US 2926027

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

Let me refine the searches for the US patent specifically and any litigation.

Let me run a few more targeted searches to confirm litigation status and any 2026 CAFC activity.

Summary of US Patent 2,926,027

Scope note on the search

The literal identifier 2926027 produces hits on several unrelated records, which I am deliberately not merging, per the strict-ID rule:

Litigation / CAFC: I found no litigation, PTAB, or Federal Circuit docket activity for US 2,926,027 in any search. I cannot confirm the absence definitively — my searches of "CAFC 2026 dockets" returned only unrelated patent cases (e.g., Greenthread IPRs, Siemens/ADC '027 patent litigation, which concerns U.S. 8,187,027 — a completely different patent whose number ends in "027" but is not the same identifier). Treat "no litigation found" as absence of evidence, not proof of no litigation. It is also consistent with the record: this patent expired for failure to pay maintenance fees / term on 1977-02-23, roughly 17 years after issue.

Bibliographic data (from the patent record)

• Title: Two-part union having left and right hand threads
• Patent number: US 2,926,027 A
• Inventor: Riley F. Marquis, Jr. (rendered "Riley F. Marquis, In" in the OCR'd cover page), of Los Angeles, California
• Original assignee / current assignee: Individual (no corporate assignee; Google Patents lists the assignee as "Individual" and notes the assignee listing "may be inaccurate")
• Application number: US 683,753 (also given as US683753A)
• Filing date: September 13, 1957 (application filed 1957-09-13)
  • ⚠️ Discrepancy noted literally: the OCR'd cover-page line reads "Filed Sept. 15, 1957," while the specification names "Application September 13, 1957," and the bibliographic filing date is 1957-09-13. I am reporting both as they appear rather than auto-correcting.
• Issue / publication date: February 23, 1960
• Legal status: Expired – Lifetime; anticipated expiration 1977-02-23
• Claims: 3 (1 independent, 2 dependent) — classified by the patent office as Class 285-32
• Classification: F16L19/00 (joints where sealing surfaces are pressed together by a member screwed on/into a joint part)
• Abstract: ⚠️ None. This is a 1957 filing; the printed patent has no abstract section. The nearest equivalent is the "object of the invention" passage (see below), not a formal abstract.

Abstract-like summary (paraphrased from the specification)

A plumbing pipe union made of only two hollow members instead of the conventional three. The first member (18) carries an internal annular flange (20) with two elastomeric sealing rings (21, 22) on opposite sides, and is internally threaded at one end to screw onto a standard tapered pipe-threaded nipple. Its other end carries external, untapered, left-hand threads (29) running all the way to its end face. The second member (19) has a shoulder (33), internal untapered left-hand threads (28) running to that shoulder engaging threads 29, and internal right-hand threads (25) of only ~2–3 turns that engage the second nipple. Because the two thread sets are of opposite hand/pitch, turning the union produces cumulative axial movement, drawing the second nipple against the outer sealing ring and seating both nipple ends against the rings — sealing independently of the pipe threads, requiring no pipe alteration, and with minimal axial length. Non-circular wrench flats and a wall-spacer flange allow tightening against a wall.

Plain-language overview of the claims

Claim 1 — the independent claim. A two-part union for joining two pipe sections that have standard tapered pipe threads, comprising:

1. First member: hollow, open at both ends; has an internal flange with a sealing ring on each side; internal threads at one end to receive pipe section #1; and external, untapered left-hand threads extending "substantially to the other end" (i.e., essentially all the way to its end face).
2. Two sealing rings against opposite flange faces — the first ring between the flange and pipe section #1, the second ring facing the open end and lying substantially flush with that end of the first member (because the flange is placed so close to the end).
3. Second member: hollow; has an intermediate shoulder; internal untapered left-hand threads running from the outer side of the shoulder toward one end, to mate with the first member's left-hand threads and running "substantially to the shoulder proper"; and internal right-hand threads running from the inner side of the shoulder the other way, extending essentially from the shoulder and for only about two or three turns, sized to engage the tapered threads of pipe section #2 without tightening on them.
4. Design relationship (the numerical/functional limitation): the length of the second member's left-hand threads must be substantially equal to, but not less than, the length of its internal right-hand threads multiplied by the ratio of the left-hand pitch to the right-hand pitch — i.e., L_left ≥ L_right × (p_left / p_right). This minimizes the required left-hand thread length while still guaranteeing the members cannot bottom out before the second pipe end reaches the second sealing ring.
5. The relative placement of the thread sets (threads 29 to the end face; threads 28 to the shoulder) plus the flush sealing ring minimize axial length, letting the union fit pipe ends that are very close together.

Claim 2 — dependent on Claim 1 (wall-clearance feature). The second member is cylindrical at one end and non-cylindrical (wrench flats) at its other end, and the cylindrical portion's axial projection lies entirely in exterior encompassing relationship to the non-cylindrical portion. Effect: the union automatically spaces itself off a wall by enough to get a wrench on the nut without damaging the wall — the "tight-against-a-wall" advantage described in the spec (flange 27 acting as a spacer, slots/recesses 45 at the corners of the non-circular portion 31).

Claim 3 — dependent on Claim 1 (single-cut/repair feature). The far end of the second member is spaced from the outer end of its left-hand threads by a distance slightly greater than the standard length of the pipe-section threads, so that end acts as a guide: aligning that end face (19a) with the base of the pipe threads positions the union correctly for a starting fit. This enables the union to be used on pipe ends formed by a single cut in an existing line, with only modest flexure of the pipe (e.g., between elbows 42/43) to insert the members and bring them into sealing relationship.

Prior art cited on the face of the patent

US 176,796 (Martin, 1876); US 861,828 (Grindrod, 1907); US 1,191,886 (Glauber/Citizens Savings & Trust, 1916); US 1,325,826 (Baker, 1919); US 1,570,155 (Laskowski, 1926); US 1,889,869 (Montgomery, 1932); US 2,187,217 (American Hard Rubber Co., 1940); US 2,374,138 (Sanford, 1945); US 2,702,715 (Andrews, 1955); plus GB 5,302 (1904) and GB 380,557 (1932).

Confidence caveats

• High confidence on title, inventor, filing/issue dates, claims text, and cited references — these come from the authoritative full patent text.
• Lower confidence / unresolved: (a) the Sept. 13 vs. Sept. 15, 1957 filing-date discrepancy (reported literally, not reconciled); (b) assignee — the record shows only "Individual," so no corporate assignee should be attributed; (c) any litigation or Federal Circuit activity — none found, but I cannot rule it out with certainty, and the unrelated "027"-ending patents (e.g., U.S. 8,187,027) must not be conflated with this one.

I'll search for litigation records involving US patent 2926027.

Findings: Litigation Involving U.S. Patent No. 2,926,027

No litigation involving U.S. Patent No. 2,926,027 was found.

I ran multiple targeted searches across the sources you specified — Unified Patents' litigation portal, CourtListener/Recap records, Justia federal district court dockets, and general web queries keyed to the exact number ("2926027" and "2,926,027"). Every hit returned unrelated patents (e.g., the '926 and '729 patents in Baby Jogger v. UPPAbaby, the '987/'092 patents in Inguran v. ABS Global, the '926 patent in DuPont v. Unifrax, etc.) — none of which is the patent at issue. I found no complaint, docket entry, judgment, or case list naming US 2,926,027 as an asserted patent.

Why this result is unsurprising

Several facts about this patent make an empty litigation record the expected outcome:

1. It is very old and long expired. Per the Google Patents record, the application (Serial No. 683,753) was filed September 13, 1957, the patent issued February 23, 1960, and its anticipated expiration is February 23, 1977 — a statutory term computed under the pre-URAA 17-years-from-issue rule. The patent's enforceable life, and therefore the window in which infringement suits could be filed, closed roughly 49 years ago.

2. No assignment or corporate owner of record. The Google Patents page lists both the original assignee and current assignee as "Individual", with inventor Riley F. Marquis, Jr. (Los Angeles, California). Small individual patentees of this era rarely generated reported litigation, and there is no large corporate portfolio owner to pursue enforcement today.

3. The "litigation" data on the Google Patents page is not litigation. The page contains only:

   • 11 cited references (prior art), e.g., US 176,796, US 861,828, US 1,191,886, US 1,325,826, US 1,570,155, US 1,889,869, US 2,187,217, US 2,374,138, US 2,702,715, and GB 5,302 / GB 380,557; and
   • 26 documents citing it (forward citations), e.g., US 3,520,561 (Global Marine), US 4,060,264 (I-T-E Imperial), US 4,258,936 (Allied Tube & Conduit), US 4,458,926 (Williamson), US 4,735,442 / US 4,809,418 (Gunzenhauser), US 5,547,229 / US 5,636,876 / US 6,200,113 (Eidsmore), US 5,520,421 (Lenwar), US 6,578,876 (Good Turns), US 6,598,908 / US 7,032,934 (Wosik), and the Mercedes Textiles fire-hose-coupling family.

   These are patent citations only — not legal proceedings.

Specified fields (per your requested format)

Important caveats and limitations

• I could not directly query PACER (it is a paywalled system requiring authentication) or run a native database search on Unified Patents' litigation case list beyond the returned web results. My conclusion rests on open-web and CourtListener/Justia searches.
• Pre-2000 litigation is poorly digitized. Because this patent's enforcement window was 1960–1977, any suits would have been filed in that era. Federal court records from that period are often not available in CourtListener, Unified Patents, or free full-text databases. It is possible that a 1960s–70s infringement action (or an interference/validity challenge) exists in paper-only archives or in West's Federal Reporter / Federal Supplement print volumes that are not indexed by the tools available here.
• Search near-miss risk. "2926027" and "2,926,027" are easily confused with similarly-numbered patents (e.g., the '926, '729, '987, '092 patents above). I deliberately excluded those as non-matches rather than treating them as results for this patent.

Bottom line

Based on the sources available to me and the specific number 2,926,027, no litigation is known. If you need a definitive, exhaustive answer, the next step would be a paid search in Derwent LitAlert, Lex Machina / Docket Navigator, or a manual review of the Federal Reporter (2d series) and Federal Supplement volumes covering 1960–1977, since pre-digital dockets for this expired individual-owned patent may simply never have been captured online. I would not represent the absence of found litigation as proof that none was ever filed.

I'll verify the structured data against live sources before writing this up.

Proceedings overview

Total AIA trial proceedings on file for US 2,926,027: 0. The USPTO Open Data Portal structured block reports no IPR, PGR, or CBM activity, and my live searches of PTAB/docket aggregators for proceeding numbers, petitioner names, and the patent number returned no petition, institution decision, Final Written Decision, or Federal Circuit appeal. The breakdown requested (active / invalidated / sustained / settled / denied) is therefore all zeros, and the defense posture is not "the patent is hardened by surviving IPRs" — it is "there is nothing left to invalidate, because the patent expired on 1977-02-23, roughly 35 years before the IPR statute existed."

That distinction matters, and it cuts against the usual heuristic. The prompt's default assumption — "the absence is itself a signal; well-asserted patents eventually attract IPRs" — does not apply here, for structural reasons set out below. A defendant holding a demand letter citing US 2,926,027 is not looking at an untested-but-live patent; it is looking at a patent that has been in the public domain for 49 years.

Key bibliographic anchors (source: Google Patents US2926027A; underlying application US 683,753):

No proceedings to report

There are no PTAB proceeding sections to write, and I will not fabricate docket numbers, petitioner names, panels, or dispositions. What I can substantiate:

• Structured source (canonical): the "PTAB proceedings on file" block states the USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest.
• Independent verification (2026-09-30): searches for "2926027" PTAB proceeding petitioner patent owner, for US 2,926,027 IPR / inter partes review, and for the patent number in combination with litigation, ex parte reexamination, and inter partes review returned no PTAB petition, Board decision, or CAFC opinion for this patent. The only hits matching the literal string "2926027" were unrelated foreign documents (e.g., an Irish journal entry listing IE 2926027, "Power transmission system for long travel distances having a guide channel," Tsubaki Kabelschlepp GmbH) — not US 2,926,027 and not a proceeding.

Why zero is the expected answer, not a data gap. AIA trials post-date this patent's entire enforceable life by decades:

• IPR (35 U.S.C. §§ 311–319) became available for petitions filed on or after 2012-09-16. The patent expired 1977-02-23 — 35 years earlier. There was no window in which an IPR could have been filed while the patent still had enforceable term.
• PGR (35 U.S.C. §§ 321–329) is limited to patents with an effective filing date on or after 2013-03-16. This patent's effective filing date is 1957-09-13, so a PGR is categorically unavailable — it is not merely unlikely, it is legally impossible.
• CBM (AIA § 18) required a patent claiming a "financial product or service" and, in any event, the transitional program sunset on 2020-09-16. Neither the subject matter (plumbing unions, Class 285) nor the sunset date permits a CBM today.
• No maintenance-fee lapse or reissue/reexam events appear in the ODP/Google legal-status timeline — consistent with a pre-1981 patent (maintenance fees apply only to patents filed on or after 1980-12-12), which simply ran its full 17-year term from grant to 1977-02-23. I flag that I did not independently pull the paper reexamination file history; the structured record shows no reexamination certificate.

Judges, panels, grounds, FWDs, settlements, appeals: none exist. The patent's own citation record — 11 references cited (US 176,796; US 861,828; US 1,191,886; US 1,325,826; US 1,570,155; US 1,889,869; US 2,187,217; US 2,374,138; US 2,702,715; GB 5,302; GB 380,557) and 16+ forward citations including US 6,578,876 (Good Turns, LLC) — reflects ordinary citation traffic over six decades, not challenge activity. No defensive aggregator (Unified Patents, RPX, AST, etc.) appears anywhere in the chain; the patent has always been individually owned.

Strategic summary

Claim status: all of claims 1, 2, and 3 are UNTESTED — but untested is irrelevant, because the patent has been expired and unenforceable since 1977-02-23. Nothing was canceled, nothing was sustained; there has never been an adversarial validity adjudication of any kind on this patent in a PTAB forum. Claim 1 (the two-member union with an internal flange, opposed sealing rings, peripheral untapered left-hand threads on the first member, a shoulder intermediate the second member, right-hand threads of "only two or three" turns, and the express thread-length-to-pitch-ratio limitation tying left-hand thread length to right-hand thread length times the pitch ratio), claim 2 (peripherally cylindrical end / non-cylindrical wrench portion with the axial-projection-and-wall-clearance limitation), and claim 3 (the rear-surface-to-thread-start guide dimensioned slightly greater than standard pipe thread length to accommodate a single-cut repair) all stand exactly as granted — and all are legally dead as causes of action.

Estoppel landscape: there is none, and there is nothing for it to bite on. Section 315(e)(2) estoppel only arises against a petitioner that obtained an institution decision and reached a final written decision. With no petitions ever filed, no petitioner, privy, or real party in interest is estopped, and no petitioner could have been. Symmetrically, the patent owner has never had a PTAB win to defend and has never appealed to the Federal Circuit. The only theoretically "available" prior-art grounds are the ones nobody has ever needed to run — the 11 cited references, prominent among them US 2,702,715 (Andrews, "Threaded extensible pipe coupling for insertion between and connection to fixed members of a pipe line"), which is facially close to the "insert the union into a single cut in an existing line" teaching of Fig. 5 and claim 3. A defendant would never need to spend money on that analysis; expiration is a complete and far cheaper defense.

Pattern signals: no pattern exists to read. No repeat petitioner, no serial petitions, no POP or Precedential Opinion Panel request, no sanctions activity, no joinder motions, no IPR-then-district-court parallel tracks, no CAFC docket. If a current demand letter cites US 2,926,027, the assertion is not a hardened monetization campaign — it is either (a) a clerical error in a patent-number list, or (b) an intentional rely-on-the-recipient-not-checking demand. Both are addressed below.

Recommended next steps

1. If you have received a demand citing US 2,926,027, your response is a one-paragraph expiration letter, not an IPR petition. The dispositive facts, quotable and verifiable:

US 2,926,027 issued 1960-02-23 from application Ser. No. 683,753, filed 1957-09-13, and its term expired on 1977-02-23. The patent is expired and in the public domain (Google Patents US2926027A, listing legal status "Expired – Lifetime" and anticipated expiration 1977-02-23).

An expired patent cannot be infringed, and 35 U.S.C. § 286 caps damages at six years before suit — which here reaches back to 2020, four decades after expiration. There is no live royalty base, no injunction available, and no claim construction worth litigating. Demand letters that cite long-expired patents are a recognized hallmark of abusive or fraudulent assertion; document the letter, the sender, and any fee demand, and consider reporting to your state AG consumer-protection unit and to the USPTO's Office of Enrollment and Discipline if a licensed practitioner signed it.

2. Do not file an IPR "for safety." It would be a waste of the filing fee and, worse, it would signal that the demand has substance. Two threshold problems make such a petition either pointless or procedurally doomed: (i) PGR is unavailable because the effective filing date (1957-09-13) predates the 2013-03-16 statutory cutoff, and CBM is unavailable both on subject matter and because the program sunset 2020-09-16; and (ii) on an expired patent the Board applies Phillips-style claim construction and, as a general matter (I state this as background law rather than as a finding about this patent), a party needs a concrete competitive injury to have Article III standing to appeal an adverse final written decision. An academic challenge to a 1977-expired drain-fitting patent is exactly the scenario where that standing problem bites.

3. If the "demand" is actually a litigation or an arbitration demand, escalate on the statute of limitations and on Rule 11 / § 285. Any count alleging infringement of US 2,926,027 is barred by expiration and § 286 on its face. If the pleading nonetheless asserts claims 1–3, that is a candidacy for a motion to dismiss with prejudice and a fee motion; the patent's 49-year post-expiration public-domain status means there is no good-faith basis to assert it. (Note there is no PTAB FWD to link to and no CAFC disposition to quote — the strongest authority is the patent's own face, which is why the citation above is to the patent document rather than to a decision.)

4. Confirm the number before you act. Because expired patent numbers are frequently recycled or mistyped in demand letters, verify that the demand actually means US 2,926,027 (plumbing union, Marquis) and not, e.g., a similarly numbered live patent. The structured ODP record and my searches both key on the literal number 2926027; if the letter cites a different number, or cites a reissue or a continuation, re-run this analysis against that document — my conclusion of "no PTAB activity, patent long expired" applies to US 2,926,027 as identified in the authoritative full text supplied above and to no other patent.

Bottom line for a defendant: zero PTAB proceedings, no estoppel to navigate, no claim-level FWD to exploit — and no need for any of it. The patent's expiration on 1977-02-23 is superior to any IPR outcome you could buy.

I'll verify the assignment record for this patent before producing the analysis.

Assignment Reconstruction — US 2,926,027

Data-source caveat up front: I could not execute a live query against the Assignment Center's search backend (it is a JavaScript application that my search tools do not index directly). What I can report is that (a) no assignment public-record extract surfaced for this number in any indexed source I queried, and (b) the Google Patents legal-events record for this patent — which mirrors recorded-assignment postings — shows only three events: application filing (1957-09-13), grant (1960-02-23), and anticipated expiration (1977-02-23). No assignment, security-interest, merger, or change-of-name event appears. Verify directly at assignmentcenter.uspto.gov and Google Patents legal events.

Inventors

Pattern note (no fire-sale precursor present): The watch-pattern you asked about — all inventors departing the original assignee within 12 months of filing — is structurally impossible here. There is exactly one inventor, he is the owner (assignee listed as "Individual"), and there is no company for him to depart from. This is a classic mid-century garage-inventor filing, not a corporate portfolio whose inventors scattered before a sale.

Original assignee

• Entity on the issued patent: "Individual" — i.e., the patent issued to the inventor personally. Google Patents flags that assignee listings "may be inaccurate," but here the absence of any corporate name in the record is consistent across sources, so no corporate assignee should be attributed.
• Product embodying the claims: Unknown. The patent itself is directed at a plumbing drainage union (Class 285-32), and the specification is written as a practical installation guide (sink 5, lavatory 6, water closet 7, neoprene rings 21/22). That is evidence of intent to commercialize, not evidence of an actual shipped product. I found no product literature, catalog, or trademark linking Riley F. Marquis, Jr. to a manufactured union.
• Primary line of business: Not determinable. No corporate record exists in the patent data.
• Current status: N/A. The relevant status is the patent's: Expired – Lifetime, anticipated expiration 1977-02-23 on the 17-year pre-1995 term. The patent has been in the public domain for ~49 years.

Assignment timeline

The Assignment Center has no records for this patent.

I found no recorded assignment, security agreement, merger, change of name, license, or correction for US 2,926,027 in any indexed assignment source. Stated plainly and per instructions, that is the finding, and the substantive chain analysis stops here.

Two independent reasons this result is expected rather than surprising — both worth recording:

1. There was likely nothing to assign. The patent issued with the assignee listed as the inventor individually. A patent issues in the name of an assignee only because an assignment was recorded during prosecution. Issuance to "Individual" therefore affirmatively indicates no pre-issue assignment was recorded. Since no post-issue assignment appears either, the inventor-owner never transferred the patent on the record.
2. The era predates reliable electronic indexing. Recording was done on paper/microfilm reels before the modern reel-and-frame electronic system (roughly complete from the 1980s forward). A 1957 filing with a 1977 expiry sits entirely outside the window where Assignment Center coverage is dependable. So the honest characterization is: no assignment found, and low confidence that the negative is exhaustive — absence of evidence here is weaker than it would be for, say, a 2005 patent.

Consequently, no reel/frame numbers, correspondent attorneys, or conveyance types exist to report. There is no correspondent of record to test for the "same lawyer, different shells" tell — the shell-LLC pattern this task is designed to catch did not exist in 1957 and required no shell to accomplish the same end (a lone inventor simply owned it).

Timeline diagram

timeline
    title Ownership of US 2926027
    1957 : Filed by Riley F Marquis Jr
         : Sole inventor and owner
    1960 : Patent issued to Individual
         : No assignment of record
    1977 : Expired for full term
         : No transfers ever recorded

NPE / troll-pattern signals

Not one of the eight signals can be grounded in a reel/frame entry, because there are no entries. Each is therefore recorded as follows, with the reason stated rather than left bare:

Explicit anti-finding: the fact that this is an anonymous-looking individual ownership with no assignee is not an NPE signal. That inference would violate the rule against inferring from naming/appearance alone. In 1957 an individual-inventor patent with no recorded assignment was the ordinary case, and the patent has been expired since before most NPE business models existed.

Verdict

Insufficient data — no assignment records exist, and only the original issuance to the inventor is documented.

The Google Patents legal-events record for US 2,926,027 contains exactly three entries (filing 1957-09-13, grant 1960-02-23, anticipated expiration 1977-02-23) and no assignment, security-interest, or name-change event, and I found no reel/frame, correspondent, or conveyance to analyze. The patent issued to "Individual" — i.e., directly to sole inventor Riley F. Marquis, Jr. — which affirmatively indicates no pre-issue assignment was ever recorded, and it expired for full term on 1977-02-23, roughly four decades before the NPE patterns this framework targets. Adding a caveat the prior section already flagged: this patent must not be conflated with the unrelated EP 2926027 (Tsubaki Kabelschlepp power-transmission system) or U.S. 8,187,027, both of which surface on naive number searches.

Recommended next step for a definitive answer: pull the paper/microfilm assignment abstract for Serial No. 683,753 through USPTO's Assignment Recordation Branch (the pre-1980 records are not fully in the electronic index), and separately confirm the pro-se filing by checking the printed "Attorney" field on the front page — which the OCR dump provided to me omits, so my "no attorney of record" statement is low confidence and should be verified against the PDF at patentimages.storage.googleapis.com before being relied on.

I'll ground this in searches on the specific cited references before analyzing them.

Scope and method

The authoritative source here is the full text you supplied (Google Patents mirror of the USPTO record, US2926027A, fetched 2026‑09‑30), specifically the patent's own "References Cited in the file of this patent" block — which is the USPTO record of prior art. I searched externally to corroborate each reference's dates/inventor. I hit my search-call budget before I could open full texts for every one of the 11 references, so I have marked clearly below which descriptions are verified from the reference's own text versus inferred from its title/classification/date. Anticipation verdicts are therefore provisional.

Two housekeeping flags:

• Statute: this application was filed 1957‑09‑13 and issued 1960‑02‑23, so the governing provision is pre‑AIA 35 U.S.C. §102 (1952). "Anticipation" = a single reference disclosing every element of a claim. All 11 references predate the filing date by well over a year, so all are available as §102(a)/(b) art (the latest, Andrews '715, issued 1955‑02‑22 — more than the one‑year §102(b) grace period before the 1957 filing).
• Date inconsistency in your prompt: the task header says "Current Date: April 26, 2026," while the tool environment and the prior sections state 2026‑09‑30. Nothing in this analysis turns on it, but I am not silently reconciling it.

The 11 references of record, with §102 assessment

1. US 176,796 — Martin — "Improvement in tubing for oil‑wells"

• Dates: issued 1876‑05‑02. No filing date appears in the retrieved record (19th‑century records often omit it).
• Description: an oil‑well tubing improvement; threaded tubing/casing joints for downhole pipe. (Inferred from title and era; full text not opened.)
• Potential §102 anticipation: none. Even assuming it shows screwed pipe sections, it cannot disclose the two‑member telescoping union, the internal flange with two sealing rings, the opposite‑hand thread pair, or the pitch‑ratio length relationship of claim 1.

2. US 861,828 — John Grindrod — "Pipe‑coupling"

• Dates: filed 1906‑03‑08; issued 1907‑07‑30.
• Description: a pipe coupling (sleeve/union type). (Inferred; full text not opened.)
• Potential §102 anticipation: none as to claims 1–3. Generic coupling art; no evidence of the '027 combination.

3. GB 5,302 (1904) — George Theodore Temple — "Improvements in Pipe Couplings"

• Dates: filed/priority 1904‑03‑03; published 1904‑04‑07. Note the '027 face prints this as "5,302 Great Britain Mar. 3, 1904" — i.e., the filing date, not the publication date. Full number as indexed: GB190405302A.
• Description: pipe couplings. (Inferred; full text not opened.)
• Potential §102 anticipation: none as to claims 1–3.

4. US 1,191,886 — (inventor Glauber) / Citizens Savings & Trust Co. — "Coupling for basins, &c."

• Dates: filed 1911‑12‑18; issued 1916‑07‑18.
• Description: a coupling for basins — the closest of the old references on subject‑matter context (plumbing/sanitary drainage fittings, as in the '027 sink/lavatory line 9). Likely a union or spud‑type fitting, possibly with a seating gasket. (Inferred; full text not opened.)
• Potential §102 anticipation: none demonstrated.

5. US 1,325,826 — Baker — "Coupling"

• Dates: the '027 face prints "Baker Dec. 23, 1919"; the index record gives 1919‑12‑23. ⚠️ Flag: that date may be the issue date rather than the filing date, and the record I retrieved carried no inventor/assignee line, so "Baker" is not independently corroborated. I am reporting both literally.
• Description: a coupling. (Inferred; full text not opened.)
• Potential §102 anticipation: none.

6. US 1,570,155 — "Pipe connection"

• Dates: filed 1924‑08‑09; published 1926‑01‑19 (application US73111424A). Classified F16L15/08 (threaded pipe connection).
• Description: threaded pipe connection with sealing. (Inferred; full text not opened.)
• Inventor‑name discrepancy — flagged, not corrected: the '027 face prints "Karbowski"; the record I retrieved lists inventor Victor Karbowski with Charles Laskowski as assignee/holder (Google's indexing renders Laskowski in the inventor slot). Treat the attribution as unresolved.
• Potential §102 anticipation: none demonstrated.

7. US 1,889,869 — Gustavus A. Montgomery — "Tube coupling" ⭐ most conceptually relevant of the older art

• Dates: filed 1930‑08‑02; issued 1932‑12‑06.
• Description: A coupling with no internal stop shoulder, in which a yielding/resilient sealing member sits centrally between the adjacent tapered pipe ends, so a seal is made at the pipe ends by compression rather than by relying on the pipe threads. I verified this characterisation from the sibling patent by the same inventor — US 1,889,868, "Double box rotary tool joint," filed 1930‑07‑21, issued 1932‑12‑06 — in which the specification states there are "no shoulders or stop means… to limit the inward screwing of the pipe ends relative to the coupling," and describes a resilient sleeve with end rings that "contact with the ends of the pipe sections and thus preserve a seal therewith." ⚠️ Caveat: this is the sibling document, not '869 itself; the correspondence is strong but I have not read '869's own text.
• Potential §102 anticipation: This is the best prior teaching of one core idea of claim 1 — sealing by a compressible member abutting the pipe ends, with sealing independent of the pipe threads. But '869/'868 is a single coupling sleeve with two tapered box ends, not a two‑member telescoping union, and it has no left‑hand/right‑hand thread pair and no pitch‑ratio relationship. → Does not anticipate claims 1–3; it is the most relevant §103 reference against the "sealing ring compressed between the flange and the pipe end" concept.

8. US 2,187,217 — American Hard Rubber Co. — "Pipe joint"

• Dates: filed 1938‑06‑29; issued 1940‑01‑16.
• Description: a pipe joint from an elastomer manufacturer — plausibly relevant to rubber/neoprene gasket or sealing‑ring pipe joints. (Inferred; full text not opened.)
• Potential §102 anticipation: arguably relevant only to the elastomeric sealing‑ring element of claim 1; it cannot anticipate claim 1 as a whole, nor claims 2–3.

9. US 2,374,138 — Hugh W. Sanford — "Pipe joint"

• Dates: filed 1941‑02‑19 (application US37968141); issued 1945‑04‑17. ⚠️ Flag: the '027 face prints "Apr. 17, 1940." The record shows 1945‑04‑17. This looks like an OCR/typo error in the printed citation; I am reporting both literally rather than silently correcting.
• Description: a pipe joint. (Inferred; full text not opened.)
• Potential §102 anticipation: none demonstrated.

10. US 2,702,715 — Forrest Andrews — "Threaded extensible pipe coupling for insertion between and connection to fixed members of a pipe line" ⭐ closest single reference

• Dates: filed 1949‑06‑24; issued 1955‑02‑22.
• Description — verified from the reference's own text: a two‑member connector: a male member 5 and a female member 6, each with a collar (7, 8) whose periphery is "polygonal, hexagonal or octagonal… to facilitate grasping with a wrench," and each with a standard tapered internal pipe‑thread socket (9, 10) for the pipe ends. The male member is externally screw‑threaded (11) and telescopes into an internally threaded bore (14) of the female member, with a tapering threaded section 13 into an enlarged threaded head 12. Its stated purpose is to avoid having to spring the pipe ends when the pipes are rigidly secured at their far ends. Notably, the specification states: "All threads in the connector are designed to be of the same pitch as pipe threading" and "all the connector threadings except one of the pipe connecting threads are designed to be of the same pitch."
• Why it is the closest: it overlaps the '027 on (a) a two‑part union replacing the old three‑member "ground type" connector (the very disadvantage the '027 states in its "Definitions" passage), (b) telescoping threaded engagement between the two members, (c) standard tapered pipe threads at both pipe ends without altering the pipe, (d) wrench flats, and (e) the problem of inserting the coupling into an already‑fixed line.
• What it lacks: no opposite‑hand threads (all one pitch/hand); no internal flange with two sealing rings; no ring flush with the member's end face; no 2–3 turn non‑tightening right‑hand thread; no L_left ≥ L_right × (p_left/p_right) relationship; no guide dimension.
• Potential §102 anticipation: none as to claims 1–3. It is the leading §103 reference — against claim 1's two‑part extensible coupling framework and against claim 3's single‑cut/fixed‑line insertion problem, and it touches claim 2's wrench‑grip feature (though not claim 2's specific "cylindrical portion encompassing the non‑cylindrical portion" wall‑spacing relationship).

11. GB 380,557 — Frederick George Jones — "Improvements in screwed pipe joints and in fittings relating thereto"

• Dates: filed 1931‑07‑30; published 1932‑09‑22 (the '027 face prints only "Sept. 22, 1932," i.e., the publication date).
• Description: screwed pipe joints and associated fittings. (Inferred; full text not opened.)
• Potential §102 anticipation: none demonstrated. Noted as "screwed" (past tense) exactly as the title is rendered in the record.

Ranked answer: the most relevant prior art

Conclusion on §102

On the record available to me, none of the 11 cited references anticipates any of claims 1, 2, or 3 under pre‑AIA 35 U.S.C. §102. The reason is structural: claim 1's novelty resides in a combination that no single one of these references shows —

1. a two‑member telescoping union (not a coupling sleeve, and not the old three‑part union);
2. an internal flange carrying two seals against the two pipe end faces, with the outer seal substantially flush with the member's end face;
3. an untapered left‑hand external thread on member 18 running to its end face, mating with an untapered left‑hand internal thread on member 19 running to its shoulder;
4. a right‑hand internal thread on member 19 of only ~2–3 turns, sized to engage tapered pipe threads without tightening; and
5. the quantitative limitation that the left‑hand thread length be not less than the right‑hand thread length × the pitch ratio (p_left/p_right) — so that opposite‑hand/opposite‑pitch engagement produces cumulative axial travel sufficient to seat the second pipe end before either thread set bottoms out.

Elements 1 and (partly) the tapered‑pipe‑thread context appear in Andrews '715; the end‑face, thread‑independent seal appears in Montgomery '869. The opposite‑hand thread pair, the flush outer seal, the 2–3 non‑tightening turns, and the pitch‑ratio length relationship, however, appear in none of the cited art. The references are therefore best characterised as the §102(a)/(b) background and §103 base, not as anticipatory art — which is consistent with the examiner's decision to allow the claims over all 11.

Caveats and contradictions

• Verified vs. inferred: Only US 2,702,715 was described above from its own text (retrieved), and US 1,889,869 was characterised from its sibling US 1,889,868. The other nine descriptions are inferred from title, classification and date and are flagged as such; I did not open their full texts. The anticipation verdicts for those nine are correspondingly provisional.
• Printed‑date discrepancies on the '027 face (reported literally, not auto‑corrected): US 2,374,138 shows "Apr. 17, 1940" vs. record 1945‑04‑17; US 1,570,155 shows "Karbowski" vs. record inventor Victor Karbowski / holder Charles Laskowski; the US 1,325,826 "Baker" attribution is uncorroborated; GB 5,302 is printed with its filing date (Mar. 3, 1904) while GB 380,557 is printed with its publication date (Sept. 22, 1932).
• Beyond the record (flag, do not rely on): my searches surfaced US 1,849,066 (1932), a sucker‑rod/pipe "duplex box" coupling whose specification expressly discusses the "hands" of the threads and arranging them so units screw up by turning in the same direction. It is not cited on the face of the '027 and I have not analysed it as art; I note it only because it shows that the broader field did contain opposite‑hand thread‑arrangement teachings that the examiner evidently did not treat as anticipatory of this combination.
• No litigation content is added here (none found previously); the 26 forward citations listed in the earlier section are citing documents, not proceedings, and I have not treated them as prior art of record for §102 — only the 11 references on the patent face qualify.

I'll research the cited prior art references to ground the obviousness analysis. Let me pull details on the most structurally relevant references.

Let me pull details on additional cited references relevant to specific claim limitations.

Obviousness Analysis — U.S. Patent No. 2,926,027 under 35 U.S.C. § 103

Note on dates/authority. The application was filed September 13, 1957 and the patent issued February 23, 1960, so this is a pre-AIA § 103 analysis and the operative "time of the invention" is 1957. The patent expired 1977-02-23, so this is a retrospective/hypothetical validity assessment (relevant to historical or defensive purposes), not a live challenge to enforceable claims. I flag that the OCR of the printed patent shows a cover-page "Filed Sept. 15, 1957" that conflicts with the bibliographic/priority date 1957-09-13; per the strict-ID rule I report both and do not reconcile them. Operative prior art must therefore predate September 13, 1957.

I build on — and do not repeat — the earlier summary and litigation sections.

1. Legal framework

• Graham v. John Deere Co., 383 U.S. 1 (1966): scope/content of the prior art, differences, level of ordinary skill, and secondary considerations.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): a claimed combination is obvious where the elements were known, the combination does no more than yield predictable results, and there existed a design incentive, market pressure, or "finite number of identified, predictable solutions." A "motivation to combine" may be found in the nature of the problem itself.
• MPEP §§ 2141–2143: the "lead" reference plus articulated reasoning (e.g., "obvious to try," "design choice," "optimization of a result-effective variable").
• In re Aller, 220 F.2d 454 (CCPA 1955): optimizing a result-effective variable through routine experimentation is obvious.
• The specification's own statements about "plumbing unions in the past" are admissions about the prior art usable against the claims (MPEP § 2129).

2. The level of ordinary skill (PHOSITA), 1957

A plumber or a plumbing-fitting designer/mechanical engineer with roughly 2–5 years' experience designing pipe joints, familiar with (a) standard tapered pipe threads (the 1957-era ASA/NPT form), (b) conventional union construction, (c) elastomeric gasket/ring seals (the patent itself names neoprene), and (d) the ubiquitous turnbuckle principle of opposite-hand threads. This is a mature, crowded mechanical art — a § 103 "combinations of old elements" case.

3. The prior art of record (11 cited references)

⚠️ Honesty caveat: I retrieved the full text of Andrews (US 2,702,715) only. My searches for the other ten returned unrelated material, so I do not assert specific disclosures for them beyond title, date, and class. Where I rely on them below, I do so as a class/field characterization and mark my uncertainty; a real reexamination/IPR-style rejection would require pulling each PDF from patentimages.storage.googleapis.com and quoting it.

Two facts about the record itself matter:

1. The examiner had all 11 references and still allowed three claims — evidence (not dispositive) that no single-reference anticipation and no obvious combination was found in 1957–1960. Any obviousness attack must overcome this.
2. The forward citations (26 documents) — Global Marine, I-T-E Imperial, Allied Tube & Conduit, Gunzenhauser, Eidsmore, Wosik, Mercedes Textiles — post-date the patent and are therefore not prior art; they show only that the field kept building on the disclosure.

4. The closest prior art: Andrews, US 2,702,715

Andrews is the single most damaging reference because it targets the same problem and the same context. Its own words (quoted from the retrieved text, https://patents.google.com/patent/US2702715):

• "Pipe connectors as known heretofore have not been entirely satisfactory, especially when the pipes to be connected are rigidly secured at their ends… In installing such connectors, it has been necessary heretofore to spring the ends of the pipes…"
• "Such pipe connectors are chiefly of the ground type, comprising three members, two of which are threaded onto the pipes… with matching ground surfaces and having a slip-over tightening ring…"
• Andrews discloses male member 5 / female member 6; polygonal/hexagonal collars 7, 8 "to facilitate grasping with a wrench"; internally threaded sockets 9, 10 with standard tapered threads matching the pipe threads; and specular threaded portions (11, 12, 13/14, 15, 16).

What Andrews is missing for claim 1: (i) it is not a two-part union in the claimed sense; (ii) it does not use opposite-hand threads — Andrews expressly says "All threads in the connector are designed to be of the same pitch as pipe threading," relying on oversize/cold-flow ("wrench tight") engagement rather than reversed pitches; (iii) it has no internal flange with two sealing rings sealing on the pipe ends. So Andrews alone does not anticipate, and no single reference appears to.

5. Claim 1 — the combination case

Claim 1 is a two-member union in which:

• A first member has an internal flange with a ring on each side, and seals independently of the pipe threads;
• B the first member's other end carries external, untapered, left-hand threads extending substantially to the end face, and the second ring sits substantially flush with that face;
• C the second member has a shoulder, internal untapered left-hand threads running to that shoulder, and internal right-hand threads of "only two or three turns" engaging the second pipe's standard tapered threads "without tightening"; and
• D the quantitative limitation L_LH ≥ L_RH × (p_LH / p_RH).

Proposed combination for Claim 1

Andrews (US 2,702,715) as the lead reference + a turnbuckle / opposite-hand-thread teaching + a gasketed-union reference.

A challenger's mapping:

• A — an internal-flange union sealing on the pipe ends with two rings reads on any of the union/pipe-joint references (US 1,889,869 Montgomery "Tube coupling"; US 2,374,138 Sanford "Pipe joint"; US 2,187,217 AHR "Pipe joint"), and even on the "ground type" three-member union Andrews describes as old. (⚠️ content of these three not independently verified.)
• B/C — the two-member telescoping body, standard tapered pipe-thread sockets, and wrench flats come straight from Andrews.
• B/C/D — the opposite-hand threads come from a turnbuckle/differential-screw teaching. Note the specification itself admits the operative principle: "This result is achieved by virtue of the fact that the threads are of opposite pitches." Opposite-hand threads in a pipe coupling to obtain cumulative take-up from a single rotation are classic turnbuckle art, and GB 380,557 (Jones, "screwed pipe joints") / GB 5,302 (Temple, pipe couplings) / US 1,325,826 (Baker, "Coupling") / US 176,796 (Martin, oil-well tubing) are the references in the record most likely to supply it. (⚠️ I did not verify which, if any, actually shows reversed pitches — this is the linchpin the challenger must prove, and I flag it as unverified.)

Why the PHOSITA would have combined them (KSR "articulated reasoning"):

1. Same field, same problem. Andrews explicitly identifies the problem set the '027 patent recites — fixed pipe ends, repaired lines, "springing." Where the problem is the same, the motivation is found in the nature of the problem (KSR).
2. Known design incentives: the '027 specification and Andrews both identify (i) eliminating the three-part union, and (ii) eliminating sealing on the pipe threads. Reducing parts count and replacing a thread seal with a face/gasket seal are classic "design choice" optimizations.
3. Predictable result. Combining a turnbuckle's opposed threads with a two-member telescoping union yields exactly the predictable result — one nut rotation produces take-up on both members. No unexpected property is claimed.
4. Explicit teaching in the field of the "without tightening" requirement: it is a mere mechanical necessity (a lock-nut too tight stalls rotation), routinely handled by making the thread loose or few in number.

The weakness of the challenge: the precise numerical relationship D is a kinematic requirement, not a discovery — the patent's own derivation (n·p′ must be greater than d, i.e., cumulative travel must exceed the gap d + f) is arithmetic any PHOSITA performs. Under In re Aller, a limitation that is an optimization of a result-effective variable — here, "make the left-hand thread at least long enough to close the gap, and no longer, for compactness" — is obvious absent proof of an unexpected result. The word "substantially equal to, but not less than" reads on a range of routine proportions with no criticality shown.

6. Claim 2 — wall-spacing / wrench-flat geometry

Claim 2 adds: the second member is cylindrical at one end, non-cylindrical (flats) at the other, and the cylindrical portion's axial projection lies entirely in exterior encompassing relationship to the flats, so the union automatically spaces itself off a wall.

Andrews already discloses the "polygonal, hexagonal or octagonal" wrench collars. Making the wrench body smaller in diameter than the adjoining flange so the flange acts as a standoff is a predictable design choice to solve a well-known problem — wrench clearance next to a wall — which the '027 specification itself concedes (slots or recesses 45 … located at the corners of the non-circular portion 31; the circular flange 27 may contact the wall to act as a spacer). There is nothing in claim 2 beyond spatial arrangement/design choice; under KSR this is a strong candidate for invalidity if the primary claim falls.

7. Claim 3 — single-cut / "guide" dimensioning

Claim 3 makes the far end of the second member sit "a distance slightly greater than the standard length of the threads" from the outer end of the left-hand threads, so the rear face 19a acts as an alignment guide for a one-cut installation.

This is the most interesting dependent claim, but again it is a sizing choice applied to a problem Andrews already frames (joining rigidly fixed pipe ends without springing). Setting a shoulder/reference-face dimension to slightly exceed a standard, published thread length (NPT thread length is standardized) is the definition of a result-effective variable optimized by routine skill. Motivation: the repair/single-cut use case is stated in both the '027 specification (Fig. 5, elbows 42/43) and Andrews.

8. Nonobviousness considerations (the counter-case)

1. Examiner allowance over all 11 references. The strongest practical point for the patentee: the examiner considered the entire record and found no anticipating or obvious combination. A challenger must produce new art or a new rationale, not merely re-list the cited references.
2. No single reference, and arguably no two, disclose the complete combination — two members + dual face-seals + reversed-pitch threads + the specific length ratio. The "invention" is real as a combination, even if every element is old.
3. Possible teaching away from left-hand threads. Mixing thread hands in plumbing invites cross-threading/confusion; the patent's own legend 32 exists to cure that. A patentee could argue the art taught away from mixed-hand threads as error-prone. This is a weak argument (mere inconvenience, easily labeled) but it is the best non-obviousness hook.
4. Secondary considerations: essentially none in the record. There is no evidence of commercial success, long-felt need, failure of others, copying, or licensing — and with an individual owner (no corporate assignee) and no found litigation, the record offers no nexus evidence. Absence of evidence is not proof of absence, but it means the patentee cannot currently rebut a prima facie § 103 case with objective indicia.

9. Bottom line

• Not anticipatable by any single one of the 11 cited references — the examiner's allowance is consistent with that.

• Claims 1–3 present a defensible but vulnerable § 103 case. The most likely invalidity theory is:

  US 2,702,715 (Andrews) — extensible, wrench-flatted, tapered-pipe-thread coupling for fixed/repaired pipe ends (states the problem and the incentives) + a turnbuckle/opposite-hand-thread teaching (GB 380,557 / GB 5,302 / US 1,325,826 / US 176,796, or the notoriously old turnbuckle itself) + a gasketed face-seal union (US 1,889,869 / US 2,374,138 / US 2,187,217) → renders claim 2 (design choice) strongest for invalidity, claim 3 (routine sizing) next, and claim 1 the hardest to invalidate because of the specific length-ratio limitation — which is itself only a result-effective-variable optimization (In re Aller).

• The pivotal unverified fact is whether any record reference actually discloses opposite-hand (left/right) threads in a pipe coupling. If yes, claim 1 is likely obvious; if the record shows only same-pitch threads (as Andrews expressly does — "All threads in the connector are designed to be of the same pitch"), the challenger must rely on outside turnbuckle art, and the patentee has a genuine (if narrow) argument that the specific combination was not suggested.

Confidence caveats

• High confidence in claim text, the Andrews disclosure, and the field/problem framing.
• Low confidence on the actual disclosures of the other ten cited references — I retrieved only titles, dates, and classes for them; any element-mapping to those references above is inferential and flagged, and should be verified against the full PDFs before being relied upon.
• Per the earlier sections, the Sept 13 vs Sept 15, 1957 filing-date discrepancy, the "Individual" assignee, and the absence of any found litigation remain as previously stated and are not contradicted by anything found here.