US 2329020

I'll research this patent number against live sources before summarizing.

Let me check for any litigation or CAFC 2026 docket activity involving this patent number.

US 2,329,020 — Summary and Claim Overview

Important preliminary caveats

1. No abstract on record. US patents issued in 1943 were not required to carry an abstract, and none appears on the face of this patent or in the Google Patents / FreePatentsOnline records. The "Definitions" block on the Google Patents page is an automatically generated CPC/keyword glossary harvested from the OCR'd description — it is not an abstract and it is badly garbled. Do not cite it as such.
2. Number collisions — do not conflate. My searches returned several unrelated documents sharing the digits "2329020": PT‑2329020‑E (Portuguese, Novartis, polypeptide surface presentation), RU 2329020 C1 (Russian, ophthalmic laser thermotherapy), and a CourtListener investment record ID 2329020 in a judicial financial disclosure. None of these relate to the US patent below.
3. No litigation found. I found no district court or CAFC docket activity for US 2,329,020. This is expected: the patent expired in 1960, more than two decades before the Court of Appeals for the Federal Circuit was created (1982), so no CAFC 2026 docket can exist for it. My search did not surface any USPTO post-grant proceeding either (none would be available for a patent of this vintage). I state this as a negative search result, not a certified clearance.

The patent

On assignee: the authoritative sources list no corporate owner. I cannot confirm from the patent record that this was assigned to Wales‑Strippit or any other company; that association is background knowledge about the inventor, not something the patent front page establishes. Treat the assignee as unassigned / individual.

Subject matter (plain language)

The patent solves the problem that a conventional press punches vertically, which is awkward or impossible for work such as angle iron, rims, flanges and channel sections. Wales puts the punch and die on opposite "horns" (parts 16, 17) of a single holder separated by an open gap (slot 18) that the work slips into, so the hole is punched horizontally (or at any chosen angle) while the press ram still moves vertically. A vertical actuating pin (35) with a tapered face (41) cams against a matching inclined face (40) on the punch head (22) to convert ram travel into horizontal punch travel.

Key mechanical features:

• The punch is guided by a sleeve (42) that also serves as the stripper, bearing on the work so the punch bit retracts after piercing.
• A die (25) that is threaded into the holder, screwdriver-slot adjustable into the gap, and lock-nut locked, with a slug-ejection passage (27).
• Stripper springs (three coil springs 50/51 plus a weaker return spring 52; or, in the Fig. 11 variant, a single spring 94 with a weaker spring 95) that strip the punch, then withdraw the whole assembly so work can be inserted.
• A gage pin (85, 87‑89) limiting how far work enters the gap.
• A hold-down mechanism using pivotally connected links that the ram actuates before the punch, pressing the work down against the gage and removing camber.
• Multiple modular units mounted on a template (60) or base plate (102) or on a T-slotted rail (102/103/104) on a press brake bed, so a complete hole pattern is pierced in one stroke and setups can be pre-built off-press.

Independent claims — plain-language overview

• Claim 1 — The core structure: a holder with two spaced apart parts forming a work-receiving gap; axially aligned through-holes terminating at the gap; a die in one hole and a punch in the other; a transverse opening across the punch hole with a stripper spring in it acting on the punch.
• Claim 2 — Similar, but framed around a longitudinal punch hole plus the transverse opening, a stripper member, and a spring.
• Claim 13 — Work-positioning via a pair of pivotally connected links on the holder; a spring joins the links; the ram strikes the link pivot before it actuates the punch, driving the free link end against the work.
• Claim 14 — Holder mountable on a press bed with the ram able to operate the movable member, plus the link-and-spring work positioner and a roller on the free link end that rolls along the holder and the work.
• Claim 15 — A gage mounted near the gap, adjustable along the gap's length, extending past the gap's open end to engage a part of the work that runs at an angle to the portion in the gap (i.e., an angle/flange gage).
• Claim 16 — A gage whose sides straddle the holder part, with a connecting member the work rests against, adjustably secured to limit insertion depth.
• Claim 17 — Same general gage/holder/punch-and-die family (the OCR of this claim is truncated in the sources; I cannot reconstruct its full scope reliably).
• Claim 18 — A guide sleeve around the punch, movable into engagement with the work; a spring urging the punch into the sleeve, that spring travelling with the punch and sleeve and compressing as the punch penetrates; plus means yieldingly urging the assembly away from the work.
• Claim 19 — Sleeve + flange + pressure pad + stripper spring between them, with adjustable means to pre-compress the spring, and a yielding member to retract punch and sleeve after stripping.
• Claim 20 — Punch hole straight through the holder in line with the die, plus a second hole intersecting it at an angle carrying an actuating member connected to the punch and retractable out of the way so the punch can be slid out for sharpening or replacement.
• Claim 21 — Same as 20 but with the inclined-face cam joint between punch head and actuating member specified.
• Claim 22 — Starts a punch/die/holder combination with a spring arrangement (text truncated in the available sources).
• Claim 23 — A unitary punching unit for horizontal punching with vertical actuation: one-piece holder, upwardly extending parts, horizontal through-hole alined with the die, an upright intersecting hole, and an actuating member in it, with the punch removable past the upright hole.
• Claim 24 — Dependent on 23: adds stripping means arranged about the punch, with a row of springs retained in the holder opening.
• Claim 25 — Punching unit with a stripping member that has an opening the punch passes through, a strong spring between punch and stripping member to pull the punch back into the stripper, and a weaker spring reacting against the holder to hold the punch out of engagement with the work.
• Claim 26 — A unitary hole punching unit: one-piece housing, horizontal through-hole, spaced die, a vertically moving member that normally blocks the hole and can be moved clear so the punch slides out for sharpening/replacement.
• Claim 27 — Method: arrange the units on a base plate in the pattern of holes needed, secure them, punch the required number of pieces, then remove the units for reuse on another plate for a different job.
• Claim 28 — Apparatus: plural units mounted on a T-slotted rail on the bed of a press brake, with each unit's gap facing forward of the bed/ram for loading and its upright actuator near the rear for the ram; bolts in the T-slots allow lengthwise adjustment.
• Claim 29 — Dependent on 28: an adapter plate on the ram extending horizontally beyond the ram to reach the several units.
• Claim 30 — Method: mount one set of units on a template for one hole pattern and another set on a second template for a different pattern; run the first template in the press, swap in the second, then reclaim the units from the first template.
• Claim 31 — Apparatus: base plate + plural units + adjustable gages + individual pressure means on each holder yieldingly urging the work against the gage before piercing, so camber in the work is taken out.
• Claim 32 — Same, with the pressure means specified as an inclined link that in its inoperative position guides work into the gap and, when depressed, presses the work against the gage before and during punching.

Prosecution-family context (for grounding, not claim scope)

The patent has only one family member (the US filing itself); there is no foreign counterpart in the record. It has been cited as prior art in roughly ten later documents, including US 2,419,534 (Burleson/Mouldings Inc., tube punching), US 2,421,864 (Boeing Aircraft, "Angle punch unit"), US 3,120,778, US 4,486,929, and EP 0 267 115 (Ecological Systems "Piercing die," where US 2,329,020 appears as an "[AD]" cited document in the search report). Those citations corroborate the horizontal-punching, angle-member niche the patent occupies, but they are examiner/third-party citations, not admissions.

Where I am uncertain

• Assignee/ownership: no assignee of record; I cannot verify any corporate assignment from the patent documents themselves.
• Claims 3–12, and the tails of claims 17 and 22: the available OCR (both Google Patents and FreePatentsOnline) is corrupted or truncated. I have not reconstructed them, and I would not represent their scope without pulling the clean USPTO PDF/USPTO PatentCenter image file.
• I found zero litigation, PTAB, or CAFC records. Given the 1960 expiry, I consider that a reliable negative for CAFC purposes, but it reflects my search results rather than an exhaustive docket audit.

I'll search for litigation records involving US patent 2329020.

Let me search additional litigation-specific databases to be thorough.

Let me run a few final targeted checks against litigation-specific sources.

Litigation Search Report — US Patent 2,329,020

Patent identification (confirmed):

• US 2,329,020 — "Punching device"
• Inventor: George F. Wales
• Application No.: US 439383 (filed 1942‑04‑17); granted 1943‑09‑07
• Assignee: Individual (no corporate assignment of record)
• Anticipated expiration: 1960‑09‑07; legal status: Expired – Lifetime
• Source: https://patents.google.com/patent/US2329020/en

Findings: No litigation known

I found no record of any litigation involving US Patent 2,329,020. This includes no district court infringement actions, no declaratory judgment actions, no CAFC appeals, no PTAB/IPR/PGR proceedings, and no ITC Section 337 investigations.

Given the patent's age and status, this is unsurprising:

• The patent issued in 1943 and reached its anticipated expiration on 1960‑09‑07, meaning it was enforceable for only a ~17‑year term ending over 65 years ago.
• Records for patent suits from the 1940s–1950s are not comprehensively digitized in the modern databases; free litigation portals (Unified Patents, CourtListener, Docket Alarm, Justia, etc.) generally cover from the 2000s forward. Even in the pre-digital era, a World War II–vintage, individually-owned sheet-metal punch tool patent is an unlikely subject of reported litigation.

What I searched and what the sources showed

Important note on false positives (do not confuse these)

Several distinct documents share the number "2329020" and are not US 2,329,020:

• EP 2329020 B1 / EE‑EP 2329020 B1 / PT‑2329020‑E — Novartis AG biological patent ("Presentation on the cellular surface of polypeptide isoforms through read-through of termination codons")
• CN 2329020 Y — Chinese utility model for a photographic-paper shifting mechanism
• RU 2329020 C1 — Russian patent (cited as prior art in a 2008 search report)
• GB 2329020 A/B — UK patent in the EP 0901229 family (electrically operable device operation)
• CourtListener record 2329020 — a judicial financial-disclosure investment entry

None of these relate to George F. Wales's punching device, and none indicate litigation involving US 2,329,020.

For each requested field

Because no case exists:

• Plaintiff(s): None known
• Defendant(s): None known
• Jurisdiction: N/A
• Case number: N/A
• Filing date: N/A
• Outcome/status: N/A

Caveats / confidence

• High confidence: US 2,329,020 was a short-lived (1943–1960), individually-owned mechanical patent; no litigation is indexed in the modern free databases I could access.
• Limitation: I could not exhaustively search paywalled or pre-2000 case archives (e.g., full PACER historical records, print-era CCH/USPQ reports), and the tool I used has limited coverage of earlier decades. If a suit was filed and dismissed in the 1940s–50s, it may simply never have been digitized. If a definitive negative is required, a manual search of historical runs of the United States Patents Quarterly and regional Federal Reporter volumes, or a commercial service such as Docket Navigator's full historical coverage, would be needed.

If a claim is being made that this specific patent was asserted in litigation, please share the case number or court, and I can attempt to verify it — but nothing in the available sources supports that.

PTAB Activity Report — US 2,329,020 (“Punching device,” George F. Wales)

Bottom line up front: there are no AIA trial proceedings on this patent. None. Not "none found" — the canonical structured list is empty, and nothing surfaced on web search.

---

Proceedings overview

Total AIA trial proceedings on file: 0 (0 IPR, 0 PGR, 0 CBM). Breakdown by status: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

There is no IPR-based defensive posture to report because there is nothing to report — and, more importantly, there is no defensible posture for a patent owner either. US 2,329,020 issued 1943-09-07, was filed 1942-04-17, and its term expired 1960-09-07 (Google Patents records the anticipated-expiration event of 1960-09-07 and a legal status of "Expired – Lifetime"). The patent is in the public domain and has been for roughly 66 years. The PTAB itself did not exist until the AIA (2012), so the entire AIA-trial regime postdates this patent's expiration by half a century. If you are a defendant and someone's demand letter cites US 2,329,020, the factual predicate of that letter is false — an expired patent cannot be infringed, and no damages claim can rest on it. Treat that letter as a serious credibility signal about the sender.

Proceedings table (canonical list)

The structured "PTAB proceedings on file" block sourced from the USPTO Open Data Portal returns no AIA trial proceedings for this patent as of the most recent ingest, and I found nothing to contradict that. No proceeding numbers are listed here because no non-fabricated ones exist.

---

Why the absence here is not the usual "signal"

In an ordinary report, a zero-proceeding patent is a yellow flag: well-asserted, commercially meaningful patents eventually attract IPR petitions, so an empty docket can mean the patent simply hasn't been monetized enough yet. That inference does not apply here. Four structural reasons:

1. The patent expired 1960-09-07. Enforceable term ended before the Federal Circuit existed in its current form, before the AIA created IPR/PGR/CBM, and before § 315(e)(2) estoppel had any meaning. There is no live right to challenge.
2. No PGR or CBM path ever existed. PGR applies only to patents with an effective filing date on or after 2013-03-16 (AIA first-inventor-to-file). CBM applies to covered business-method patents — a mechanical punch-and-die fixture classified in B21D 28/30 (perforating annular parts) is not a covered business method under § 18(d)(1) by any reading.
3. IPR would be technically available but pointless. An IPR can be filed against an expired patent (the Board has addressed expired-claim IPRs), but the only realistic reason to do so — freeing yourself from an infringement claim — cannot arise. The record contains no petition.
4. The "Cited By" and "Similar Documents" lists are not PTAB records. Google Patents shows 10–11 later patents citing 2,329,020 (e.g., US 2,421,864 "Angle punch unit" to Boeing Aircraft Co., filed 1945-09-04; US 2,419,534 to Mouldings Inc.), and the "Similar Documents" list is a citation/classification-similarity artifact (US 2,363,431, US 2,365,344, US 2,325,990, etc.). These are prior-art citations of the '020 patent by later patents, which is the opposite relationship from an AIA proceeding. Do not mistake them for challenges.

---

Per-proceeding analysis

There are no proceedings to analyze. I am deliberately not supplying the template fields (judge panel, petition grounds, institution decision, FWD, settlement, appeal) because populating them would require inventing a proceeding number and a record, which the constraints prohibit.

---

Strategic summary

Claim status: all claims UNTESTED, and all claims EXPIRED. Independent claims 1, 14, 15, 16, 18, 19, 20, 21, 22, 23, 25, 26, 28, 31 and 32, plus their dependents (including claim 24 and claim 29), were never subjected to an AIA trial. No claim was canceled, no claim was sustained. The distinction "canceled vs. sustained vs. untested" collapses here because the claims left the enforceable set through ordinary statutory expiration in 1960-09-07, not through adjudication. Note also the odd claim structure visible in the printed text — claim 27 and claim 30 are method claims directed to arranging multiple punching units on a base plate/template for press-brake operation, and claims 28–29 and 31–32 are apparatus/system claims to the multi-unit press-brake setup — none of which changes the expiration analysis.

Estoppel landscape: inapplicable. Section 315(e)(2) estoppel attaches only to a petitioner that obtains a final written decision. With zero petitions and zero FWDs, there is no petitioner, no privy, and no estopped ground. That means every § 102/§ 103 ground that the patent's own prosecution history left open is technically "available" — but there is nothing to assert them against. Practically, the § 315(e)(2) question is moot: you cannot be barred from challenging claims that are no longer enforceable, and you have no reason to spend money challenging them.

Pattern signals: none. No repeat petitioner (there is no petitioner), no PTAB appeal history, no defensive aggregator such as Unified Patents in the chain. A search of PTAB E2E / PTAB Decisions and CourtListener-style full-text queries for "2,329,020" / "2329020" returns no Board or Federal Circuit activity. If you want to double-check the negative independently, query the USPTO PTAB E2E search at https://ptacts.uspto.gov/ptabweb#/search on patent number 2329020, the PTAB Decisions page, and CourtListener (https://www.courtlistener.com/?q=%222%2C329%2C020%22) for the patent number as a docket string. I ran a web search and it surfaced only the patent's own publication records (Google Patents https://patents.google.com/patent/US2329020/en, FreePatentsOnline https://www.freepatentsonline.com/2329020.html) and unrelated hits on higher-numbered '232 patents — no PTAB or CAFC material.

---

Recommended next steps

• If you are a defendant and a demand letter or complaint cites US 2,329,020: do not negotiate. Pull the Google Patents record linked above, which shows the 1960-09-07 anticipated expiration and "Expired – Lifetime" status, and quote it back. An expired patent supports no infringement theory, so any § 271 claim premised on it is subject to a Rule 11 / § 285 exposure argument. There is no FWD to link to here — the disposal mechanism was expiration, not adjudication.
• If you are evaluating a portfolio that lists 2,329,020: treat it as a historical/documentary asset only. Its value is as prior art (it is a 1943 publication and is being cited by later patents), not as an assertion vehicle. Its § 102(b) date of 1943-09-07 makes it useful defensively against later punch-unit patents — check the Cited By list (US 2,421,864, US 2,419,534, etc.) if you are defending a modern punch/die case and need an anticipatory or obviousness reference.
• If someone tells you there are PTAB proceedings on this patent: ask for the proceeding number and verify it in PTAB E2E before relying on it. I found none, and the ODP structured data shows none.
• One accuracy caveat I want to be explicit about: the source snippet for the "Similar Documents" table contains a line reading "US2323431A … 1943-07-06 … Punching device," which is a different patent from a different inventor and is easy to confuse with 2,329,020 in a text-scan. Likewise, search noise around "HTS 9403.99.9020" (wire decking) and "GB 2329020 A" (a UK patent application on an electrically operable device) are unrelated documents that share the digits 2329020. None of these is a PTAB proceeding on US 2,329,020.

I'll research the assignment/ownership record for this patent before writing anything.

Assignment / Ownership Reconstruction — US 2,329,020 ("Punching device," George F. Wales)

Headline finding (stated up front, as required): The USPTO Assignment Center / Patent Assignment Search returns no recorded assignment records for US 2,329,020, and Google Patents' Legal Events table for the patent (as fetched) shows only Application filed (1942‑04‑17) → Priority to US439383A → Application granted (1943‑09‑07) → Publication of US2329020A → Anticipated expiration (1960‑09‑07), with no "Assigned to" event.

That is not merely an empty search. There is a structural reason the online record is empty: the USPTO's searchable assignment database (now Assignment Center, formerly EPAS; bulk XML backfile) only contains recordations from August 1980 forward (see the USPTO's own Assignment Database webinar material: "This searchable database contains all recorded Patent Assignment information back to August 1980," https://www.uspto.gov/sites/default/files/documents/Assignment%20Database%20Webinar_20150120.pdf, and the USPTO bulk-data catalog entry "Patent Assignment XML (1980 – Present) … backfile – August 1980 – December 31, 2024," https://catalog.data.gov/dataset/patent-assignment-xml-1980-present). US 2,329,020 issued 1943 and expired 1960 — twenty years before the online database begins. Any assignment that was ever recorded for this patent would exist only as a pre‑1980 paper/microfilm recordation in the USPTO's Assignment Recordation Branch, retrievable today via a paid Assignment Abstract of Title order or a FOIA request, not via the public web search.

Consequently I am not going to populate reel/frame fields, correspondents, or conveyance types. I have none, and inventing plausible-looking reel numbers for a 1943 patent would be the worst possible failure mode in this task.

---

Inventors

Unusual patterns: none of the classic fire-sale precursors apply.

• Sole inventor, single-patent family — no co-inventor departures to track, so the "all inventors left within 12 months" tell is structurally inapplicable.
• The one genuinely notable pattern is the assignee field itself. The patent names no assignee ("Individual" per Google Patents), yet Wales's later patents are assigned to Wales Strippit Corp. on their faces — e.g., US 2,720,264 ("Apparatus for blanking and perforating work-pieces," Wales + Paul H. Taylor, filed 1950‑09‑19, assignee Wales Strippit Corp.), US 2,778,426 ("Perforating apparatus," filed 1953‑09‑04, assignee Wales Strippit Corp.), US 2,815,814 and US 2,837,161 (both filed 1955‑04‑14, assignee Wales Strippit Corp.). By contrast, Wales's 1938–1944 filings — including this one, plus US 2,163,641 (1938), US 2,317,424 (filed 1942‑03‑12), US 2,355,344 (filed 1942‑01‑12) and US 2,363,431 (filed 1942‑05‑12) — all issue in his personal name.
• Read carefully, this suggests the corporate assignee did not yet exist (or was not yet being used as assignee) when '020 was filed in April 1942, and that rights to the Wales punch-tooling portfolio may have been personally held at that time and thereafter conveyed to the corporation. I cannot verify that from any record. It is an inference from the contrast between the 1942–1944 patents (individual) and the 1950s patents (corporate assignee), not a documented assignment.

---

Original assignee

None. No assignee is named on the issued patent. Google Patents lists both "Original Assignee" and "Current Assignee" as Individual — i.e., the patent issued to George F. Wales personally, unassigned.

• Product embodying the claims: The patent is a component/tooling patent (a horizontal-punching unit for press brakes), not a consumer product. Background context — the relevant commercial vehicle was Wales's own company. That company did build and sell the machinery line this disclosure belongs to (the later Wales Strippit Corp. patents cover perforating implements, punches, dies and stripping media for exactly this machine family, e.g. US 2,740,475 "Perforating apparatus having a stripping medium mounted within a chamber in the punch member"). I cannot confirm from any assignment record that the company ever held title to '020.
• Primary line of business: sheet-metal fabricating machinery and punch/die tooling, Buffalo, N.Y.
• Current status: Undetermined. The corporate genealogy (Wales‑Strippit → Strippit → later industry consolidation of the Strippit name) is background knowledge, and I am deliberately not asserting specific acquisition dates or acquirers because I could not verify them against a primary source in this session. What is verifiable: the corporate name existed by June 1945; it was an operating manufacturer, not a licensing vehicle; and it was not an assignee of '020 on the patent's face. Whatever happened to the company afterwards is irrelevant to the '020 right, which expired 1960‑09‑07.

---

Assignment timeline

No recorded assignment can be retrieved for US 2,329,020.

• USPTO Assignment Center (https://assignmentcenter.uspto.gov/) — no records returned for patent 2329020.
• Legacy assignment search (https://assignment.uspto.gov/patent/index.html) — same underlying data, same result.
• Google Patents Legal Events — no assignment conveyance listed; sole ownership event of note is the anticipated expiration of 1960‑09‑07.

This is not the same as "there was no assignment." It means: no post‑August‑1980 recordation exists (expected, since the patent left the enforceable set in 1960), and pre‑1980 recordations are outside the online database entirely. For a definitive pre‑1980 abstract of title, the routes are:

1. Order an Assignment Abstract of Title from the USPTO Assignment Recordation Branch (571‑272‑3350 / AssignmentCenter@uspto.gov), which searches the full recorded history including microfilm-era records; or
2. File a FOIA request for the recorded documents on reel/frame for patent 2,329,020; or
3. Consult the print-era Official Gazette assignment notices and the USPTO's published assignment digests / microfilm assignment reels (available at a Patent and Trademark Resource Center) for 1943–1960 conveyances.

Because there are no records to itemize, the per-assignment fields (reel/frame, conveyance type, assignor, assignee, correspondent) are left empty rather than guessed. There is also no correspondent of record to analyze — meaning the "repeat-player attorney" tell, the single most useful signal in this kind of investigation, is simply not available for this patent.

---

Timeline diagram

timeline
    title Ownership of US 2329020
    1942 : Application filed by George F Wales
         : No assignee named
    1943 : Patent issued to Wales personally
         : Patent number 2329020
    1960 : Patent term expires
    2026 : No online assignment records found
         : Database coverage begins 1980

---

NPE / troll-pattern signals

Explicit non-findings (guardrails). There is no Marathon Patent Group reel, no shell LLC, and no repeat correspondent to report — and I decline to construct the appearance of one. Notably, the word "MARATHON" that appears in one of my search results is a Mate Precision Tooling punch-holder product line in a 2009 tooling catalog (https://mecos.es/files/productos/punzonadora/mate-wiedemann.pdf), not Marathon Patent Group, and it has nothing to do with this patent. Similarly irrelevant to ownership: EP 2329020 B1 / PT‑2329020‑E (Novartis), RU 2329020 C1 (Russian ophthalmic laser), CN 2329020 Y, GB 2329020 A/B, and a CourtListener financial-disclosure record ID 2329020. None is US 2,329,020.

---

Verdict

Insufficient data (no records).

The Assignment Center returns zero recorded assignments for US 2,329,020, and the patent issued with no assignee named — George F. Wales held it personally, per the "Original Assignee: Individual" entry on the issued patent and the absence of any "Assigned to" Legal Event on Google Patents. No reel/frame, no assignor, no assignee, and no correspondent exists to analyze, so not one of the eight NPE signals can be affirmatively established; I rate them "not present" only in the sense that no evidence supports them, which is materially different from evidence of absence. The decisive context is that the USPTO's online assignment corpus begins August 1980, while this patent expired 1960‑09‑07 — so the negative here reflects a database coverage boundary, not a demonstrated clean chain, and a definitive answer requires a manual Assignment Abstract of Title order or a FOIA request for the pre‑1980 recordation.

Verify (and, if you need the pre-1980 record, escalate from here):

• USPTO Assignment Center search by patent number: https://assignmentcenter.uspto.gov/
• Legacy assignment search (same data): https://assignment.uspto.gov/patent/index.html
• Patent record of reference: https://patents.google.com/patent/US2329020/en
• Manual pre-1980 route: USPTO Assignment Recordation Branch — 571‑272‑3350 / AssignmentCenter@uspto.gov (Assignment Abstract of Title, or FOIA for the recorded document).

Confidence: High that no online assignment record exists for this patent. High that the 1980 coverage floor explains it in material part. Low-to-none as to whether any pre‑1980 paper assignment (for example, a personal-to-Wales‑Strippit conveyance, which the assignee-field contrast between his 1942–44 and 1950s patents makes plausible) was ever recorded — that question is genuinely open and cannot be closed with the tools used here.

I'll start by pulling the authoritative USPTO/patent-office records for this specific number and checking whether the patent's own front page lists cited references (as opposed to later patents that cite it).

The searches surfaced a critical distinction I need to nail down: the "Cited By" list on Google Patents for this patent is a forward citation list (later patents citing the '020 patent), not the prior art cited against it during prosecution. Let me verify whether the patent's own front page carries a "References Cited" list.

Prior-Art Analysis — US Patent No. 2,329,020 ("Punching device," George F. Wales)

0. What I searched, and one correction to the question as posed

I searched for the literal number 2329020 / US 2,329,020 against the accessible patent registries (Google Patents, FreePatentsOnline, EPO Global Patent Index), and I ignored near-miss numbers (GB 2329020, RU 2329020, PT‑2329020‑E, CN 2329020 Y, EP 2329020, CourtListener record 2329020 — none of which are this patent).

Correction to the premise. The task asks me to "look at each patent citation for 2329020" and state which claims each "potentially anticipates under 35 U.S.C. § 102." Two problems:

1. There is no backward citation list on this patent. The patent issued 1943‑09‑07. The USPTO did not print a "References Cited in the file of this patent" block on the face of patents until roughly the late 1940s; the 1943 '020 patent therefore carries no front‑page reference list, and neither Google Patents nor FreePatentsOnline renders a "Patent Citations" block for it. (Confidence: moderate‑high on the dating convention; high on the factual absence of the block in every record I retrieved.) The only citation data attached to this number is forward — later patents that cite '020. The Google Patents page shows only "Cited By" and "Similar Documents": https://patents.google.com/patent/US2329020/en
2. The available citations cannot anticipate '020 as a matter of law. Every document in the "Cited By" list postdates '020's 1942‑04‑17 filing. A reference published or filed after the applicant's filing date cannot be § 102(a)/(b) prior art against that applicant's claims. So the honest answer to the literal question, reference by reference, is "none — none of these can anticipate any claim of US 2,329,020." I give the table and the reasoning below, then flip the analysis to the direction that is legally meaningful: where '020 itself stands as prior art.

(Minor metadata note: the system clock reads 2026‑09‑27; the task header says April 26, 2026. Not material to this analysis, but flagging the inconsistency.)

---

1. The only citations of record for 2329020 are forward citations

Table 1 reproduces the "Cited By" list exactly as it appears on the authoritative record (dates as given there; Google Patents reports priority date and publication date). These are later documents that cite '020 as background art.

Two independent corroborations that these are forward citations (i.e., '020 cited by them, not against it):

• The EPO Global Patent Index search report for EP 0 267 115 lists "[AD] US 2329020 A 19430907 – WALES GEORGE F" — the "AD" tag denotes background/auxiliary art cited in the search report: http://data.epo.org/gpi/EP0267115A3-PIERCING-DIE.html
• The FreePatentsOnline page for US 2,875,829 (Patrick) prints, under "US Patent References": "2329020 | Punching device | 1943‑09‑07": https://www.freepatentsonline.com/[2875829](/patent/2875829).html

Why none of them can anticipate a claim of '020 (the § 102 mechanics)

A further framing caveat worth stating plainly: 35 U.S.C. § 102 did not exist when '020 was examined. The statute as we use it came from the 1952 Act (effective 1953‑01‑01). In 1942‑43 the novelty provisions were R.S. § 4886 / § 4920. A modern invalidity analysis of '020 would apply pre‑AIA § 102 by convention, but it is an anachronistic frame for the original prosecution.

---

2. What is not a citation (do not treat these as § 102 art)

The Google Patents record also shows a "Similar Documents" table — US 2,363,431; US 2,163,641; US 2,355,344; US 2,317,424; US 2,410,372; US 2,379,271; US 2,708,970; GB 965,521; US 2,325,990; US 1,715,138; US 2,732,015; US 2,323,431; US 2,325,290; JP H07185692 A; US 2,367,490; US 1,570,994; US 2,326,588; US 2,350,436; US 2,320,205; US 1,964,752; US 3,029,677; US 2,013,976; US 2,341,131; US 2,378,603; US 2,575,945. These are machine-generated similarity/classification neighbors, not citations and not examiner art. They carry no § 102 weight on their own.

That said, several of them do predate the 1942‑04‑17 filing and are legitimate candidates for a freelance anticipation search (I am labelling these as candidates only, because nothing in the record shows they were before the examiner):

---

3. The legally meaningful direction: '020 as § 102 art against these later patents

Because '020 published 1943‑09‑07, it is a § 102(b)-type printed publication/patent for every document filed after 1944‑09‑07, and a § 102(a)/§ 102(e) reference for earlier-filed, later-issued ones. Where the subject matter overlaps, '020 is a plausible anticipation or obviousness reference. The strongest overlaps, given '020's disclosure (horizontal punch and die on opposed horns with a work gap, vertical ram converted by a cam into horizontal punch motion, spring stripper sleeve, threaded adjustable die, slug‑ejection passage, gage, template/base‑plate modularity):

This is the analysis a defendant in a modern punch‑unit case would actually want — i.e., using '020 defensively. I am not asserting that any of these were actually rejected over '020; the record does not show the examiners' grounds, and I will not invent them.

---

4. Bottom line

• Prior art cited against '020: none retrievable. The 1943 patent has no printed "References Cited" front‑page block, and neither Google Patents nor FreePatentsOnline carries a backward‑citation table for US 2,329,020. The file wrapper (USPTO PatentCenter / NARA hard copy) is the only place the examiner's actual citations would live, and I could not retrieve it in this pass.
• The 11 "Cited By" patents (Table 1) are all later than '020 and therefore anticipate nothing in '020. For each, the § 102 answer is the same: no claim of US 2,329,020 can be anticipated by a post‑1942 reference.
• The useful § 102 relationship runs the other way: '020 (published 1943‑09‑07) is prior art against the later punch‑unit patents, with the closest overlaps being US 2,421,864 and the Ecological Systems "Piercing die" family (US 4,802,393 / EP 0 267 115), both of which expressly cite it.
• "Similar Documents" are not citations and were excluded from the anticipation analysis; a shortlist of pre‑filing candidates among them is given for a proper search, clearly flagged as unverified provenance.

If a definitive § 102 record is required

1. Pull the file wrapper for application 439,383 (USPTO PatentCenter; the printed patent itself will have no reference list).
2. Run a pre‑1941‑04‑17 search in CPC B21D 28/24 / 28/30 and legacy USPC class 83 (punch/stripper subclasses) and class 29 (Y10T29/49897 "registering mating opposed tool parts").
3. Search the inventor's own prior art: George F. Wales's earlier patents (e.g., US 1,812,468 and US 1,811,987, both 1931, per classification indexes) are self‑art candidates under § 102(a)/(b).

Confidence: high that no backward citations exist in the accessible digital records and that the Table 1 references cannot anticipate '020; moderate on the ~late‑1940s USPTO practice of printing front‑page references (it explains the absence but I did not verify it against a primary USPTO source in this session).

Sources: https://patents.google.com/patent/US2329020/en · https://www.freepatentsonline.com/2329020.html · http://data.epo.org/gpi/EP0267115A3-PIERCING-DIE.html · https://www.freepatentsonline.com/2875829.html · https://www.freepatentsonline.com/[2882971](/patent/2882971).html

I'll research the prior art landscape around this 1942 filing before running the § 103 analysis, since the "Prior Art" material on the Google Patents page needs careful date-filtering.

§ 103 Obviousness Analysis — US 2,329,020 ("Punching device," George F. Wales)

Analysis date: 2026-04-26 · Patent filed: 1942-04-17 · Issued: 1943-09-07 · Expired: 1960-09-07
Authority: Pre‑AIA 35 U.S.C. § 103 (the 1952 Act's § 103 is the retrospective vehicle for evaluating a 1942 invention; the substantive test is Graham v. John Deere, 383 U.S. 1 (1966), as later refined by KSR Int'l v. Teleflex, 550 U.S. 398 (2007)).

---

0. Framing, and three corrections to the record as previously generated

Three flags before the analysis begins:

1. Claim-text gap (material to this task). The OCR of the patent as supplied contains no text whatever for claims 3–12, and the text of claim 13 is truncated to a fragment ("…along said holder into and out of position to engage the work, a spring connecting said links…") that actually matches the tail of the later-printed claim 14. Likewise claim 17 is truncated and claim 22's tail is interleaved with claim 23's preamble in the OCR. I therefore cannot perform a claim-by-claim § 103 analysis on claims 3–12, 17 or 22. I say so in the table rather than guessing.
2. Claim-independence inconsistency between my two prior sections. The "Patent summary" listed claims 2, 13, 17, 27, 30 as independent; the "PTAB challenges" section's list omitted 2, 13, 17, 27, 30. On the face of the printed text, claims 2, 13, 17, 27 and 30 each begin with an independent-form preamble ("In a punching device, the combination of…"; "A method of punching…"), so the broader list is correct and the PTAB-section list was incomplete. This matters because claims 27 and 30 are the method claims that carry the whole "modular unit" commercial concept, and they were the ones missing from that list.
3. The "Prior Art" material on the patent page is not what its label suggests. The block headed "Prior art keywords: punch, work, die, holder, gap" is a keyword tag cloud, not a reference list. The "Cited By" tables are forward citations and every one of them post-dates the '020 filing — they are legally incapable of being § 102/§ 103 prior art against '020. The "Similar Documents" table is a citation/classification-similarity artifact, only part of which predates the critical date. The "Definitions" block is NLP/OCR noise (it literally contains "venting," "rolling process," and a scrap of an unrelated benzamide chemical structure), and must not be treated as disclosure.

Consequence: the record as supplied supplies no anticipation of the core claims, and the genuinely operative § 103 art has to be assembled from the inventor's own earlier patents (which are § 102(b) art — see § 2.1) plus the pre-1942 printed publications identified below. That is what this report does.

Practical value caveat: because the claims expired 1960-09-07 and no litigation or PTAB proceeding has ever attached to them (per the two prior sections), this analysis is defensive/prior-art value and historical validity assessment, not a live freedom-to-operate or damages question. It is nonetheless the right frame for anyone using '020 as a § 102(b) reference against a later patent, or evaluating a demand letter that wrongly cites it.

---

1. Critical dates and the statutory filter

Why the "by others" distinction is the hinge of this case. Almost all the closest art is Wales's own. Under § 102(a), a reference is not prior art unless it is "by another." But under § 102(b), there is no "by another" requirement, and an inventor's own patent issued more than one year before his filing is a statutory bar. See MPEP § 2133.01; In re Katz, 687 F.2d 450 (CCPA 1982). And crucially, pre-AIA § 103(c) (which can disqualify commonly-owned § 102(e)/(f)/(g) art) has no application to § 102(b) art, so common inventorship does not rescue Wales's own 1934–1939 patents.

Therefore the following are § 103 prior art against '020, notwithstanding that they name the same inventor:

And these are expressly NOT § 103 art — flag this because they are the most tempting to misuse:

Analyst note on a common error: the "Similar Documents" table on the patent page lists US 2,323,431 — "Punching device" — 1943-07-06, which is a different patent by the same inventor and is easy to confuse with '020 in a text scan. It is not prior art to '020 (later-filed, same inventor), and it is not a § 103 reference.

---

2. The operative prior art, and what each reference actually teaches

2.1 US 2,013,976 (Wales, 1935-09-10) — the gap-holder / C-frame architecture

This is the single most damaging reference, because it discloses almost the entire holder architecture of claims 1, 2, 15–17, 23, 26 and 28 — with the axis rotated 90°.

From the record:

• A holder comprising a lower base 11 and an upper head 22, "separated at their front opposing ends by an intervening gap 23, while their rear ends are held apart from each other by means consisting preferably of a foot 25… These two members of the holder are preferably detachably connected with each other and held against relative movement in a horizontal direction by means of screws 24 … and also by means of dowels 26."
• A die member 31 seated in a rabbet with a dowel pin 29, having a die opening 32 whose "vertical centerline… is the vertical axis of its cooperating punch," plus a discharge chute 35 for slugs that "extends at an angle toward the rear of the base."
• A taper-headed clamp screw 39 ("serves to clamp the die member firmly to the holder and also to locate said die member so that the die opening 32 will always line up directly on the axis of the punch") — i.e., a removable, alignable, adjustable die, the functional equivalent of '020's threaded die 25/30 with lock nut 32.
• A "vertical guideway 42 of circular form which is axially in line with the die opening 32" in the head 22, in which "a punching unit … is capable of sliding vertically … toward and from the die member 31."
• A shoe 10 with a T-shaped head and clamp 37/36, secured by clamping screw 16, with an optional calibrated scale 21 and a template and nut lock (Fig. 7), plus "a self-contained punch and stripper which can be readily removed from the holder without disturbing any other part of the set-up."

What it does not have: the punch axis is vertical, not horizontal; there is no actuator-pin/wedge; and the gap opens front-to-rear rather than upwardly.

2.2 US 2,168,377 (Wales, 1939) — the stripper sleeve + spring, and the gage

From the record, this teaches the exact stripper mechanism of claims 1, 2, 18, 19, 25 and 26:

• "During the initial part of the downward stroke of the ram the stripper and punch move in unison therewith until the lower ends of the punch and stripper engage with the upper side of the work-piece. Thereafter the punch continues its downward movement independently of the stripper … the spring 41 is compressed between the collar or flange 35 of the punch and the stripper … During the first part of the subsequent upward movement of the ram … the stripper remains at rest on the work sheet … thereby causing the sheet of material which has been perforated to be pushed or stripped off from the tip of the punch."
• A guide 40 that both guides and strips — the functional twin of '020's guide sleeve 42.
• A punch collar/flange 35 cooperating with the stripper shoulders — the twin of '020's lock nut 23 bearing on pressure pad 48.
• A discharge opening 42 from the female die 40 downward and outwardly to the side of the head for slugs.
• Gages 45, each "comprising an angular structure having a horizontal arm 46, slotted as at 47, and a vertical arm 48," bolted via T-shaped retainer grooves and adjustable along them.
• And, for claim 20/24/26 removability: the express teaching of removing the stripper spring, raising the guide, and using the punch bit itself as a pilot to align punch and die, then reassembling — i.e., the punch is serviceable and removable without disturbing the alignment.

2.3 US 2,163,641 (Wales, 1939-06-27) — the press-brake rail, back gage, lifter spring, and slug handling

Directly on point for claims 24, 27, 28, 29 and 31:

• "a rail on which the die holders of the perforating apparatus are mounted together with means for attaching the holders to said rail, to permit a quick adjustment of the holders on said rail and provide clearance in the rail for removal of the slugs"; "the two sections of the rail are connected with each other at intervals by transverse tie bars… and thus hold the rail against lateral displacement on the bed"; "clamping screws… working in the front part of the bed and bearing against the adjacent tie bars of the rail."
• "the shoes of a plurality of die and punch holders with calibrated means for locating the several holders in proper relation to each other, together with a gage bar having proper markings for making accurate adjustments of holders."
• "an improved adjustable back gage combined as a unit with the holder." → the gage integrated into the holder is exactly '020's claims 15/16/31 approach.
• "a punch member which can be readily removed from its position in the punch guide and can be replaced into the guide, without the stripper spring in position… an improved punch, punch guide, and stripper spring which can be readily removed from the holder without disturbing any other part of the set up." → claim 24 verbatim in substance.
• "an improved lifter spring for raising the punch guide a suitable distance upward from the die to permit insertion of the parts to be perforated." → this is precisely the second, weaker spring function of '020 claims 19 and 25 (a spring that holds the assembly clear of the work so work can be inserted).
• Holder 15 with a lower base/shoe 151 and an upper head or horn 152 — the same two-horn geometry — with a groove fitting a tongue on the rail; and the whole thing operated on a press brake on a Z-bar.

2.4 US 1,964,752 (1934-07-03) — "Apparatus for perforating sheet material"

(Inventor not confirmed from the retrieved record — verify before formal citation. It appears in the '020 Google Patents page's "Similar Documents" table and in the citation list of Wales's US 2,317,424, and reads in Wales's style.)

Teaches a frame independent of and removable into and out of the space between press bed and ram, comprising:

• a lower bridge-shaped shoe with legs resting on the bed and a clearance space 25 beneath the die plate for slugs;
• an upper holder plate 26 with legs 27 so that a gap 23 is formed between the lower base plate and the upper holder plate which extends from the front to the rear of the frame … into which the sheets of material to be perforated are introduced;
• the two plates "rigidly but detachably connected with each other by means of a plurality of screws 29 … and a plurality of dowels 30";
• die buttons 36 seated in sockets 38, with a discharge passage 34; a vertical guide opening 40 in the head with a punching unit slidable in it;
• a gage pin slidable vertically in the corresponding openings 60, 61;
• and — critically — the work is pushed into the gap, where "the front edge of the sheet by engaging with the convex underside of the supporting [sleeve]…" lifts a floating punching unit, the preferred construction including "a spring interposed between said disk and follower."

2.5 The 1919 Iron Age printed publication — multi-unit setups on templets

From the retrieved text of Sim Chilton's Iron Age, 1919-07-03 (https://archive.org/download/sim_chiltons-iron-age_1919-07-03_104_1/sim_chiltons-iron-age_1919-07-03_104_1.pdf):

• A Thomas Spacing Machine Co. punch "arrangement of punches making four tools which move symmetrically… It has sliding punches which can be made to move in and out from the center of the ram."
• "punching to templet and indicator. In this case, the plate is not marked, and the spacing lengthwise is obtained from a wood templet traveling with the plate…"
• "The punches and dies are kept in alignment by levers… The strippers are arranged in the same manner as on the ordinary multiple punch…"
• "The movement of the plate is controlled by travelling carriage, to which the plate is attached by means of a clamp."

This is a printed publication more than 22 years before the critical date, establishing that setting up multiple punches in a pattern defined by a templet, with clamping, stripping and alignment means, was notorious.

2.6 The 1934 Prentice record (Canada S.C. / U.K. Privy Council) — "customary in all punch press work"

From the retrieved factum (http://knyvet.bailii.org/uk/cases/UKPC/1934/1934_29(image4).pdf): "Such an alteration is customary in all punch press work — for each shape of punching to be turned out. The forming and punching mechanisms involved nothing but standard, routine, punch press practice…" The record also documents the notorious use of "stationary inclined planes" to convert motion/apply clamping force in punch presses.

This is exactly the evidentiary posture KSR endorses for a § 103 case: an admitted, printed, contemporaneous statement that reconfiguring a punch press for a new workpiece geometry was routine shop practice, plus documented use of inclined planes as motion-converting elements.

2.7 Field evidence from '020's own classifications (not prior art, but probative of a crowded, predictable art)

The examiner placed '020 in B21D 28/24 (perforating) and B21D 28/30 ("perforating… in annular parts, e.g. rims"), and the cross-reference collections include:

• Y10T 83/384 — "By tool inside hollow work"
• Y10T 83/0596 — "Cutting wall of hollow work"
• Y10T 83/395 / 83/40 / 83/401 — "One tool having only rectilinear motion"; "Multiple external active tools"; "Tools operate in a substantially common transverse plane of cut"
• Y10T 83/2135–2159 — "Moving stripper timed with tool stroke"; "Carried by moving tool element or its support"; "Stripper biased against product"; "By spring means"
• Y10T 29/49897 — "Registering mating opposed tool parts [e.g., registering a punch and a cooperating die]"

Those Y10T buckets are the PTO's own classification of the art and are useful evidence that (i) horizontal/inside-hollow-work punching and (ii) spring-biased strippers carried by the moving tool were established, densely populated sub-arts before 1942 — i.e., a finite, well-mapped set of predictable solutions for the problem '020 addresses.

---

3. The person of ordinary skill in the art (POSITA)

A press-tool/punch-and-die designer with several years of experience in sheet-metal perforating unit design as of 1942 — someone who would have known Wales's own Walsco/Wales-Strippit-style self-contained perforating units, press-brake rail mountings, and the standard stripping-spring design of US 2,168,377. This is not a PhD-level artisan; it is a skilled tool designer, which lowers the § 103 threshold and makes KSR's "combination of familiar elements according to known methods" and "obvious to try" rationales directly applicable.

---

4. The problem, stated as the art would state it

As '020's own specification concedes: "The punching of holes in sheet metal and other materials is generally accomplished by means of a vertically movable punch… There are times when the work to be operated on is of such shape or form that it is either impossible or difficult to operate on the same by means of a vertically operable punch."

Stated that way, the inventive problem is a work-access problem: the web/flange of an angle, rim, channel or Z-bar obstructs a vertical punch. The art as of 1942 had already solved the work-access problem twice over:

• by the two-part gap holder ('976, '752, '641) — the work enters a gap between a base and a head, so awkward shapes can be presented to the tools; and
• by re-indexing the tool axis — '020 itself says the punch/die "may be arranged to operate in any other desired angular relation to the vertical," and its own sibling patent US 2,323,431 expressly discloses adapters that let the holder "swing relatively to the adapter on an axis which coincides with the axis of the punch 31" "if it is desired to arrange the holes in different angular relations to the work." (Not art, but it is contemporaneous, same-inventor corroboration that swinging/rotating the tool axis to suit the work was understood as an ordinary design variable.)

Once the problem is framed that way, the remaining step — use a wedge/cam to convert the vertical ram stroke into the chosen tool-axis direction — is the definition of a predictable mechanical solution, and the record shows the art already used "stationary inclined planes" as motion/clamping converters in presses.

---

5. Claim-by-claim § 103 analysis

Legend for verdicts: ● = strong obviousness case (single-reference or two-reference, high confidence) · ◐ = good obviousness case, some gap · ○ = weaker case; would need a third reference or a "common sense"/design-choice rationale · N/A = claim text unavailable in the record.

Group A — Gap-holder and horizontal-punch structure

Group B — Stripper sleeve / springs / pre-compression

Group C — Actuator, cam conversion, punch removability

Group D — Gages

Group E — Work-positioning / hold-down / camber removal

Group F — Multi-unit systems, bases, and methods

---

6. Combination summary table

---

7. The best available nonobviousness arguments — and how they fare

I state these fairly, because a competent § 103 report should identify where the case is weak:

1. "No single reference discloses a horizontal punch in a gap holder driven by a vertical ram."
   True, and this is the claim family's best point. But it is an argument about combination, not about elements. Under KSR, "the combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results." Rotating a tool axis and adding a wedge to redirect a ram stroke are precisely "familiar elements… known methods." The inventor's own specification concedes the rotation is "obvious" and the actuating connection may be "any other" type.

2. "All the closest art is the inventor's own."
   This argument fails as a matter of law for the 1934–1939 patents, because § 102(b) contains no "by another" limitation and § 103(c) (the common-ownership disqualification) does not reach § 102(b) art. In re Katz; MPEP § 2133.01. It does succeed for US 2,317,424, US 2,355,344, US 2,323,431 and US 2,381,476 — I have excluded them above for exactly that reason.

3. "The specific ram-actuated two-link hold-down with a roller (claims 13, 14, 32) is not shown."
   The strongest surviving nonobviousness position. But it is a narrow position covering three claims, and the functional motivation (camber removal before piercing) is stated in the claims themselves, which cuts against it.

4. "The angle-engaging gage of claim 15 is not shown."
   Also a fair point. The best rebuttal art (Boeing's US 2,421,864 "Angle punch unit") is post-dated and cannot be used. So claim 15 is the claim with the thinnest § 103 record.

5. Secondary considerations. I have no verified evidence of long-felt need, failure of others, copying, or commercial success tied to the claimed subject matter. Wales did found a substantial business (Wales-Strippit), but (a) the patent record shows no assignee, and (b) the record shows a cluster of near-simultaneous Wales filings (US 2,355,344 filed 1942-01-12; US 2,317,424 filed 1942-03-12; '020 filed 1942-04-17; US 2,323,431 filed 1942-05-12; US 2,381,476 filed 1942-09-18), which reads as a busy development program in a crowded, incremental field — the opposite of the "long-felt, unsolved need" narrative. Even if secondary considerations were proven, the nexus requirement would have to be met claim-by-claim, and for claims 1, 2, 16, 18–21, 23–28 and 30 I would expect the elements to be shown as individually available in the art.

---

8. Bottom line

No single reference anticipates. But on the record assembled here, and applying the pre‑AIA § 103 standard, I assess the following:

Practical recommendation for anyone relying on this:

• If you are defending against a demand letter citing '020 — the expiration point (1960-09-07) is dispositive and needs no § 103 analysis at all.
• If you are using '020 as prior art against a later punch-unit patent, the far better references are the § 102(b)-eligible Wales patents, US 2,013,976, US 2,168,377 and US 2,163,641, whose disclosures map more directly onto modern self-contained perforating units than '020's claims do.
• If you are evaluating the validity of '020 itself, publish-quality copies of claims 3–12, 17 and 22 are the single highest-value next step, followed by a pre-1942 search on press hold-down pads sequenced ahead of the punch (US class 83 subclasses corresponding to today's Y10T 83/2135–2159) and on angle/flange back gages, which are the two areas where my record is thinnest.

---

9. Caveats and confidence

• Dates verified from the retrieved records: US 2,013,976 (filed 1934-09-25 / issued 1935-09-10); US 2,163,641 (filed 1938-04-25 / issued 1939-06-27); US 1,955,866 (filed 1932-02-04 / issued 1934-04-24); US 1,964,752 (listed 1934-07-03). Not fully verified: US 2,168,377's exact 1939 issue date (record header shows only "Aug"); US 2,268,787's issue date; US 1,964,752's inventor.
• I did not retrieve the full text of US 1,955,866, US 2,268,787, US 1,715,138 or the 1919 Iron Age article's complete punch-press discussion; my characterizations are based on the snippets quoted above and are flagged where thin.
• No claim charts have been built against claims 3–12, 17 or 22 because the text is not available. I will not infer their scope.
• No litigation or PTAB record exists for this patent (consistent with the two prior sections), so this § 103 analysis is academic/defensive rather than case-driven.
• Search-tool limitation: my web tool returned a step-budget cutoff on one query, so the pre-1942 non-Wales punch-press art (especially hold-down pads and angle gages) is not exhaustively covered here. Treat the "○ / ◐" verdicts as provisional pending a targeted class-based search.