US 9003852

I'll verify the patent's litigation/docket status via search, then compile the summary from the authoritative full text provided.

US 9,003,852 B2 — Verification & Summary

Sourcing note: I could not query USPTO PatentCenter / CAFC docket systems directly. My searches returned the Google Patents record (fetched 2026-09-21), the patentimages PDF, FreePatentsOnline, and Patents-Review. Searches of CAFC 2026 scheduled-case listings and Federal Circuit opinions/orders, and general litigation searches, returned no docket, appeal, IPR, or enforcement activity involving 9003852. I found no authority for any such proceeding; treat that as "no evidence found," not as a confirmed negative. There is also a date inconsistency in your instructions (task header says April 26, 2026; system date is 2026-09-21) — I flag it rather than resolve it.

Bibliographic data (as recorded literally)

Abstract (verbatim)

A sheet metal blank for the manufacture of a casing of a gas turbine engine comprising a sheet of substantially constant wall thickness and a raised or thicker region extending along substantially the full length of the sheet metal blank. A raised or thicker region is provided between edges of the sheet metal blank. The raised or thicker region comprises at least one boss and at least one ridge extending way from the or each boss.

Independent claims (there are two: claims 1 and 10)

Claim 1 — the blank. A sheet metal blank for making a gas-turbine-engine casing, comprising:

• a metal sheet of substantially constant wall thickness; and
• at least one raised/thicker region protruding along substantially the full length of the sheet;
• wherein the raised/thicker region is between the edges of the sheet;
• the raised/thicker region comprises two or more bosses plus at least one ridge connecting two of those bosses; and
• the blank is adapted to be rolled into a gas-turbine-engine casing.

Plain language: a flat sheet of uniform gauge that carries raised pads (bosses) linked to one another by raised strips (ridges), running the sheet's length and body (not just at the rim) — pre-formed features that survive being rolled into a casing shell.

Claim 10 — the method. A method of forming a casing:

• (a) manufacture a sheet metal blank with two sides and two ends, a metal sheet of substantially constant wall thickness, and at least one raised/thicker region protruding along substantially the full sheet length, located between the sheet edges, comprising two or more bosses and at least one ridge connecting two of the bosses;
• (b) deform the blank to be substantially circular in cross-section;
• (c) weld the blank at its two ends;
• wherein step (b) is performed by passing the blank along a series of rollers whose through-path is wider than the sheet wall thickness but narrower than the raised-region thickness, to form a gas-turbine-engine casing.

Plain language: the raised features act as the rolling contact surface — because the roller gap is set between the thin-wall gauge and the thick-region gauge, the rollers ride on the bosses/ridges rather than on the thin panel, so the sheet is bent consistently along its full length instead of deforming unevenly between features.

Dependent claims (concise)

• 2 — further raised/thicker region forming a flange along at least one edge.
• 3 / 4 — blank is substantially rectangular / is a sector of a ring.
• 5 — sheet wall thickness greater than 0.7 mm but not greater than 5 mm.
• 6 — raised/thicker region thickness no greater than twice the wall thickness.
• 7 — raised/thicker region width no less than 7 mm.
• 8 — raised/thicker region protrudes from one side; the other side is planar.
• 9 — raised/thicker region of substantially constant thickness along the full length.
• 11 — raised region formed by machining and/or chemical etching.
• 12 — further deforming the casing on a sizing tool.
• 13 — gradual (splayed) change in width in the ridge-to-boss transition.
• 14 / 15 / 16 — cylindrical boss; plural bosses with differing diameters; plural bosses of varying sizes.

Points worth flagging

• Claim 1 is narrower than the written description's "first aspect." The summary of invention recites "at least one boss and at least one ridge extending away from the or each boss"; granted claim 1 requires two or more bosses and at least one ridge connecting two of them. A single isolated boss with a trailing ridge — expressly contemplated in the description — is not within claim 1 as issued. Claims 13–16 are all dependent on claim 1, so the "single boss" embodiment is unclaimed.
• The stated technical rationale is that the raised regions keep the rollers in contact with the blank throughout bending, and that thickening between bosses stiffens the casing while remaining within acceptable stress limits — described as "counterintuitive" because such features would ordinarily be stress concentrators.
• Claim 10's deformation step and the rollers limitation are linked; the roller-gap geometry is a positive limitation of the claim, not merely a described embodiment.
• Uncertainty: I have no authoritative confirmation of current USPTO fee/status transactions beyond the legal-events listing shown, no claim-construction or validity rulings, and no litigation or PTAB record for this patent. If you need certified status or a litigation clearance, that requires a direct PatentCenter/PTAB/PACER query, which I could not perform here.

I'll search for litigation involving US patent 9003852.

Let me search more targeted litigation databases.

Litigation Search Report — U.S. Patent No. 9,003,852

Patent identified: U.S. 9,003,852 B2 — "Sheet metal blank"

• Application No. 11/808,692; filed June 12, 2007; granted April 14, 2015
• Inventors: Richard Green, Michael Annear
• Assignee: Rolls-Royce plc (original and current)
• Priority: GB0619361.9, Sept. 29, 2006 (GB 2442238 B)
• CPC: F01D25/24, F23R3/00, B21D53/84, etc. (gas turbine engine casings)
• Status per Google Patents: Active; adjusted expiration Feb. 13, 2032

Bottom line

No litigation involving U.S. Patent No. 9,003,852 was found. I could not locate any district court action, ITC investigation, CAFC appeal, PTAB proceeding, or arbitration/opposition referencing this specific patent number as an asserted or challenged patent.

Given that the patent is owned by Rolls-Royce plc — a practicing, vertically integrated OEM that does not license or assert patents as a monetization strategy — and that the subject matter is an internal manufacturing feedstock blank for gas turbine casings (a process/manufacturing-adjacent article claim), the absence of enforcement litigation is consistent with what one would expect.

Search methodology and what was checked

Important caveat

A negative result from web-indexed sources is not conclusive proof that no case exists. District court complaints, protective-order-sealed filings, and recently filed dockets may not be indexed. To confirm, a Westlaw/Lexis KeyCite "patent history" report, a Docket Navigator patent-litigation report, or a direct PACER/RECAP docket query on the patent number should be run. I was unable to complete those live queries in this session, so I state the conclusion as "none found" rather than "none exists."

False positives — do NOT attribute these to US 9,003,852

Several similarly numbered but entirely different patents surfaced during the search. To avoid mis-attribution:

• U.S. 9,036,852 — the "'852 patent" in the U.S. International Trade Commission's general exclusion order on certain earpiece devices (Fed. Reg. notice, Nov. 6, 2019). Different patent, different technology, different owner.
• U.S. 6,711,385 — Satius Holding v. Samsung; also MeshDynamics/Cisco matters involving 7,894,385 and 8,514,852. Again, unrelated numbers.
• U.S. 11,266,852 — BTL Industries v. Rejuva Fresh (medical device). Unrelated.
• GB 9003852 (1990, skid steer vehicle), BR PI 9003852-5 (Petrobrás), and various non-patent "9003852" identifiers (trench coat SKU, Texas real-estate license, COVID isolate EPI_ISL_9003852) — all unrelated to the patent at issue.

These are listed only to demonstrate that search results for the digit string "9003852" are dominated by other subjects, and none of them are lawsuits against or about U.S. 9,003,852.

Related activity that may be of interest (but is not litigation)

• GB 2442238 B — the UK priority sibling of this patent, listed on Google Patents as "Expired – Fee Related" (not active, but no revocation or opposition recorded).
• US 9,003,852 — still showing Active, with maintenance fees paid at the 4th year (2018-10-15) and 8th year (2022-10-04). No litigation-related legal events on the register; only assignment (2007) and fee payments.
• Forward citations by other OEMs (Pratt & Whitney Canada's structural case patents, GE's travelling-wave inhibitor) reflect technological citation, not disputes.

Summary table of known litigation

If you need a defensible "no litigation" opinion for an IP due-diligence or freedom-to-operate file, I recommend confirming against: (1) a Westlaw KeyCite patent history report, (2) a Docket Navigator "Patent Litigation" report by patent number, (3) PACER/RECAP full-text docket search, and (4) the Unified Patents litigation portal — none of which I was able to query live in this session.

I'll verify the ODP result against live sources before reporting.

Let me run two more targeted checks before reporting.

PTAB Proceedings — US 9,003,852 B2 ("Sheet metal blank")

Patent Owner: Rolls-Royce plc · Inventors: Richard Green, Michael Annear · Priority: 2006-09-29 (GB0619361.9) · Filed: 2007-06-12 · Granted: 2015-04-14 · Listed status: Active, adjusted expiration 2032-02-13 · Claims: 16 (2 independent — claim 1 apparatus, claim 10 method)

Verification date: 2026-09-21. Sources checked: (1) the structured "PTAB proceedings on file" block sourced from the USPTO Open Data Portal; (2) targeted web searches for IPR/PGR/CBM references to this patent number, its assignee, and its title. Neither the ODP record nor any public source surfaced an AIA trial naming this patent.

Proceedings overview

Total AIA trial proceedings on file: 0. Breakdown by status: 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials. There is no proceeding to rank "most impactful," so no per-proceeding entries follow — inventing an IPR number, panel, or claim-level holding here would be exactly the kind of fabrication this task forbids. The bottom-line defensive posture for a defendant is therefore "untested patent, open season on validity": no claim of the '852 patent has ever been construed by the Board, no claim has been canceled, no amended claim has been substituted, and — critically — no § 315(e)(2) estoppel has attached to anyone. All 16 claims are available to be challenged, and the prosecution file itself hands you a ready-made prior-art roadmap.

No proceeding entries

I cannot populate the per-proceeding template (judge panel, grounds, institution decision, FWD, settlement, appeal) because there is nothing to populate it with. For the avoidance of doubt on the two things a defendant most wants to know:

• No Final Written Decision exists for this patent, so there is no claim-level disposition to quote and no FWD link to provide. Any claim that claims 1–16 were canceled or sustained in an IPR would be false.
• No Federal Circuit appeal exists arising from an AIA review of this patent. I found no CAFC docket (and no CourtListener entry) tying an appeal to a PTAB decision on the '852 patent. The only Rolls-Royce/UTC Federal Circuit and E.D. Va. litigation I encountered concerns swept fan blade technology (U.S. Patent No. 6,077,077 and the '931 reissue application) — unrelated to the '852 patent — and I am flagging it only so it is not mistaken for activity on this patent.
• No defensive aggregator (Unified Patents, RPX, IPR-focused fund) appears in the chain. There is no petitioner history to report.

Strategic summary

Claim status: everything is UNTESTED. All 16 claims — including independent claim 1 ("a metal sheet of substantially constant wall thickness … at least one raised or thicker region … comprises two or more bosses and at least one ridge connecting two of the bosses, and the blank is adapted to be rolled to form a casing of a gas turbine engine") and independent claim 10 (the method counterpart, with the step-b) roller-through-path limitation) — stand exactly as they issued on 2015-04-14. There are no canceled claims, no surviving-after-IPR claims, and no certificate of correction or reexamination certificate narrowing them that I could find. The only adverse family-level signal is foreign: the GB counterpart, GB2442238B, is listed as "not_active — Expired – Fee Related," which suggests the family has not been aggressively policed outside the US.

Estoppel landscape: a blank slate, which is the single most valuable fact in this report. Because no IPR has ever been filed, no petitioner is estopped, no privity chain exists, and there is no IPR record for a court to lean on under § 315(e)(2). A defendant currently facing assertion has the full statutory toolkit available:

• IPR under § 311 is the only available AIA vehicle. This is a pre-AIA patent (filed 2007-06-12), so PGR is unavailable, and the claims are directed to gas-turbine sheet-metal blanks and casings, not financial-services data processing — so CBM review was never available (and the program has since sunset for petitions filed after 2020-09-16 regardless).
• Grounds are limited to § 102 and § 103 "only on the basis of prior art consisting of patents or printed publications" (§ 311(b)). No § 112 grounds, no public-use or on-sale art, no system art. This matters: the '852 specification is broad and functional, and a § 112 written-description/enablement attack — often the softest spot in a claim like "adapted to be rolled" — is simply not on the menu at the Board. Preserve it for the district court.
• Watch the § 315(b) clock: one year from service of the complaint on your client. If you are already inside that window, the deadline is jurisdictional in effect and non-extendable.

Pattern signals: none, which is itself the signal. There is no repeat petitioner, no serial-filing pattern, no Director review activity, and no PTAB appeal history. Well-asserted patents in aerospace are routinely hit with IPRs by competitors — the Board's own docket shows Pratt & Whitney Canada, GE, and UTC filing in the gas-turbine casing space (e.g., the '852 patent is cited by Pratt & Whitney Canada's structural-case family, US9498850/US10180084, and by GE's US20180073522/US10808574 and Pratt & Whitney Canada's EP4488495 "Local stiffening for gas turbine engine casing"). The absence of any IPR against '852 most likely means the patent has not been asserted in litigation that would trigger a § 315(b) response — not that it is invulnerable. Treat the zero count as a signal about assertion history, not about validity strength.

Recommended next steps

1. Do not expect an FWD to lean on. There is no PTAB decision to cite and no claim to quote as canceled. If a demand letter or complaint against your client cites claim 1 or claim 10, your validity case must be built from scratch — you are not riding anyone's coattails and you are not blocked by anyone's estoppel.
2. Mine the prosecution file first — it is your roadmap. The 18 references cited on the face of the patent include several directly on point and are the natural starting set:
   • US4725334 (Chem-Tronics, 1988) — "Method of forming integrally stiffened structures." This is the closest conceptual hit to the "raised or thicker region on a constant-thickness sheet, then formed" idea.
   • US7093470 (Boeing, 2006) — "Methods of making integrally stiffened axial load carrying skin panels for primary aircraft structure and fuel tank structures." Directly relevant to claim 1's "adapted to be rolled" and to the thickness-ratio limitations of claims 5–6.
   • US4699567 (Rolls-Royce, 1987) — fan duct casing; GB2110306 (Rolls-Royce, 1983) — turbomachine housing; US5605438 (GE, 1997) — casing distortion control; US5413456 and US6059524 (UTC) — fan casing structures.
   • Caveat on § 325(d): because these references were before the examiner, re-asserting them invites a § 325(d) discretionary denial. Under current Office practice (Director Squires's 2026 discretionary-denial decisions and the heightened § 325(d) scrutiny now applied in reexamination as well), you should lead with new references and new combinations, using the cited art only to frame the state of the art.
3. Consider running IPR and ex parte reexamination in parallel. In the current environment (AIA filings down sharply, reexam requests up, ~90% reexam grant rate, and no requester estoppel or settlement-stoppage from reexam), a reexam is a useful hedge: it is anonymous, it survives settlement, and it lets you raise art the examiner never saw. The trade-off is that reexam does not create the § 315(e)(2) estoppel leverage that an IPR does. Neither route is precluded here — the slate is genuinely clean.
4. Claim-scope check for non-infringement, in parallel. Claim 1 is narrower than the specification's disclosure: it affirmatively requires "two or more bosses and at least one ridge connecting two of the bosses." The specification itself confirms the point — it states that where "bosses are not required, only ridges and/or a flange extending along the full length of the blank … are provided." A blank or casing with isolated bosses and no inter-boss ridge, or with only a full-length edge flange, sits outside claim 1 on its face. Likewise, claim 13's "gradual change in width in a transition between the at least one ridge and the at least one boss" and claim 8's one-side-raised/other-side-planar limitation are concrete, infringement-avoidance-relevant limits.
5. Confirm the zero count before you rely on it, and diarize it. ODP ingest can lag and recently filed petitions (especially those filed within the last quarter) may not yet appear. Before finalizing a validity budget, run a docket search on the patent number in USPTO PTAB E2E (https://ptacts.uspto.gov/ptabweb#/dashboard) and a party search for "Rolls-Royce" restricted to AIA trials, and repeat it at each case milestone. Check CourtListener's docket search (https://www.courtlistener.com/?type=o&q=%229003852%22) for any concurrent district-court assertion, which is what would generate the § 315(b) trigger and any real IPR risk.

Disclosure of limitations: Everything above regarding claim scope, dates, and cited references is drawn from the authoritative full text you supplied for US 9,003,852. The zero-proceeding finding rests on the ODP structured block plus my own web searches on 2026-09-21, and I did not have direct programmatic access to PTAB E2E to query it independently — hence the recommendation to re-confirm. I did not identify any judge panel, petition ground, institution decision, FWD, settlement, or appeal for this patent, because none exists in the public record I could reach.

I'll verify the assignment record against the USPTO Assignment Center and related sources before writing the analysis.

Let me check for any post-issuance transfers, litigation, or NPE activity involving this patent.

Assignment & Ownership Analysis — US 9003852 B2

Patent: US 9003852 B2, "Sheet metal blank" — Rolls-Royce plc
App. 11/808,692 filed 2007-06-12 · Priority GB0619361.9 (2006-09-29) · Granted 2015-04-14
Verification sources: USPTO Assignment Center · Google Patents legal events

Inventors

Both inventors executed an assignment of all right, title and interest to Rolls-Royce plc roughly eight months after the 2006-09-29 GB priority filing and before the US filing, which is the ordinary employment/obligation-to-assign pattern for a UK aerospace OEM. No unusual pattern detected: I found no evidence of either inventor departing Rolls-Royce within 12 months of filing, and no subsequent inventor-side assignment (which is what a departure-driven chain would normally generate). I could not independently verify current inventor whereabouts; treat "no departure" as an absence of evidence rather than affirmative confirmation.

Note on data quality: the Google Patents "Family Cites Families" table for this patent is polluted with unrelated pharmaceutical art (e.g., carvedilol, aryloxypropanolamines) that has no relationship to this patent family. That is a Google family-linking artifact, not an assignment record, and is disregarded here.

Original assignee

Rolls-Royce plc (registered in Great Britain; subsidiary of Rolls-Royce Holdings plc, LSE: RR.) is the entity named on the issued patent and, per the only recorded assignment, remains the owner.

• Primary line of business: design, manufacture and service of gas turbine engines for civil aerospace, defence aerospace, and power systems. The patented subject matter — a sheet-metal blank with integral raised/thicker boss-and-ridge regions for rolling into a compressor/turbine/combustor casing (claims 1–16) — sits squarely in Rolls-Royce's core casing manufacturing lines.
• Product embodying the claims: yes, as an operating matter — the patent is a manufacturing-blank claim, and Rolls-Royce builds the gas-turbine engine casings that these claims describe. (I have not independently confirmed a specific production part number traces to this patent.)
• Current status: operating. Not acquired, not dissolved, not in bankruptcy. The company weathered a severe COVID-era downturn (2020 restructuring and job reductions), but there is no US Chapter 7/11 filing — Rolls-Royce is a UK entity — and no recorded patent sale in connection with any restructuring.
• Family status variance worth noting: the GB family member GB2442238B is recorded as "not_active — Expired - Fee Related" (lapsed for non-payment of renewal fees), while the US patent remains Active with maintenance fees paid. Divergent maintenance behaviour between family members is common and is not itself an NPE signal.

Assignment timeline

The USPTO Assignment Center shows exactly ONE recorded assignment for US 9003852, and it is the original inventor→company assignment. There are no post-issuance transfers, no security interests, no name changes, and no releases recorded.

• 2007-05-30 (executed) / recorded 2007-06-12 — Reel 019454/0946
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: GREEN, RICHARD; ANNEAR, MICHAEL (joint)
  • Assignee: ROLLS-ROYCE PLC, GREAT BRITAIN
  • Correspondent: Not determinable from the sources retrieved. Google Patents' legal-events feed for this patent carries the reel/frame, parties, and effective date but omits the correspondent of record; the underlying reel-019454/0946 cover letter is only obtainable via the Assignment Center document viewer. With a single-link chain there is no recurrence to test anyway, so this omission does not change any signal below.
  • Context: Standard employment/obligation-to-assign capture of inventor rights by the operating company, executed pre-filing and recorded on the US filing date. Not a fire sale, not a reorg, not a transfer to an asserter.

Non-assignment legal events (for completeness — these are maintenance/grant events, NOT conveyances):

• 2014-12-03 — FEPP, fee payment procedure (large-entity status confirmed)
• 2015-03-25 — STCF, patent grant recorded
• 2018-10-15 — MAFP, 4th-year maintenance fee paid (large entity)
• 2022-10-04 — MAFP, 8th-year maintenance fee paid (large entity)
• Adjusted expiration per Google/Patents: 2032-02-13

Timeline diagram

timeline
    title Ownership of US 9003852
    2006 : GB priority application filed
         : Inventors Green and Annear
    2007 : US application filed
         : Assigned to Rolls-Royce plc
    2015 : US patent granted
    2018 : Maintenance fee 4th year paid
    2022 : Maintenance fee 8th year paid

NPE / troll-pattern signals

Inverse observation: the only assertion posture in this record runs the other direction — the patent owner is itself a repeat patent plaintiff against genuine aerospace competitors (Rolls-Royce plc v. United Technologies Corp.). That is the classic operating-company-assertion profile, not an NPE profile.

Verdict

Insufficient data — within the meaning of this rubric, "only the original assignment."

Only one conveyance exists in the entire chain (inventor→Rolls-Royce plc, executed 2007-05-30, recorded 2007-06-12, reel 019454/0946), so there is no post-issuance ownership sequence to analyze for NPE behaviour. Substantively, every one of the eight signals is not present: the sole assignee is a public, product-shipping aerospace OEM that has paid maintenance fees through the 8th year (2022-10-04), and the only litigation record touching Rolls-Royce in this domain was brought by the original assignee on a different patent — a clean operating-company posture. Residual uncertainty is limited to (a) the un-retrieved correspondent of record on reel 019454/0946, which is immaterial on a one-link chain, and (b) the absence of any data after the last fee payment, since an unrecorded or later transfer cannot be ruled out from a single snapshot.

Verify at: https://assignmentcenter.uspto.gov/ — search by patent number 9003852; the sole hit should be reel 019454/0946.

I'll verify the patent against USPTO sources and then examine each cited reference.

Now let me pull details on the most technically significant cited references.

Prior-Art Analysis — US 9,003,852 B2 ("Sheet metal blank")

0. Scope, verification, and two flagged inconsistencies

What I did and did not do. I searched for the identifier 9003852 and confirmed the document via Google Patents, the granted-PDF host, FreePatentsOnline, uspto.report and Espacenet. I could not reach USPTO PatentCenter / PAIR directly, so bibliographic data below is cross-checked against those mirrors rather than a certified USPTO fee/status record. This is a limitation, not a claim of certification.

Identifier hygiene (per your literal-interpretation rule). The search for 9003852 surfaced three incidental near-matches that are not this patent and are excluded from all analysis:

Two inconsistencies, flagged rather than resolved:

1. Date conflict. Your task header states "Current Date: April 26, 2026," while my system date is 2026-09-21. The previously generated section already flagged an April-26-vs-September-21 discrepancy. This does not change the §102 analysis (all cited references pre-date the 2006-09-29 priority date by years), but I note it because I was instructed not to auto-correct.
2. Citation count vs. list length. Google Patents renders the front page as "Citations (18)" and "Patent Citations (18)," yet the enumerated list contains 18 entries, of which I can date 17. GB1095376A is listed with no date, assignee or title fields populated. I therefore treat one entry (GB1095376A) as insufficiently characterized in the underlying record and mark it as such rather than guessing.

1. The reference set, and a note on provenance

The 18 front-page references split, per Google Patents' own asterisk convention (* = "cited by examiner"; unmarked = cited on the face of the patent), as follows. Because the asterisk rendering is scraped, treat the split as indicative, not certified:

• Examiner-flagged (*): US 4,699,567; US 4,725,334; US 5,413,456; US 5,605,438; US 6,059,524; US 6,497,105; US 2003/0039510; US 6,681,577; US 6,792,757; US 7,093,470.
• Unmarked: GB 1,095,376; GB 1,176,053; GB 1,211,313; GB 1,303,573; GB 1,548,704; GB 2,110,306; US 4,502,809; GB 2,397,343.

Critically, all 17 datable references pre-date the 2006-09-29 priority date (latest publication: US 7,093,470 B2, 2006-08-22; latest priority: US 6,681,577 B2, 2002-01-16). They are therefore facially available as prior art under §102(a)/(b) (publication) or §102(e) (US filings). The forward citations on the record — US 9,498,850; US 10,180,084; US 2018/0073522; US 10,808,574; EP 4,488,495; US 2025/0012204; US 12,448,900 — post-date the priority date and are not prior art; I exclude them from §102 consideration.

2. Master table — full citations as rendered

3. §102 analysis — reference by reference

Framing rule I apply throughout. Anticipation under §102 requires a single reference to disclose every element of a claim, arranged as claimed (no reliance on the knowledge of a skilled artisan to supply a missing element). Claim 1 requires, in combination: (i) a gas-turbine-engine casing blank; (ii) a metal sheet of substantially constant wall thickness; (iii) a raised/thicker region protruding along substantially the full length; (iv) that region between the edges; (v) two or more bosses and at least one ridge connecting two of those bosses; (vi) the blank adapted to be rolled. Claim 10 adds the roller-gap geometry (through-path wider than the sheet wall, narrower than the thick region) plus welding. Applying that rule:

Tier 1 — Closest single reference

US 4,725,334 A — Chem-Tronics, Inc., "Method of forming integrally stiffened structures" (filed 1985-05-15; granted 1988-02-16). URL: https://patents.google.com/patent/US4725334

• Description. Discloses a lightweight structure of a skin with integral reinforcing ribs, produced from plate stock by chemical milling/etching through masking steps. The specification expressly names jet-engine compressor housings and a fan case 20 as the object; ribs run in geodetic/waffle patterns and intersect at nodes; "attachment features such as pads and bosses can be formed integrally and at the same time as the reinforcing ribs" (pad 50, shown before/after final machining); flanges 44, 46 are welded to the shell, which is then resized and stress-relieved with final machining.
• Potential §102 read. This is the only cited reference that simultaneously touches: gas-turbine casing (compressor housing/fan case), constant-thickness skin, integrally formed raised features, bosses/pads, ribs, welded flanges, and post-forming resizing. On the broadest reading it maps onto claims 1, 2, 6, 8, 9 and 11 and onto much of claim 10's preamble.
• Why it likely still fails §102 for claims 1 and 10. (a) It does not describe a blank that is adapted to be rolled, nor the roller through-path limitation of claim 10 — its forming is chemical milling of plate, plus welding and resizing. (b) The rib network is described as geodetic/waffle patterns of intersecting ribs at nodes, which is not the claimed "at least one ridge connecting two of the bosses" architecture, and the record does not show ribs connecting two or more bosses in the claimed sense. (c) The "raised region protruding along substantially the full length" limitation is not shown in those words. Net: strongest §103 reference; not a clean §102 anticipation.

Tier 2 — Materially relevant to specific elements

US 7,093,470 B2 — The Boeing Company (El-Soudani), filed 2002-09-24, granted 2006-08-22. URL: https://patents.google.com/patent/US7093470

• Description. Methods of making integrally stiffened unitized skin panels for aircraft/fuel-tank structures. Two variants: (1) machine a starting plate to final part geometry, then explosive-form to uniaxial/biaxial curvature; or (2) explosive-form a plate first, then machine to final geometry. Discloses stiffening features described as "ribs, stiffeners, frame outer chords, stringers" and flanges, integrally formed with the skin.
• Potential §102 read. Maps onto the forming-after-machining concept and onto claim 11 (forming the raised region by machining) and the general notion of an integrally stiffened panel later deformed to curvature (relevant to claim 1's "adapted to be rolled"). It is not anticipatory of claim 1: no gas-turbine casing, no boss-and-ridge architecture, and the deformation is explosive forming, not roller forming.
• Correct characterization: §103 art against claims 1/10/11 when combined with the roller-forming teachings of the primary reference; not §102.

US 4,699,567 A — Rolls-Royce plc (Stewart), priority GB 8414510 (1984-06-07), granted 1987-10-13. URL: https://patents.google.com/patent/US4699567

• Description. A gas-turbine fan duct casing / containment ring: a relatively thin-section substantially rigid cylinder 23 with flanged ends 24, 25, wrapped in discrete-piece woven fibrous layers plus continuous wound layers, finished with an impervious skin. Claims are directed to the ring/lay-up, not to any boss/ridge blank.
• Potential §102 read. Touches the thin-wall gas-turbine casing and flange environment (relevant to claim 2). It discloses no raised/thicker region, no bosses-and-ridge structure, and no rolling of a flat blank. No §102 anticipation of any claim; relevant only as background/§103 context.

GB 2,110,306 A — Rolls-Royce Limited, "Turbomachine housing", priority 1981-11-26, published 1983-06-15.

• Description. Same-assignee family lineage as the target's problem statement (turbomachine housings). I could not retrieve the specification body in this session — I do not know its disclosure with high confidence, so I decline to assert a §102 mapping. On its title and provenance it is at best §103 context for the "casing formed as a shell rather than a casting/forging" concept. Flagged as unverified.

US 5,605,438 A — General Electric Co., "Casing distortion control for rotating machinery", filed 1995-12-29, granted 1997-02-25.

• Description. Addresses distortion of thin casings — directly the problem the target patent recites (weld-induced distortion of thin (1–5 mm) casings). Relevant as motivation/§103 art against claim 1's stated purpose. No boss/ridge blank disclosure is reflected in the front-page characterization, so no §102.

US 6,681,577 B2 — General Electric Co., "Method and apparatus for relieving stress in a combustion case in a gas turbine engine", priority 2002-01-16, granted 2004-01-27.

• Description. Stress relief in a gas-turbine combustion case — i.e., manipulating stress concentrations at case features. This is the closest art to the target's stated rationale (thickened features as stress concentrators that must be "tuned"). Relevant to claim 1's functional setting and to claim 13 (splayed transition to manage stress), but the front-page characterization shows no raised-region/ridged-blank structure. §103, not §102.

US 5,413,456 A — United Technologies Corp., "Aircraft fan containment structure" (1994-04-29 / 1995-05-09) and US 6,059,524 A — United Technologies Corp., "Penetration resistant fan casing for a turbine engine" (1998-04-20 / 2000-05-09).

• Description. Fan-case containment structures — thin metal/composite casing shells needing stiffness and impact resistance. Relevant to the structural-stiffening motivation for claim 1 and to claim 2 (flange) context. Neither is characterized as disclosing a flat blank with integrally raised bosses joined by ridges, and neither is a rolling method. No §102.

GB 2,397,343 A — Rolls-Royce plc, "Gas turbine engine viscoelastic blade containment assembly", priority 2003-01-16, published 2004-07-21.

• Description. Viscoelastic blade-containment assembly; same-assignee casing technology. Background/§103 only. No §102.

Tier 3 — Peripheral; cited for context, not for the inventive combination

Because these do not, on the record, touch the boss-plus-connecting-ridge blank or the roller-gap forming step, I map them only to the limited claims noted:

4. Bottom line on §102

• No reference among the 18 is a clean, complete §102 anticipation of independent claim 1 or independent claim 10. The only serious single-reference candidate is US 4,725,334 A (Chem-Tronics), and it fails at minimum on the "adapted to be rolled"/roller-gap limitation (claim 10) and on the "ridge connecting two of the bosses" architecture (claim 1). Its real force is under §103.
• The strongest realistic invalidity theories are §103 combinations, not §102:
  • US 4,725,334 A (integral skin + ribs + integrally formed bosses/pads + welded flanges + chemical-milling/machining) in view of US 7,093,470 B2 (machine-then-form integrally stiffened panel) and/or the roller-forming teaching already in the target's own FIG. 4 "known rolling arrangement" → claims 1, 6, 8, 9, 11.
  • US 4,699,567 A (thin-section flanged cylindrical gas-turbine casing) + US 5,605,438 A / US 6,681,577 B2 (distortion and stress control at case features) → claims 2, 13.
• Claims most exposed to prior art generally: 1, 2, 6, 8, 9, 11. Claims least exposed: 3, 4 (blank geometry: rectangular / ring sector), 5 (0.7–5 mm window), 7 (≥7 mm width), 12 (sizing tool), 14–16 (cylindrical/varying-diameter bosses) — these recite specific numeric or structural parameters that, on the present record, no cited reference discloses.

5. Explicit uncertainties

• I could not retrieve the full specifications of GB 1,095,376 A, GB 1,303,573 A, GB 2,110,306 A, US 4,502,809 A, US 5,413,456 A, US 5,605,438 A and US 6,059,524 A in this session; for those I relied on the front-page characterization in the record and therefore decline to assert §102 mappings. GB 1,095,376 A has no dates or title populated at all.
• The examiner/applicant split (§1) rests on scraped asterisk rendering and is indicative only.
• I have no claim-construction, IPR/PTAB, or litigation record for US 9,003,852, and I could not query PatentCenter/PAIR directly. Certified status and any post-grant challenge require those direct queries.
• Per your rule, I did not auto-correct any identifier; note that 9003852 also resolves to unrelated Brazilian and UK documents, which I excluded by design.

I'll verify the key prior-art references before building the obviousness analysis.

Obviousness Analysis — U.S. 9,003,852 B2 under 35 U.S.C. § 103

Date/personnel note: This task header says April 26, 2026; the system clock says 2026-09-21. That discrepancy was already flagged in the earlier sections and is not repeated here.

Governing framework. Priority is 2006-09-29 (GB 0619361.9), so this is a pre-AIA case: pre-AIA § 102/§ 103, with the Graham v. John Deere factors (scope and content of the prior art; differences; PHOSITA level; secondary considerations). Because the art at issue is both (i) of record in the file and (ii) in the same field and directed at the same problem, KSR Int'l v. Teleflex supplies the controlling mode of analysis: an express teaching, suggestion or motivation to combine is not required where the combination of known elements yields predictable results, or is the mere application of a known technique to a known device ready for improvement.

POSITA. A mechanical/aerospace manufacturing engineer with ~3–5 years' experience in gas-turbine casing design and sheet-metal fabrication (rolling, chemical milling, machining, welding), familiar with integrally stiffened skin/panel technology.

Limiting caveat on sources. I verified the full text of US 4,725,334 (Chem-Tronics), US 7,093,470 (Boeing), US 5,605,438 (GE), GB 2 110 306 A (Rolls-Royce) and US 4,699,567 (Rolls-Royce). I did not retrieve the substance of GB 1,095,376, GB 1,176,053, GB 1,211,313, GB 1,303,573, GB 1,548,704 (Stal-Laval), US 4,502,809, US 5,413,456, US 6,059,524, US 6,497,105, US 6,681,577, US 6,792,757, GB 2,397,343 or US 2003/0039510 A1, and I do not assert mappings from them.

1. The two strongest single-reference attacks

A. US 4,725,334 (Chem-Tronics, "Brimm") — primary reference against claim 1

https://patents.google.com/patent/[US4725334A](/patent/US4725334A)/en · https://www.freepatentsonline.com/[4725334](/patent/4725334).html

US 4,725,334 is therefore close to a one-reference case on claim 1. The two points that a challenger must shore up are (i) whether a rib-intersection node/pad reads on a "boss," and (ii) the "adapted to be rolled" capability. Both are supplied by the secondary art below.

B. US 7,093,470 B2 (Boeing) — primary reference against claim 10

https://patents.google.com/patent/US7093470

Verbatim: "machining a starting plate to a final part geometry; and explosive forming the machined plate to at least one of a uniaxial curvature and a biaxial curvature"; "machining at least one stiffener into the starting plate 12 and machining the starting plate 12 from a first thickness down to a second or suitable skin thickness"; stiffeners "integrally engaged or formed with the inner skin surface"; a filler material placed in the pockets to protect the stiffeners from "distorting or buckling during the explosive deformation process"; final curvature "dictated by the die surfaces"; and the finished part has an outer mold line surface that is substantially smooth (stiffeners on one side only).

That reference supplies the order of operations rationale that is the entire stated advantage of the '852 patent ("it is easier to machine a flat surface than, say, a curved surface"), and it discloses the reverse order (form, then machine) as well.

2. Specific combinations and the motivation to combine

Combination I — Chem-Tronics + Boeing (claims 1, 5–9, 11, 12; claim 10 sub-steps (a)–(c))

Both references are in metal lightweight structural panels for aerospace, both teach integral (not welded-on) stiffeners/bosses produced by material removal from plate, and both expressly contemplate forming the machined panel to curvature. Motivation: (1) Boeing expressly frames machining-then-forming (and forming-then-machining) as the affordable route to lightweight unitized stiffened panels — the same cost/quality problem the '852 patent addresses; (2) Boeing's use of filler to protect the features during forming provides a specific teaching for how to carry pre-formed raised features through a forming step without losing them. Reasonable expectation of success: high — both are metal panel-forming processes and the combination merely orders known steps.

Combination II — Chem-Tronics + GE US 5,605,438 (claims 1, 13, 16; supports motivation)

https://patents.google.com/patent/[US5605438A](/patent/US5605438A)

GE teaches placing a circumferentially extending rib about each casing half "to prevent radial distortion and meridional roll," plus a plurality of axially extending ribs whose stiffness "substantially match[es] the radial stiffness of said flanges" (claims 5–6). This is the exact technical rationale the '852 specification asserts for running a thickened ridge between bosses ("the thicker/raised regions stiffen the casing"). The motivation to add ridges between bosses in a thin-walled gas-turbine casing is therefore supplied by a reference in the same field, addressing the same failure mode, with a predictable stiffening result.

Combination III — Chem-Tronics + US 4,699,567 and/or GB 2 110 306 A (claims 2, 3, 4)

• US 4,699,567: "relatively thin section substantially rigid cylinder 23 including flanged ends 24 and 25," for a gas-turbine fan duct — supports claim 2 (edge flange) and the thin-wall premise.
• GB 2 110 306 A: an "outer hollow cylindrical casing 18 made up of a plurality of cylindrical sections 18(a)–18(f) … fabricated from one part or a plurality of parts bolted together along a joint or flange which extends in a direction along the length of the compressor." This supports claim 3 (rectangular blank) and claim 4 (sector-of-a-ring blank), because the segment preform for a cylinder is rectangular and for a frusto-cone is an annular sector. Motivation: forming a casing from flat/sectorial preforms and joining along a longitudinal joint or flange is the standard alternative to casting, as the '852 background itself concedes.

Combination IV — for claim 10's welding step

The '852 background section is an admission: casings are fabricated from sheet metal "with end flanges and bosses being welded into place." Under MPEP 2129, applicant's own admission of what was known is available as prior art for § 103. Chem-Tronics additionally discloses welded flanges and electron-beam welding of a formed shell. The welding step of claim 10 is therefore the weakest link in any attempted defence.

3. Dependent-claim mapping

4. Counterarguments, and where the attack fails

1. The examiner had all of this art. US 4,725,334, US 7,093,470 and US 5,605,438 are all in the "Citations" list marked as cited by the examiner, and the claims issued over them on 2015-04-14. A § 103 attack must therefore show the examiner failed to appreciate a combination, not merely failed to find references. That is a materially harder posture, and it is the single biggest weakness in an invalidity theory built only on the of-record art.

2. The roller-gap limitation (claim 10) is the strongest non-obviousness feature. Claim 10 positively requires that the through-path defined by the rollers be "wider than the wall thickness of the metal sheet and narrower than a thickness of the at least one raised or thicker region." The technical effect the specification claims is that the rollers ride only on the raised features, so the sheet is bent along its full length. I found no verified reference in the cited art that discloses this specific roll-gap geometry, and this element does real work (it is what makes claim 10 a method claim rather than a product-by-process restatement). Unless GB 1,548,704 (Stal-Laval, "Method of assembling a turbo-machine") or GB 1,303,573 discloses a roll-pass line equal to the thick-region gauge, claim 10 is defensible. I could not verify those references and will not assert a mapping.

3. Numerical ranges (claims 5, 6, 7). These are classic obvious-to-optimize ranges. The specification's only support is the conclusory statement that 7 mm width "has been shown to distribute stress … to within optimal limits" — no data, no criticality, no unexpected result. Under KSR, a challenger will argue routine optimization. But note the rub: absent record evidence of criticality, claims 5–7 are vulnerable; conversely, absent record evidence from the challenger of why the ranges would have been selected, the applicant can argue the ranges were not predictable. Claim 7 (≥ 7 mm, with no upper bound) is the most exposed.

4. The specification's "counterintuitive" argument is attorney argument, not evidence. The '852 description asserts that "providing a casing with thicker/raised regions extending between bosses is counterintuitive, as it will be appreciated that ordinarily such features act as stress concentration features." That is an unsupported statement in the specification, not a secondary-consideration showing, and it cuts against the applicant because Chem-Tronics expressly solved exactly that problem — "the transitions are gentle, and stress concentrations are therefore avoided" — which is the very feature recited in granted claim 13. A teaching-away defence therefore fails: no reference discourages thickening between bosses; the art instead teaches how to do it without concentrating stress.

5. No secondary considerations on the record. I found no evidence in the file of commercial success, long-felt need, unexpected results, or licensing attributable to the claimed blank. The patent is owned by Rolls-Royce plc, a vertically integrated OEM; the absence of any enforcement activity (as detailed in the earlier Litigation section) means there is also no adjudicated validity record either way.

5. Bottom line

Cleanest theory: a single-reference-plus-KSR attack at claim 1 built on US 4,725,334, in which a rib-intersection pad/node is the "boss" and the intersecting/tapered ribs are the "connecting ridge(s)." Add US 7,093,470 for the "adapted to be rolled / machine-flat-then-form" element and for the method, and US 5,605,438 for the motivation to run a stiffening rib between discrete features on a thin-walled gas-turbine casing.

Where I would not overclaim: I have no verified basis to say the cited art discloses (a) the claim 10 roller-gap geometry, or (b) a 7 mm minimum raised-region width. I also did not retrieve eight of the eighteen cited references, and one of my best references (US 4,725,334) was of record and was overcome during prosecution. Any invalidity opinion should be run against the full text of GB 1,548,704 and GB 1,303,573 — precisely the two references I could not open and the two most likely to bear on the rolling step.