own portfolio. That fact pattern is not an NPE signal, but it is a genuine oddity in the ownership picture. No suit, IPR, or ITC action on the '900 patent was found in the previously generated sections of this analysis.

---

Assignment timeline

One (1) recorded assignment exists for US 10,190,900. There are no security agreements, no mergers, no change-of-name filings, no licenses recorded against the patent, and no post-issuance transfers.

• 2018-04-27 (executed) / recorded 2018-04-27 — Reel 045653/0261
  • Conveyance: Assignment (per USPTO event code AS; full text: "ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:SAWCHUK, BLAINE;SAWCHUK, DALE;SAWCHUK, DANIEL;REEL/FRAME:045653/0261")
  • Assignor: Sawchuk, Blaine; Sawchuk, Dale; Sawchuk, Daniel (all three joint inventors)
  • Assignee: Canada Pipeline Accessories Company Ltd., Canada
  • Correspondent: NOT RETRIEVED. The patent's legal-events record carries only the reel/frame and the assignor/assignee names; it does not reproduce the correspondent-of-record field, and I could not execute a live Assignment Center query in this session. Do not treat the correspondent as blank in the underlying record. The highest-probability correspondent is Frederick N. Samuels, Cahn & Samuels, LLP, 1100 17th St., NW, Suite 401, Washington, DC 20036 — that firm is the recurring US prosecution and trademark correspondent for the entire CPA portfolio (e.g., attorney-of-record firm listed on CPA's US 10,677,632; legal correspondent on CPA trademark serials 98737366, 88361285, 97816058; 279 filings recorded under that correspondent). I flag this as probable, not confirmed, for the assignment recording itself. Recurrence note: Cahn & Samuels is a repeat player in this chain by definition (it handles the whole portfolio), but it is an operating-company firm here — it is not a name that appears on Unified Patents / RPX / Patent Progress NPE-assertion lists in my searches, so recurrence is not an NPE tell in this chain.
  • Context: Routine inventor-to-company assignment for a corporate-filed application — the three individual inventors conveying title to their employer three weeks after the 2018-04-02 filing. Not a fire-sale, not a securitization, not a transfer to an asserter. This is the only conveyance in the chain.
  • Effective date of record: 2018-04-27 (same date as execution and recording).

If the Assignment Center shows no records for this patent — it does not; there is exactly one. The finding is therefore: one link, and it terminates at the original operating-company assignee, which still owns the patent.

Also on the record (fee events, not assignments — do not conflate): 2018-04-02 entity status undiscounted→(?); 2018-04-27 entity status set to SMALL; 2019-01-09 PATENTED CASE; 2022-01-31 MAFP 4th year; 2026-01-29 MAFP 8th year. A maintenance-fee payment is not an assignment and should not be read as one.

---

Timeline diagram

timeline
    title Ownership of US 10190900
    2017 : Provisional filed by three Sawchuk inventors
    2018 : Non-provisional filed 2018-04-02
         : Assignment to Canada Pipeline Accessories Company Ltd
         : Reel 045653 frame 0261
    2019 : Patent issued 2019-01-29
    2022 : Maintenance fee paid year 4
    2026 : Maintenance fee paid year 8

---

NPE / troll-pattern signals

---

Verdict

Insufficient data — and that call is driven by the rubric's own definition ("no records, or only the original assignment"), not by any ambiguity about who owns the patent. The record contains exactly one conveyance: Reel 045653/0261, executed and recorded 2018-04-27, moving title from inventors Blaine, Dale, and Daniel Sawchuk to Canada Pipeline Accessories Company Ltd., and nothing since. Substantively, that is the opposite of an NPE chain: zero of the eight signals are present, the assignee is a Calgary operating manufacturer of flow-measurement hardware with real products in the relevant field (including custom restriction/sonic nozzles), the small-entity 8th-year maintenance fee was paid 2026-01-29, and the only litigation touching CPA concerns its 50E trademark/trade dress and the expired Laws Patent — never US 10,190,900.

Because the rubric has no "operating company, unasserted" bucket, I designate Insufficient data rather than force this into Operating-company assertion (which requires the assignee to be suing actual competitors on this patent — it is not) or Defensive / non-asserting (which requires a defensive-aggregator termination — there is none). If your scoring convention treats "original assignee still owns, no transfers" as the inverse-NPE case, then the correct substantive label is operating-company ownership, non-asserted, no NPE indicia.

Verification links:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent number 10190900; alternative index: https://assignment.uspto.gov/patent/index.html)
• Do one thing I could not do: open Reel 045653/0261 in Assignment Center and read the Correspondent and Execution Date fields directly. Those two fields are the only material gaps in this reconstruction, and the correspondent is the field the prompt (correctly) cares most about.

Gaps and confidence: High confidence in the number of assignments (one) and in the assignor/assignee identities/dates, because that comes from the authoritative legal-events text. Lower confidence in the correspondent-of-record (unretrieved; Cahn & Samuels offered as a probable, flagged inference), in the employer attribution for Blaine and Dale Sawchuk (inferential from the assignment plus the RJ Machine/CEO record), and in the "does CPA ship an embodiment of claims 1–7" question (it ships sonic/restriction nozzles, which is relevant-field but not a documented nozzle-bank product). I did not access SEC filings, and I found no evidence CPA is public or SEC-reporting, so there is no 10-K/8-K trail to cross-reference — the private-company status is why the paper trail here is one link long.

I'll verify the patent and each cited reference against live sources before mapping them to § 102.

Now let me verify the two remaining patent citations and the USPTO record for the patent itself.

Verification note (USPTO record for 10,190,900)

I could not open patentcenter.uspto.gov directly in this session, so the "USPTO database" result I can stand behind is the mirrored record at patents.google.com/patent/US10190900 and the Justia copy at patents.justia.com/patent/10190900 — both of which reproduce the same application number (15/942,653), priority (62/481,871, 2017-04-05), grant date (2019-01-29), and the same five-document citation list you supplied. The citation list is the controlling input for this task, and I searched on the exact identifiers rather than on look-alike numbers (e.g., I did not let results for US 10,451,457, US 9,605,695, US 9,625,293, or US 9,506,484 masquerade as the '900 patent — those are the granted parents/children of the cited publications, and are treated separately below).

This builds on the earlier claim overview and obviousness sections. No contradiction found between those sections and the live data retrieved here; the findings below reinforce the earlier conclusion that the five patent citations are weak art and the two NPL references carry the real § 102/§ 103 weight.

---

The § 102 legal frame (why the answer below is "none anticipate")

The '900 patent has an effective filing date of 2017-04-05 and is a post-AIA patent. Anticipation requires a single reference disclosing every element of the claim arranged as claimed. Claims 1–7 (reproduced in the earlier section) reduce to: (i) a nozzle bank of a plurality of stepped nozzles with a predetermined number open; (ii) mass-flow measurement of one partitioned stream; (iii) scaling to total flow; and (iv) the specific step geometry (single step ~1 throat diameter from the inlet plane, ~10% diameter increase, length ~throat diameter ÷ 2).

Every cited document must therefore be tested against that conjunction — not merely against "flow measurement" or "a nozzle."

---

The five patent citations — full entry-by-entry treatment

All five are available as prior art under § 102(a)(1)/(a)(2) (each published or issued before 2017-04-05), so availability is not the problem. The problem is content: none of them discloses a stepped sonic-nozzle bank used as a flow partitioner, which is the whole of claim 1.

1. US 7,823,462 B2 — Turbulence conditioner for transit time ultrasonic flow meters and method

• Dates: filed 2007-12-14 (Ser. No. 12/002,270); granted 2010-11-02. Publication US 2009/0151472 A1.
• Disclosure (verified): An apparatus for determining fluid flow in a pipe comprising a transit-time ultrasonic flowmeter plus a turbulence conditioner (honeycomb, "egg-crate," or "mini-tube" perforated plate) disposed in the pipe, where the pitch between holes is a function of the effective diameter of the meter's transducer apertures. The conditioner reduces eddy sizes so turbulence is averaged within the acoustic beams, allowing proving within ±0.027% in five runs. One embodiment adds a reducing nozzle downstream of the conditioner and upstream of the meter, plus an expanding nozzle to restore line diameter.
• § 102 mapping: Anticipates nothing in claims 1–7. The device is a flow conditioner to make an ultrasonic meter's measurement repeatable — not a bank of choked, stepped nozzles; there is no partitioning of flow among nozzles, no sonic/choked throat, no mass-flow device, and no "number of open nozzles × measured flow" arithmetic. The "reducing nozzle" it mentions is a smooth contraction, the antithesis of the claimed stepped throat. Cited only for the generic pipeline-measurement-component context noted in the earlier section.

2. US 2016/0054158 A1 — Water deluge system testing apparatus and method

• Dates: PCT filed 2013-03-21; published 2016-02-25; granted as US 10,451,457 B2 (2019-10-22). Applicant Paradigm Flow Services Ltd.; inventors Mackenzie, Thomson, Macaulay.
• Disclosure (verified): A method of testing a water deluge (fire-suppression) system by supplying gas instead of water: a gas source (accumulator/air cylinders) is coupled to the deluge pipework, sensors are installed at a subset of the deluge nozzles, fluid parameters (flow rate/pressure) are measured over a short "gas test period" (≈5–120 s), and the data set is compared to a baseline. The nozzle sensor in the embodiment contains a turbine flow meter and a pressure gauge; the system may also mention a flow straightener at the test station.
• § 102 mapping: Anticipates nothing in claims 1–7. This is a fire-protection compliance-test method. The "nozzles" are deluge sprinkler outlets, not sonic flow nozzles; there is no choked-throat step, no nozzle bank partitioning a pipeline flow, no single-step-at-1-throat-diameter geometry, and no scaling of one stream to total pipeline flow. Its only § 102-adjacent trait — multiple nozzles with flow sensed at a subset — does not map onto the claimed element because claim 1 requires the stepped nozzle geometry in the bank that is doing the partitioning. As the earlier section concluded, this is the weakest of the five and contributes essentially nothing.

3. US 2016/0097411 A1 — Flow conditioner and method of designing same

• Dates: priority US 61/825,559 (2013-05-21); PCT/CA2014/050416 filed 2014-05-01; published 2016-04-07; granted as US 9,605,695 B2 (2017-03-28). Applicant Canada Pipeline Accessories; inventors Daniel Sawchuk, Reginald Selirio.
• Disclosure (verified): A single-disk perforated-plate flow conditioner having a central hole, an outer ring of holes, and at least one inner ring "recessed or stepped" from the outer ring, with hole diameters/lengths graded so that fluid passages have unequal length; plus a design method using the swirl formula D = C·tan⁻¹(V_T/V_x) and a pipe assembly with the conditioner disposed perpendicular to the pipe axis with a flow meter downstream.
• § 102 mapping: Anticipates nothing in claims 1–7 — but note the trap. The word "stepped" appears repeatedly, and the "pipe assembly for flow measurement … perpendicular to an axis of the pipe … and a flow meter" language superficially resembles claim 1's setting. It is not anticipation: the "steps" here are recessed concentric rings of holes in a conditioner disk, not a single step in a nozzle throat; there is no sonic/choked flow, no nozzle bank partitioning flow into measurable sub-streams, and no total-flow calculation. Counsel reading the '900 patent should expect an examiner or defendant to quote these "stepped" passages — the distinction to preserve is conditioner-plate steps (profile conditioning) vs. nozzle-throat steps (oblique-shock boundary-layer closure). This one is also a same-family CPA document (see the earlier assignment section), which is worth knowing for § 102(b) statutory-bar and self-collision analysis.

4. US 2016/0334249 A1 — Flow Conditioner Having Integral Pressure Tap

• Dates: priority/filing 2015-05-14; published 2016-11-17; granted as US 9,625,293 B2 (2017-04-18). Inventor/applicant Daniel Sawchuk.
• Disclosure (partial — flagged): I confirmed the bibliographic record and the grant (US 9,625,293 B2, 2015-05-14 → 2017-04-18) in the live results, and its subject per the citation list of the '900 patent is a flow conditioner with an integral pressure tap. I did not retrieve its full claim text in this session (search limit reached), so my description of its disclosure is limited to the title/record and the family context — I am stating that limitation rather than reconstructing claims I did not read.
• § 102 mapping: On the record available, anticipates nothing in claims 1–7. A pressure tap integrated into a flow conditioner is a conditioner + sensing-port feature; it supplies none of the claimed method steps (partitioning with a stepped nozzle bank; measuring a partitioned stream with a mass-flow device; computing total flow) and none of the step geometry. Its § 102 usefulness is contextual only — evidence that integrating pressure taps into flow-modifying pipeline hardware was routine (a point already made in the earlier obviousness section).

5. US 9,506,484 B2 — Flow conditioner and method for optimization

• Dates: appl. US 13/896,947 filed 2013-05-17; published as US 2014/0338771 A1; granted 2016-11-29. Assignee Cameron International; inventors Gregor Brown, Herbert Estrada, Donald Augenstein (per the retrieved record); counterpart CA 2912197 A1 / WO 2014/186377 A2.
• Disclosure (verified): A flow conditioner plate for a circular pipe with a central circular passage surrounded by two or more concentric arrays of segmented annular passages defined by separating and subdividing ligaments of varied widths, giving a graded resistance; plus a computerized optimization method (store target pressure-loss coefficient and velocity-profile goals, set ring/subdivision widths and plate thickness based on passage-length/hydraulic-diameter ratio, iterate). Verified data: average loss coefficient ≈0.91 vs ≈2 for the Nova/CPA 50E Laws variant; the document also expressly discusses Laws (CPA 50E), Spearman, Gallagher and Zanker conditioners.
• § 102 mapping: Anticipates nothing in claims 1–7. This is a velocity-profile conditioner and a design-automation method. No sonic/choked nozzle, no throat step, no nozzle bank, no mass-flow device, no flow-scaling step. Its substantive relevance is background/§ 103: it shows the art's preoccupation with pressure-loss optimization of conditioners — which the earlier section already used to frame the motivation analysis.

---

Bottom line on the patent citations

Not one of the five cited US patent documents anticipates any of claims 1–7. Ranked by technical proximity to the '900 claims:

1. US 2016/0097411 A1 — closest, because of the shared "stepped / pipe assembly / perpendicular to the axis / flow meter" vocabulary and the shared CPA authorship; still not anticipatory (conditioner steps ≠ nozzle-throat step).
2. US 7,823,462 B2 — same field (pipeline flow measurement hardware) but a turbulence conditioner for an ultrasonic meter.
3. US 9,506,484 B2 — same field, conditioner optimization.
4. US 2016/0334249 A1 — same family neighborhood; a tapped conditioner.
5. US 2016/0054158 A1 — different field entirely (fire deluge testing); the weakest.

The genuinely § 102-relevant art remains the non-patent citations of record, exactly as the earlier section flagged:

• Meier et al., "Oscillations of the Supersonic Flow Downstream of an Abrupt Increase in Duct Cross-section," Mitt. Max-Planck-Institut Nr. 65, Göttingen 1978, pp. 7–172; and Meier, AIAA Journal, Apr. 1980, 18:394–395 — the oblique-shock / abrupt-enlargement physics that the '900 specification relies on ("an oblique shock wave in the throat").
• Studzinski et al., "Nova's Gravimetric Meter Prover and Sonic Nozzle Facility," International Fluid Flow Symposium, San Antonio, TX, 1995-03-20/22 — cited on the face of the patent, and (per the companion NOVA inter-lab literature) describing a sonic-nozzle facility with a gravimetric prover and a nozzle having "an intentional step change in the diameter at the throat." This is the single reference that comes closest to the § 102/§ 103 boundary for claim 1, because it supplies both the stepped-throat nozzle and the gravimetric measurement in a multi-nozzle facility.

---

Flag — relevant art not of record (clearly labeled as such)

My searches surfaced a document not listed on the '900 patent face that is materially closer to claim 2's arithmetic than any cited patent: US 2002/0078734 A1, which describes a critical-flow-nozzle system with nozzles 100, 102 whose individual flows are computed from inlet absolute pressure and temperature and where "the total mass flow through the critical flow nozzles 100, 102 is the sum of the mass flows calculated for each of the critical flow nozzles through which air is flowing, i.e. the nozzles whose respective valves 118, 120 are open." That is, in substance, claim 2's "number of open nozzles × measured flow." I have not verified its assignee, exact filing/publication dates, or whether it was before the examiner in this case, so I flag it as a candidate for a further § 102/§ 103 search, not as an established ground.

---

Confidence and caveats

• High confidence: the five cited patent documents are flow conditioners, a turbulence conditioner, and a deluge-test apparatus; none discloses a stepped sonic-nozzle bank with choked-flow partitioning plus mass-flow scaling, so none anticipates claims 1–7. The descriptions of US 7,823,462, US 2016/0097411, US 9,506,484 and US 2016/0054158 above are grounded in the retrieved full texts/records cited inline.
• Explicit uncertainty: I did not retrieve the full claim text of US 2016/0334249 A1 / US 9,625,293 B2 before the tool limit; its description here rests on the bibliographic record and title, and should be confirmed against the granted claims before being relied on. Likewise, I could not run a direct USPTO PatentCenter query, so the "USPTO database" leg rests on the mirrored Google Patents/Justia record.
• Reporting convention observed: identifiers are reproduced literally, including the specification's own "Canada Pipieline Accessories (CPA) no step" typo, which is not corrected.

I'll ground a few technical points with live searches before drafting the § 103 analysis (building on the prior-art inventory already generated).

Preliminary verification and scope notes

• Date discrepancy (flagged, not corrected): this task states "Current Date: April 26, 2026." The session header reads 2026-09-29, and the earlier sections of this file used a Google Patents fetch dated 2026-09-24. I use the references as of the patent's 2017-04-05 priority date and note the discrepancy rather than resolving it.
• Builds on, does not repeat: the citation inventory (P1–P5, N1–N3), the claim-1 element template (A–D), and the "no anticipation" conclusion are taken as given from the Prior Art section. This section picks up at § 103.
• Important new material: this analysis identifies several references not on the face of the patent. They are flagged as such throughout, and their verification status is disclosed at the end.

---

1. Governing law and the § 103 framework applied here

AIA 35 U.S.C. § 103 governs (filing 2018-04-02; priority 2017-04-05). The Graham v. John Deere inquiries — scope/content of the prior art, differences from the claims, PHOSITA level, and objective indicia — control, applied through the flexible KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) approach. Cornell LII case page; USPTO 2007 KSR Examination Guidelines, 72 Fed. Reg. 57,526, updated 2024 (Finnegan summary of the Feb. 27, 2024 update).

The MPEP 2143 rationales relied on below are: (A) combining known elements by known methods for predictable results; (C) using a known technique to improve similar devices in the same way; (D) applying a known technique to a known device ready for improvement; (E) "obvious to try" among a finite number of identified, predictable solutions; and (F) known work in one field prompting predictable variation driven by design incentives. KSR supplies the operative language: "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious unless its actual application is beyond his or her skill."

PHOSITA definition (proposed): a person with a B.S. in mechanical or aerospace engineering (or equivalent) and 3–5 years' experience in gas flow measurement, custody transfer, or critical-flow-nozzle (CFN) metrology — i.e., someone fluent in ISO 9300 / ASME 7M practice, discharge-coefficient (C_d) correction for boundary layer, and mass-flow-meter (including Coriolis) selection. This is a predictable, mature, well-standardized mechanical metrology art, which weighs toward obviousness. Nothing in the '900 specification suggests the art was unpredictable.

---

2. A threshold construction issue that shapes every § 103 theory

Under Phillips, the numeric limitations are read against the specification's own express definition:

"'substantially', 'relatively', 'generally', 'about', and 'approximately' are relative modifiers intended to indicate permissible variation (e.g., ±1%, ±2%, ±5%, ±10%, ±20%) …"

That self-defined tolerance means "about 1 throat diameter" plausibly reaches 0.8–1.2 θ; "about 10%" reaches 8–12%; "about θ/2" reaches 0.4–0.6 θ. Two consequences:

1. The claim is broad enough that prior-art step geometries need not match the numbers exactly.
2. Simultaneously, the "±20%" definition is a § 112 indefiniteness vulnerability and a § 103 vulnerability — a claim range that diffuse is hard to defend as a "critical" or "result-effective" range under In re Aller / In re Woodruff.

---

3. References newly identified beyond the citation record

These are not among P1–P5/N1–N3 and are flagged as additions:

Two further "similar documents" flagged in the Prior Art section remain unread and are not relied on: US 4,753,114 ("Critical flow detection"), US 4,027,523 ("Methods and apparatus for proof testing gas meters"), US 3,686,946, JP S61-266922 A.

The single most damaging item is R1. Chevron '492 is the closest thing to a method-level preview of claim 1: partition a pipeline flow with a critical-flow nozzle, measure only the smaller partition on a low-pressure device, and compute the total from that measurement. That is elements (A)+(B)+(C) of claim 1, plus the meter prover of claim 5 — and claim 1's only remaining novelty is the step geometry (element D).

---

4. Combination theories

Combination 1 — R1 (Chevron '492) + R2 (AIAA 2015-2311) [+ R4 as the baseline] — primary theory

Motivation (KSR rationales C and D). The known problem in the field is exactly what the '900 specification admits: boundary-layer (C_d) sensitivity causes mass-flow drift with pressure ratio. R4 (ISO 9300 / ASME 7M) embodies that problem and is the owner's own admission of the deficiency as of 2017. R2 supplies the known technique — a recessed step at the throat — that produces a local shock/boundary-layer interaction. Applying a known technique (throat step) to a known device ready for improvement (a smooth-throat CFN used as the partitioner in R1) is KSR rationale (D).

Reasonable expectation of success. R2 reports the step produces only a local disturbance and that the pressure signal is stable; R3 reports the shock "did not propagate to the throat, so that the nozzle remained choked at all times." A PHOSITA reading R2/R3 would expect the stepped nozzle to remain choked and stable — i.e., the very property the '900 patent asserts. There is no teaching away; the art affirmatively predicted the claimed outcome.

Combination 2 — R1 + R3 (+R4) — same theory, different shock-at-step evidence

R3 is the stronger shock-physics reference: it identifies an oblique shock at the step and confirms sustained choking. Combined with R1's partitioning/measurement arithmetic, this covers (A)–(D). R3's "stepped CFVN" also shows the step concept was already being applied to critical flow venturi nozzles — not merely to rocket-duct expansions as in Meier (N1/N2), which is the gap in the record art.

Combination 3 — R1 + N1/N2 (Meier) + R4 (ISO 9300 / ASME 7M)

This is the § 325(d)-aware variant built only on references already considered by the examiner. Meier supplies the abrupt-expansion oblique-shock/boundary-layer teaching; ISO 9300/ASME 7M supplies the smooth-throat sonic nozzle with a cylindrical throat of ~1θ axial length; R1 (Chevron — not of record) supplies the partitioning/measurement/totaling method and the prover. This theory's weakness is that Meier is about a duct, not a nozzle throat, so the "to a nozzzle throat" leap is argued rather than shown — which is precisely why R2/R3 are the better primary pinning references.

Combination 4 — P3/P4 (CPA's own flow-conditioner family) as secondary evidence

P3 (US 2016/0097411 → US 9,605,695) and P4 (US 2016/0334249 → US 9,625,293) are the applicant's own publications and teach (i) a "stepped configuration" fluid element mounted perpendicular to the longitudinal axis, (ii) a flow meter downstream, and (iii) calculation of flow across the element from pressures. Their relevance is not element D (their "step" is a stepped plate, as the Prior Art section correctly notes) but motivation: they show the same inventive entity was, contemporaneously, mounting stepped flow elements perpendicular to a pipe with a meter downstream and computing flow — i.e., the "arrangement of old elements" was squarely within the applicant's own ordinary practice. KSR's "common sense / familiar items" reasoning applies.

---

5. Dependent claims — § 103

---

6. Motivation-to-combine — consolidated

A PHOSITA would have been motivated by at least five independent, articulated reasons, none of which requires hindsight:

1. Known problem, admitted in the patent itself. The '900 specification states the existing nozzles "all indicate a sensitivity in mass stability (mass flow rate) to pressure ratio." That is an express admission that a long-felt, recognized deficiency existed in the nozzle bank art as of 2017 — the first Graham/KSR hook ("any need or problem known in the field … can provide a reason for combining").
2. Design incentive / market pressure (KSR rationale F). Custody-transfer gas metering is a high-value, high-accuracy field; R1's own background notes "errors of only a few tenths of a percent in the meters can result in millions of dollars." The incentive to eliminate C_d sensitivity is economic and manifest.
3. Finite, identified, predictable solutions (KSR rationale E). The designer's options to defeat throat boundary-layer flow were few and known: alter the inlet curvature, alter the throat axial length (Ishibashi's work on cylindrical-part length), or introduce a discontinuity/step. Choosing the step among a small set of known options, with R2/R3 reporting predictable results, is KSR "obvious to try."
4. Same technique, same field, same function (rationales A and C). R2 and R3 apply the throat-step technique to critical flow nozzles themselves — not to a remote art. The step performs the same function (boundary-layer/shock control at a supersonic throat) it was known to perform; the combination yields nothing more than the expected result.
5. Adaptation of a known partition-and-scale method. R1 already taught partitioning with a sonic nozzle and scaling a measured small stream to a total. Substituting a stepped nozzle for the smooth nozzle inside R1's existing architecture is a substitution of one known element (a CFN) for another known element (a stepped CFN) with predictable results.

---

7. Objective indicia (secondary considerations)

On the record available to me, no evidence of secondary considerations exists:

• No litigation asserting the '900 patent and no PTAB proceeding (per the earlier sections), so there is no commercial-success or copying record tied to the claims and no nexus evidence.
• No unexpected-results evidence in the specification. The '900 specification asserts accuracy as a goal ("extremely accurate, accounted for, and stable") but supplies no comparative data table, no error budget, and no statistical showing that the stepped nozzle outperforms R4/Meier beyond the FIG. 4 plot — and FIG. 4 is in significant tension with R2's finding that the step produced "no significant differences" from the clean configuration at the observable frequencies. A patentee relying on FIG. 4 will have to establish that the plotted advantage is attributable to the claimed geometry rather than to test conditions.
• No long-felt-need rebuttal is available where the need was known and the solution was among a handful of known options.
• Possible nexus point for the patentee: CPA is a practicing manufacturer (per the Assignment section) selling nozzle products; if a competitor's product were shown to copy the stepped-throat design, that would be KSR-style "copying" evidence — but none is on the record, and copying alone is weak absent nexus.

---

8. Counterarguments, weaknesses, and candor

1. Hindsight is the central risk in Combination 1. No single reference discloses the triple of numbers (1θ / +10% / θ2). The § 103 case therefore rides on "optimization of result-effective variables" (In re Aller) plus the specification's own ±1–20% "about" tolerance. A patentee will argue: (i) the prior art does not identify which variables are critical or which of many step positions/sizes is likely to succeed; (ii) the values are not shown to be result-effective by the art; and (iii) the combination reconstructs the claim only by selecting values from the claim itself — the Cyclobenzaprine/O'Farrell "no direction as to which of many choices" objection.
2. Meier (N1/N2) is about oscillations — a possible teaching-away. The 1978/1980 work concerns oscillatory supersonic flow downstream of an abrupt expansion. A patentee could argue the art taught that a step destabilizes, not stabilizes. R2 blunts this: it reports stable, unchoked-disturbance-free behavior with a recessed throat step. If R2 is properly authenticated, the teaching-away argument fails.
3. The step may be geometry-neutral. R2 states that at the observable frequencies there were "no significant differences between the recessed step and the clean configuration," and that in choked conditions the step "does not influence significantly the flow in the divergent section." Read literally, R2 could be argued either way: as establishing predictable, non-adverse behavior (good for a challenger) or as undermining the patent's asserted mechanism of "closing off the unaccounted for boundary-layer flow" (also good for a challenger, via enablement/utility).
4. § 325(d) exposure. Combination 3 rests on N1/N2 (Meier) and R4 items discussed on the face of the patent, and the examiner allowed over them. Any petition built on considered art must show the Office "materially erred" or that the new theory is not cumulative. This is the strongest reason to lead with R1 and R2/R3 — none of which appears to be of record — and to use Meier only to explain the physics.
5. The prior allowance is not dispositive but is evidence. The claims issued over Meier, Studzinski and the CPA/Cameron conditioner art. That does not defeat the R1+R2 theory (different references, different theory of the invention), but it does mean a challenger must articulate why the new combination was not before the examiner.
6. "About" cuts both ways. If the patentee narrows construction to preserve validity (e.g., "about 10%" = 10% exactly, "about 1 throat diameter" = 1.0θ), the prior-art overlap shrinks — but then the claim is exposed to § 112 written-description/enablement attack, because the specification discloses no data supporting a critical range and no algorithm for selecting within it.

---

9. Bottom line

The '900 claims are, in my judgment, likely invalid under § 103 as obvious, with Combination 1 or 2 (Chevron US 4,706,492 + AIAA 2015-2311 and/or von Lavante & Kaya 2015, over the ISO 9300/ASME 7M baseline) as the strongest theory, and Combination 3 (Chevron + Meier + ISO 9300/ASME 7M) as the § 325(d)-conscious fallback. Claim 1's method steps (A)–(C) are essentially disclosed by Chevron '492, and element (D) — the only true gap in the record art — is supplied by the 2015 stepped/recessed-throat-nozzle literature, with a strongly articulated, non-hindsight motivation (§ 1 above) and a documented, positive expectation of success (the nozzle remains choked; the disturbance is local and stable). Claims 2–7 fall to routine arithmetic, parallel-nozzle bank practice, and the express "meter prover" teaching of Chevron '492 claim 3 and Studzinski (N3). The principal obstacles are the hindsight/result-effective-variable attack on the numeric geometry, any teaching-away reading of Meier, and § 325(d).

---

10. Verification status and confidence (do not gloss over)

• Verified this session from live sources: the content of US 4,706,492 / US 4,753,095 (Chevron) — abstract, description, and the full claim set including the express "meter prover" claim, retrieved from Google Patents, FreePatentsOnline and OSTI; the ISO 9300 discharge-coefficient regime and the "cylindrical throat length typically of one throat diameter d" statement (from a peer-reviewed Flow, Turbulence and Combustion article on cylindrical CFVNs and the IMEKO TC9-2019 paper); the von Lavante & Kaya 2015 "stepped CFVN" passage and the oblique-shock/remains-choked finding; the KSR / MPEP 2143 rationale framework.
• Partially verified — treat with caution. R2 (AIAA 2015-2311) was retrieved only as a partial PDF snippet (a DOI path 10.2514/6.2015-2311 and extracted figure text), not as a complete document; I could not confirm its authors, full pagination, or official publication date. Its content as quoted (recessed step at throat, oblique shock away from the step, stable choked behavior) is reported as-is. Do not cite R2 in a filing without pulling the complete paper. Bibliographic details of R3's internal reference list also appear garbled in the retrieved snippet and should be re-sourced.
• Not verified / not relied upon: US 4,753,114; US 4,027,523; US 3,686,946; JP S61-266922 A. These remain "worth a look" only.
• This is an obviousness analysis, not a validity opinion or a freedom-to-operate conclusion. It assumes the references are prior art under § 102(a)(1)/(a)(2) (all predate 2017-04-05) and assumes the R2/R3 documents are authentic printed publications. Any actual challenge should re-verify each reference's date, public accessibility, and completeness, and should confirm whether R1/R2/R3 were before the examiner in application 15/942,653 via the USPTO file wrapper — a step I could not perform here.

Confidence: high that the claims are vulnerable to a well-articulated § 103 case; moderate that any single combination would succeed on the numeric limitations without expert declaration evidence mapping the step dimensions to the prior art's result-effective variables; low that the patent would survive a properly supported petition or summary-judgment motion built on Combination 1 or 2.