US 11402414

I'll verify against live sources for any litigation or CAFC activity, then summarize.

USPTO / CAFC docket check first

I ran targeted searches for litigation, PTAB, and CAFC activity tied specifically to 11402414 / US 11,402,414. No CAFC 2026 docket and no PTAB (IPR/PGR) proceeding involving this patent number surfaced. My hits for "11402414"-adjacent terms resolved to unrelated matters (e.g., Apple cases against Masimo and CText IP, and PTAB institution notices covering other IPRs such as IPR2025-01451, IPR2026-00149, etc.). I am not aware of any authoritative record of this patent being asserted or challenged, and I could not confirm one way or the other. Treat "no known litigation" as unverified rather than established.

---

Bibliographic data

Abstract (as issued)

A circuit and method for current sensing in DC‑DC converters having coupled inductors. The DC‑DC converter includes first and second inductors coupled to first and second switching nodes, respectively, and a current sensing circuit. The current sensing circuit includes a first current sensing circuit coupled to the first switching node and having a first sense node, and a second sensing circuit coupled to the second switching node and having a second sense node. The circuit further includes a first capacitor coupled between the first sensing circuit and the second sensing circuit.

Independent claims in plain language

Claim 1 — Circuit. A first switching network drives a first inductor at a first switching node, and a second switching network drives a second inductor at a second switching node, where the two inductors share their magnetic fields (i.e., coupled inductors). A current sensing circuit is coupled to both switching nodes and comprises (a) a first sensing network tied to the first inductor at the first switching node, (b) a second sensing network tied to the second inductor at the second switching node, and (c) a capacitor connected between the two sensing networks. The core point: the cross‑coupling capacitor is claimed structurally between the two sensing networks, not limited to sense‑node‑to‑sense‑node.

Claim 11 — Method. Determine the current through each of two coupled inductors by measuring voltages at two sense nodes, where each sense network has a resistor between its switching node and its sense node. Then, using a capacitor coupled between the first and second sense nodes, cause the two sense voltages to be proportional to the currents through the first and second inductors, respectively. This is the DCR‑sensing correction that cancels the cross‑inductor coupling term.

Claim 16 — Apparatus. A load circuit plus a DC‑DC converter providing an output voltage to that load, the converter having two magnetically coupled inductors. The converter includes a first sensing network coupled to the first inductor at a first switching node with a first sense node, a second sensing network coupled to the second inductor at a second switching node with a second sense node, and a capacitor coupled between the first and second sense nodes.

Notable dependent claims

• 3/4 — resistor between switching node and sense node in each network; cross‑capacitor specifically between the two sense nodes.
• 5 — additional capacitor from each sense node to ground.
• 6/7/15/18 — the cross‑capacitor's value is set based at least in part on the coupling factor between the inductors (and on the values of the sense capacitors).
• 8 — recites a DC‑DC converter with the inductors as a coupled‑inductor pair.
• 10/19 — high‑side transistor to Vin, low‑side transistor to ground, per switching network.
• 20 — measurement circuits feeding control circuits that act on the sense voltage and a peak current threshold.

Technical substance / the disclosure's key insight

The patent solves the problem that plain DCR sensing (a series R–C network across an inductor) gives a wrong answer in a coupled‑inductor buck because each inductor's current is perturbed by the other's. Adding capacitor C3 between the two sense nodes produces, in the spec's derivation, a sense voltage equal to the resistive drop of the inductor (V_sns = R_S·i_L) provided two matching conditions hold:

• R_S ≈ R_f·[C1 + C3] = L
• k ≈ −C3 / (C1 + C3) (k = coupling factor)

So C3 is sized as a function of k — the disclosed design rule reflected in claims 6/7/15/18. This is a purely passive DCR‑sense technique requiring no active compensation or injected slope.

Citations and citation status (from the front page)

Seven U.S. references cited, four marked as examiner‑cited (*): US 6,424,129 B1 (Semtech), US 2003/0102849 A1 (Analog Devices), US 7,233,132 B1 (Virginia Tech — current sensing in multiple coupled inductors by time‑constant matching to leakage inductance), US 2003/… (AD), and US 2016/0187386 A1 (Texas Instruments — coupled inductor current sensing); plus US 8,228,014 B2 (GM), US 9,281,748 B2 (Lenovo Enterprise Solutions), and US 9,627,969 B2 (Infineon — inductor current measurement compensation). Notable as the closest‑seeming art is the TI application and the Virginia Tech patent on coupled‑inductor current sensing.

Cited by: US 2022/0109368 A1 and US 11,876,445 B2 (both Infineon Technologies Austria, trans‑inductance multi‑phase power converters), the latter granting 2024‑01‑16.

Uncertainty flags

• Litigation/docket: none found for this specific number; no CAFC 2026 activity identified. Not verified as nonexistent.
• Prosecution history: the file wrapper (office actions, prior‑art rejections, any appeals) was not retrievable from the sources I could reach; I can't say which references drove allowance or what arguments were made.
• Equation 9 discrepancy: the granted text labels the C3 current as "i_C2" using C2 (rather than C3) in the formula, an apparent typographical inconsistency in the specification. The claim and the ultimate result equations use C3 as the cross‑capacitor. I'm reporting the document literally rather than correcting it.
• Applicability of the coupled‑inductor pair: while the spec contemplates multiple coupled pairs (2‑phase is what's drawn), claims 1/11/16 recite only the first/second (two‑inductor) arrangement.

Principal source: US11402414B2 on Google Patents; corroborated by the granted‑PDF text at patentimages US11402414.pdf.

I'll search for litigation involving US patent 11402414 across patent litigation databases.

Initial results are irrelevant (they concern unrelated patents like Wyeth's '024 vaccine patent). Let me search litigation-specific databases.

Let me try more targeted queries on litigation databases and IPR proceedings.

Summary of Findings

I found no litigation involving U.S. Patent No. 11,402,414 ("Direct current sensing for coupled inductors").

What I searched

Using the litigation-oriented queries you specified (including targeted queries pairing "11402414" / "US11402414" / "11,402,414" with litigation terms, and searches referencing the patent title and Unified Patents / litigation databases), none of the results returned any case in which U.S. Patent No. 11,402,414 was asserted, challenged, or otherwise at issue.

Every hit that mentioned a "…02414"-type number or a "'414"-type number traced to different patents entirely, for example:

• Wyeth's pneumococcal conjugate vaccine "'024 patent" (Merck v. Wyeth declaratory judgment litigation) — an unrelated patent whose shorthand is '024, not 11,402,414.
• Various unrelated patents cited in other matters ('991, '834, '816, '114, '564, '599, etc.).
• The CourtListener and PTAB documents returned were unrelated to this patent's technology (LED displays, video streaming, wearables, NFC, pulse oximetry).

I applied your strict rule: I did not conflate any of these similar-looking numbers with 11,402,414, and I did not auto-correct any identifier.

What the patent record itself shows

The authoritative Google Patents record (fetched 2026-09-24) for US 11,402,414 B2 shows:

• Patentee/Assignee: Apple Inc. (original assignee Apple Inc.)
• Inventor: Jonathan F. Bolus
• Application: US 16/820,658; filed: 2020-03-16; granted: 2022-08-02
• Status: Active; adjusted expiration 2040-12-28
• Cited-by (forward citations): only US 2022/0109368 A1 and US 11,876,445 B2 (both Infineon Technologies Austria AG, "Trans-inductance multi-phase power converters and control") — these are patent citations, not litigation.
• Patent-citation list (7 references): US 6,424,129 B1; US 2003/0102849 A1; US 7,233,132 B1; US 8,228,014 B2; US 9,281,748 B2; US 9,627,969 B2; US 2016/0187386 A1 — again, prior-art citations, not litigation.
• The record lists no litigation, no IPR/PGR, and no reexamination entries for this patent.

Conclusion

Known litigation involving U.S. Patent No. 11,402,414: none identified. There is no plaintiff, defendant, jurisdiction, case number, filing date, or outcome to report because no such case appears in the sources searched.

Important caveats (stated explicitly rather than assumed)

• My searches did not return live confirmations from Docket Navigator, PACER, the Unified Patents litigation portal, or the CAFC docket specifically for this patent number. The queries returned either zero results or irrelevant matches. I therefore cannot claim a positive verification of "no cases exist" from those databases the way I could if I had retrieved an empty docket listing from each.
• Because my tool usage was capped before I could complete every planned database-specific query, this is a good-faith negative finding based on the searches performed, not an exhaustive clearance search.
• The definitive way to confirm would be to pull the full docket/assertion history for patent number 11,402,414 directly from PACER/RECAP, Docket Navigator, and the Unified Patents litigation case list, and to check the USPTO PTAB (PTAB E2E / P-TACTS) for any IPR/PGR filings. On the evidence gathered here, those sources would be expected to show no activity for this patent.

If you'd like, I can attempt those specific database lookups again (Unified Patents case list, PACER/RECAP, and PTAB) with additional search steps.

I'll verify the structured data against live sources before drafting.

Let me run a few more targeted checks to make sure I'm not missing a recently-filed or non-indexed proceeding.

Proceedings overview

Zero AIA trial proceedings are on file against US 11,402,414 as of 2026-09-29 — no IPRs, no PGRs, no CBM reviews, no derivation proceedings, no active trials, no claim cancellations, no settlements, and no Federal Circuit appeals arising from a PTAB FWD. Bottom line for a defendant: there is no PTAB record to lean on. The patent is a completely unlitigated-at-the-Board asset — all 20 claims (1–20) stand exactly as issued on 2022-08-02, and its defensive posture is "untested," not "hardened" and not "hollowed out."

That is an important distinction. A patent with zero IPR history is neither validated nor weakened by the Board. Every validity argument is still on the table for a first petitioner, and conversely there is no institution decision, no FWD, and no Board claim construction to borrow for a § 282 defense or an IPR petition.

---

Proceedings on file

None. There are no per-proceeding entries to report; the section template is empty because the underlying data set is empty.

Verification performed

• Structured source (canonical): The USPTO Open Data Portal "PTAB proceedings on file" block returns no AIA trial proceedings for US 11,402,414 / application 16/820,658. Per the task's operating rule, this is the default and controlling answer.
• Live web cross-check (2026-09-29): Targeted searches for "11402414", "11,402,414", and combinations with IPR2021–IPR2025, "Apple," "PTAB," and "trial proceeding" returned no petition, institution decision, FWD, or settlement referencing this patent. Hits were generic PTAB documents that merely mention other Apple or Fintiv-related matters — none of them concern the '414 patent.
• Federal Circuit cross-check: No CAFC docket or opinion was located naming US 11,402,414 or application 16/820,658. No PTAB appeal exists because no PTAB trial exists.
• Prosecution/ownership context confirmed from the patent record: Filed 2020-03-16; assigned to Apple Inc. by assignment effective 2020-03-23 (Reel/Frame 052284/0896); published as US 2021/0285989 A1 on 2021-09-16; granted 2022-08-02; 4th-year maintenance fee paid 2026-01-21; adjusted expiration 2040-12-28. Sole named inventor Jonathan F. Bolus.

⚠️ Caveat, stated plainly: the ODP ingest may lag recently-filed petitions by weeks, and very recent filings (e.g., a petition filed in the last 60–90 days) can be invisible to both ODP and to search-engine indexing. I found no such filing, but I cannot affirmatively rule out a petition filed very recently. Before relying on this for a litigation position, re-run the search directly in PTAB E2E / PatentCenter (https://ptacts.uspto.gov/ptacts/) and on CourtListener (https://www.courtlistener.com/?q=%2211%2C402%2C414%22).

---

Strategic summary

Claim status: all 20 claims UNTESTED. No claim of US 11,402,414 has been canceled, confirmed, or even challenged at the Board. The claim set as it stands for a defendant is the full issued set: independent claims 1 (circuit), 11 (method), and 16 (apparatus), plus dependents 2–10, 12–15, and 17–20. Notably, the patent recites DCR-style sense nodes (claims 3–5, 17), a coupling-factor-dependent capacitance for the inter-sense-node capacitor (claims 6, 7, 18), and a peak-current-limit control loop (claim 20) — all fairly specific limitations that give a defendant real claim-construction and non-infringement room, but none of which have been construed by any tribunal.

Estoppel landscape: a clean slate. Because no IPR has been instituted, 35 U.S.C. § 315(e)(2) estoppel has not attached to anyone. A defendant being asserted against today may file a first IPR on any § 102/§ 103 ground available. Practical constraints to watch are (a) the § 315(b) one-year bar — a petition must be filed within one year of service of a complaint alleging infringement of this patent, or the party is barred; and (b) § 315(a)(1) if that party previously filed a DJ action of invalidity. Neither has been triggered on this record as far as the public docket shows. The absence of any earlier petitioner also means no General Plastic serial-petition risk and no § 325(d) "same art previously considered" estoppel problem beyond what the examiner considered.

Pattern signals: none, and that is itself meaningful. There is no serial-petitioner pattern (no petitioner has filed once, let alone twice), no patent-owner appeal activity (Apple has never had to defend these claims at the Board or the CAFC), and no defensive aggregator (no Unified Patents or similar entity) in the chain. This is a young Apple-owned power-conversion patent — grant 2022-08-02, ~4.5 years old — that has not yet attracted a challenger. That is consistent with a patent that has not been asserted in litigation (asserted patents reliably draw IPRs), rather than one that has been asserted and survived.

Art worth noting for a prospective petitioner. The face of the patent and its citation record identify adjacent coupled-inductor current-sensing work that has never been applied by the Board against these claims:

• US 7,233,132 B1 (Virginia Tech Intellectual Properties) — "Current sensing in multiple coupled inductors by time constant matching to leakage inductance" (cited by examiner).
• US 2016/0187386 A1 (Texas Instruments) — "Coupled inductor current sensing apparatus and systems" (cited by examiner).
• US 9,627,969 B2 (Infineon Technologies Austria) — inductor current measurement compensation for switching regulators.
• US 10,122,269 B2 (Gozzini et al.) / US 2018/0083532 A1 — "Current mode control of a buck converter using coupled inductance," priority to provisional 62/398,264 filed 2016-09-22: this has an Apple assignee chain, so its availability as § 102(a)(2) art is subject to the § 102(b)(2)(C) common-ownership exception, though its pre-2020 publication date may still make it available as § 102(a)(1) printed publication art. Treat the availability analysis as a real, non-trivial question rather than assuming it is clean.

Only one third-party document cites the patent: Infineon's US 2022/0109368 A1 (priority 2020-10-05), a trans-inductance multi-phase converter application — a citation, not a challenge.

---

Recommended next steps

If you are a defendant being asserted against today:

1. Do not assume the patent is weak because it has no IPR history — and do not assume it is strong. There is no FWD to quote and no canceled claim to point to. Your invalidity case must be built from scratch. Verify current status in PTAB E2E before filing anything, in case a petition was filed recently and is not yet indexed.
2. Calendar the § 315(b) bar immediately. The date you (or your privy/RPI) were served with the complaint starts a hard one-year clock to file an IPR. Because § 315(e)(2) estoppel has not yet attached to anyone, a well-built first petition is unusually valuable here — there is no Fintiv/General Plastic baggage from prior petitioners.
3. Mine the examiner-cited references first. US 7,233,132 and US 2016/0187386 map directly onto coupled-inductor current sensing and were cited but (apparently) not substantively applied to the sense-node/coupling-factor limitations of claims 3–7 and 17–18. That is the classic § 325(d) setup where the Board will accept a petition because the art was not "previously considered" in substance.
4. Pressure-test independent claim 1 and claim 16 on the "first capacitor coupled between the first current sensing network and the second current sensing network" limitation. That element — plus the coupling-factor-dependent capacitance of claims 6/7/18 — is the narrowest, most construction-dependent feature in the set, and it has never been construed by any adjudicator.
5. Expect Apple to be the adversary (assignee of record since 2020-03-23), not a non-practicing entity — so an IPR decision will be defended aggressively on appeal if you win. Budget for a CAFC phase.

If you represent the patent owner / are evaluating the asset: the patent is clean but unproven. It has never been tested at the Board, has no reexamination history surfaced, and is on track to expire 2040-12-28 (adjusted). Its value depends entirely on the strength of the coupling-factor-compensated DCR sensing claims 6/7/18 and the peak-current-limit loop of claim 20, none of which have been validated by any tribunal.

Links for verification:

• Patent: https://patents.google.com/patent/US11402414/en
• PTAB E2E proceeding search: https://ptacts.uspto.gov/ptacts/
• CourtListener full-text search: https://www.courtlistener.com/?q=%2211%2C402%2C414%22
• CAFC opinions & orders: https://cafc.uscourts.gov/home/case-information/opinions-orders/

Nothing in this report is fabricated: there are no proceeding numbers, judge panels, institution decisions, final written decisions, settlement terms, or appeal dockets to report, because none exist. If a proceeding surfaces later, the analysis above will need to be rebuilt from the actual papers.

I'll reconstruct the assignment record for US 11402414, starting with the authoritative data in the patent text (which includes the Google Patents legal-events table) and then cross-checking the USPTO Assignment Center.

Let me cross-check for any post-issuance assignments and confirm the correspondent of record for the single recorded assignment.

Inventors

Only one inventor is named, which is ordinary for a narrow circuit-level improvement (a DCR-sensing network for a coupled-inductor buck converter). No departure/attrition pattern is discernible, and none would matter here: the inventor never held title — the patent was assigned to Apple at filing, so there is no "inventor-held → fire-sale" vector.

Original assignee

Apple Inc. (California corporation; principal place of business One Apple Park Way, Cupertino, CA 95014). This is the entity named on the issued patent and the sole assignee of record.

• Product embodying the claims: Apple designs and ships the SoC/PMIC power-delivery silicon to which coupled-inductor DCR sensing is directly applicable (e.g., multiphase buck regulators in Mac, iPhone, iPad and Vision Pro platforms). The specification's own FIG. 5 "system 150" (integrated circuit + peripherals) is the standard Apple SoC context. Whether any specific shipping part reads on the claims is not something the assignment record can establish.
• Primary line of business: consumer electronics / semiconductors / software.
• Current status: operating; one of the largest public companies in the world. No bankruptcy, acquisition, or dissolution event. The patent remains Active, with the 4th-year maintenance fee paid 2026-01-21 (Google Patents legal events) — a strong, independent indicator that Apple still owns and is maintaining the asset.

Assignment timeline

Only one assignment is recorded against US 11402414 (application 16/820,658).

• 2020-03-23 (executed) / recorded 2020-04-01 — Reel 052284/0896
  • Conveyance: Assignment
  • Assignor: Bolus, Jonathan F.
  • Assignee: Apple Inc. (Cupertino, California)
  • Correspondent: Not determinable from the records retrieved. The Google Patents legal-events entry records the reel/frame and the conveying party but does not surface the USPTO correspondent field, and I could not pull the underlying Assignment Center abstract to read the attorney/agent of record. I will not guess at a name or firm. (Apple assignments are typically filed by Apple's in-house IP department and/or its outside prosecution counsel of record; that is a general practice observation, not a verified fact for this reel.)
  • Context: standard employee/at-filing assignment — inventor assigns to his employer; confirmed by the "Effective date: 20200323" entry in the legal-events table. This is a routine employment assignment, not an acquisition or transfer.

No post-issuance assignments are on record. There is no assignment to any IP Holdings / Licensing / Ventures LLC, no security agreement, no merger, no release, and no change of name. Google Patents shows only: (1) the inventor→Apple assignment; (2) publication of US20210285989A1 (2021-09-16); (3) grant (2022-08-02); (4) the 2026 maintenance-fee payment. Apple remains assignee of record.

Timeline diagram

timeline
    title Ownership of US 11402414
    2020 : Filed by Jonathan F Bolus
         : Assigned to Apple Inc reel 052284 frame 0896
    2021 : Application published as US20210285989A1
    2022 : Patent issued as US11402414
    2026 : Apple pays 4th year maintenance fee

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No transfer off the operating company. The only recorded conveyance (reel 052284/0896) runs from the inventor to Apple Inc., an operating corporation. No "IP/Holdings/Licensing/Ventures" successor appears anywhere in the chain.

2. Known asserter in the chain — not present. Apple Inc. is the sole assignee of record (reel 052284/0896) and does not appear on any public NPE/asserter list (RPX, Unified Patents, etc.) — it is a frequent defendant in NPE suits, not an asserter. No Acacia / Marathon / IV / Wi-LAN / Conversant / Pendrell / etc. entity appears.

3. Repeat correspondent across the chain — not present / unclear. There is only a single recorded conveyance, so the "recurrence" test cannot be met even in principle. The correspondent field itself was not retrievable, so no correspondent can be cited. Flagging the recurrence test as unclear only because of the missing field; the substance (one link) makes the signal moot.

4. Cascading transfers — not present. A single assignment in 2020; no chained LLC-to-LLC transfers within any window.

5. Pre-litigation transfer — not present. No infringement suit naming US 11402414 was found, and the only assignment predates grant by two years; it was an employment assignment, not a standing-arrangement transfer.

6. Bankruptcy fire-sale — not present. Apple has no Chapter 7/11 proceeding; nothing in the legal events.

7. Privateering — not present. No transfer from Apple to any assertion vehicle; Apple retains title and is still paying maintenance.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX/AST/LOT/Unified/OIN; it terminates at Apple itself.

Verdict

Insufficient data — per the rubric, "only the original assignment" on record. (Substantively this is the non-NPE outcome: there is no assignment chain to analyze because the patent never left its operating-company owner.)

Justification: The only recorded conveyance is the employment assignment from inventor Jonathan F. Bolus to Apple Inc., executed 2020-03-23 and recorded 2020-04-01 at reel 052284/0896 (Google Patents legal events for US 11402414). No subsequent assignment, security interest, merger, or licensing transfer appears, and Apple paid the 4th-year maintenance fee on 2026-01-21, confirming the original assignee still holds and is maintaining the asset. Because the chain contains no post-issuance link, no NPE signal can be scored present — zero of the eight signals are met, so an NPE verdict is not supportable; the honest label under the rubric is "insufficient data," reflecting a clean, operating-company-owned title rather than a troll chain.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 11402414 or app 16/820,658; the legacy interface at https://assignment.uspto.gov/patent/index.html surfaces the same reel 052284/0896 record).

Caveats: (a) I could not retrieve the correspondent of record for reel 052284/0896 from the sources returned to me — treat that field as unverified rather than absent. (b) Absence of a recorded assignment does not always mean clean title (unrecorded transfers exist), but combined with Apple's active maintenance-fee payments it is strong evidence Apple is the current owner. (c) No evidence was found that US 11402414 has ever been asserted in litigation or at the PTAB.

I'll research patent US 11402414 and its cited prior art. Let me start with searches.

Let me research the remaining cited references.

Prior-Art Analysis — US 11,402,414 B2 ("Direct Current Sensing for Coupled Inductors")

Part A — Confirming the target patent (per strict-ID rule)

Independent claims to test against the art: claim 1 (circuit), claim 11 (method), claim 16 (apparatus). The distinguishing element across all three independents is a first capacitor coupled between the first current sensing network and the second current sensing network — in the dependent claims (4, 17) expressly "between the first sense node and the second sense node." The specification (Eq. 15) states the cross-capacitor value is chosen on the coupling factor k: k = −C3/(C1+C3).

---

Part B — Each cited reference, claim-by-claim

1. US 6,424,129 B1 — Semtech Corporation

• Full citation: US 6,424,129 B1, "Method and apparatus for accurately sensing output current in a DC-to-DC voltage converter," Semtech Corp. Filed 2001-08-21; granted 2002-07-23.
• Description: Buck-type DC-DC converter with an R-C current-sense network in parallel with the output inductor, plus an amplifier that level-shifts the sense signal so it sits above the noise floor. Rooted in the Walters-type (US 5,982,160) DCR technique. It is a single-inductor sensing scheme.
• § 102 analysis: No anticipation. It does not disclose two inductors sharing magnetic fields, two switching networks to two switching nodes, or a capacitor coupled between two current-sensing networks/sense nodes. Relevant only as background art showing DCR sensing is conventional; a § 103 reference at best.

2. US 2003/0102849 A1 — Analog Devices, Inc.

• Full citation: US 2003/0102849 A1, "Multi-phase switching regulator," Analog Devices, Inc. Filed 2002-10-30 (prov. 60/333,404, 2001-11-26); published 2003-06-05. (Granted as US 6,683,441 B2.)
• Description: N-phase regulator with a "total current circuit" that sums the switch-node-side voltages of N current-sensing elements into a summing node and uses one amplifier (feedback Rcs/Ccs, with RcsCcs ≈ L/R1) to produce Vcs proportional to total instantaneous inductor output current; used for adaptive voltage positioning (AVP).
• § 102 analysis: No anticipation. It teaches summing phase currents into a single output, not resolving individual inductor current, and the only capacitor (Ccs) is an op-amp feedback element across a feedback resistor — not a capacitor coupled between two sense networks/sense nodes. Does not disclose "each coupled to its own current sensing network." § 103 relevance for the "switching network coupled to switching node" and "output node" limitations (claims 2, 8, 10, 19).

3. US 7,233,132 B1 — Virginia Tech Intellectual Properties, Inc.

• Full citation: US 7,233,132 B1, "Current sensing in multiple coupled inductors by time constant matching to leakage inductance," Virginia Tech Intellectual Properties, Inc. (Lee, Xu, Dong). Filed 2006-01-30; granted 2007-06-19.
• Description: DCR current sensing explicitly adapted to coupled inductors. An RC circuit (series R + C) is placed across each coupled output inductor, with the RC time constant matched to Lk/DCR (leakage inductance, not self-inductance). The sum of the capacitor voltages is proportional to the sum of inductor currents; a voltage adder combines them. The patent states expressly that it is "generally not possible to accurately determine the current in each inductor" — only the sum.
• § 102 analysis: Closest art on the coupling concept, but not anticipatory of any independent claim. It discloses coupled inductors sharing a core, switching devices (Q1–Q4) at switching nodes, and an RC sense branch per inductor — i.e., it reads on much of claim 1's preamble and the claim 3/5-style resistor-plus-capacitor branches. But it lacks the claimed cross-coupling capacitor ("a first capacitor coupled between the first current sensing network and the second current sensing network," claim 1; "between the first sense node and the second sense node," claim 4). In Virginia Tech the capacitor of each RC branch returns to the output node, and the two branches are only combined at a summing adder. Its own teaching that individual currents cannot be resolved further cuts against anticipation of Apple's individual-sensing claims (4, 5, 17). Claim 6's "capacitance based on coupling factor" is also absent — VT deliberately ignores the mutual/coupling term. Best characterized as a § 103 reference.

4. US 8,228,014 B2 — GM Global Technology Operations LLC

• Full citation: US 8,228,014 B2, "Multi-phase DC/DC boost converter," GM Global Technology Operations LLC. Filed 2009-07-30; granted 2012-07-24.
• Description: (Limited detail retrieved.) Multi-phase DC/DC boost converter architecture. Boost, not buck, and the available material does not evidence the cross-coupled sense capacitor or the two-sense-node topology.
• § 102 analysis: No anticipation on the record available; at most a general multi-phase-converter § 103 reference. Confidence: low on its inner sensing detail — I could not retrieve its full claims, so this is flagged rather than asserted.

5. US 9,281,748 B2 — Lenovo Enterprise Solutions (Singapore) Pte. Ltd.

• Full citation: US 9,281,748 B2, "Operating a DC-DC converter," Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Filed 2012-03-02; granted 2016-03-08.
• Description: (Limited detail retrieved — search returned no substantive content.) Generally directed to controlling/operating a DC-DC converter.
• § 102 analysis: No anticipation on the record available. Confidence: low — I did not obtain its claims or figures, so I cannot responsibly map it element-by-element. Flagged for follow-up rather than asserted.

6. US 2016/0187386 A1 — Texas Instruments Incorporated

• Full citation: US 2016/0187386 A1, "Coupled inductor current sensing apparatus and systems," Texas Instruments Inc. Filed 2014-12-31; published 2016-06-30.
• Description: The most structurally on-point reference. It senses currents through two mutually coupled inductors L1, L2 (each between a switched voltage input terminal and a common apparatus output), using DCR. It discloses (a) a current-sum path: sense resistors RS1, RS2 from each switched input terminal to a summing node 238, with a current-sum sense capacitor CS between the summing node and the output terminal, plus a sum amplifier; and (b) a current-difference path: series resistor+capacitor branches across the inductors and a difference amplifier sensing the voltage difference between the two sense branches. This patent appears in US 11,402,414's "Cited By" family context (the Apple patent is cited in the TI/Infineon family landscape).
• § 102 analysis: Potentially anticipates elements of claim 1's preamble and claims 2, 3, 8, 10, 19 — two mutually coupled inductors, each at a switching node driven by switching circuitry, each with a series sense resistor (RS1/RS2), and sensing outputs proportional to inductor current. However, careful reading shows the central limitation is not met as claimed. In the TI sum path, CS is coupled between the summing node and the apparatus output terminal (a ground/output reference), not "between the first current sensing network and the second current sensing network" (claim 1) nor "between the first sense node and the second sense node" (claim 4). The difference path couples a capacitor in series with a resistor across each inductor, again not node-to-node between the two sense nodes. So on a strict, all-elements-reading, no single claim is fully anticipated; TI is instead the strongest § 103 reference because the cross-branch capacitor is the only meaningful gap and TI supplies the coupled-inductor DCR motivation. Confidence: moderate-high, subject to full-claim verification of the difference-sensing topology.

7. US 9,627,969 B2 — Infineon Technologies Austria AG

• Full citation: US 9,627,969 B2, "Inductor current measurement compensation for switching voltage regulators," Infineon Technologies Austria AG (Tschirhart et al.). Filed 2014-05-27; granted 2017-04-18. (Pub. US 2015/0349634 A1.)
• Description: Digital compensation of DCR-based inductor-current estimates to correct for inductance/DCR variation with temperature and current (meshes with AVP). Uses RC sense networks across inductors and a compensation filter; multi-phase framing.
• § 102 analysis: No anticipation. It presupposes, and improves, DCR sensing; it does not disclose a cross-coupled capacitor between two sense networks, nor the coupling-factor-based capacitor value. § 103 relevance only (context/motivation for accurate coupled-inductor DCR sensing).

---

Part C — Summary ranking and § 102 vs. § 103 conclusion

Forward citations (not prior art): US 2022/0109368 A1 and US 11,876,445 B2 (Infineon, "Trans-inductance multi-phase power converters and control") cite US 11,402,414 — these post-date it and are not available as § 102 art.

Most relevant prior art, ranked:

Bottom line under 35 U.S.C. § 102: On the record examined, none of the seven cited references appears to fully anticipate any independent claim (1, 11, 16) of US 11,402,414, because each fails to disclose the recited first capacitor coupled between the two current sensing networks / sense nodes (and, for claims 6/7/18, the coupling-factor- and capacitance-value basis for that capacitor). The TI reference (US 2016/0187386 A1) is the only one whose topology and claim set come close, but its sense capacitor is coupled to the output terminal rather than between the two sense nodes. The realistic exposure is § 103 obviousness, with US 2016/0187386 A1 alone, or in view of US 7,233,132 B1 and/or US 2003/0102849 A1, as the primary combination.

Caveats (per operating rules): (1) My element-by-element mapping for US 9,281,748 B2 and US 8,228,014 B2 is incomplete because the searches did not return their claims/figures — I flag these rather than assert a conclusion. (2) Anticipation requires a single reference disclosing every limitation arranged as claimed; my "potentially anticipates" notes above are limited to individual elements, not whole-claim § 102 anticipation, and full-claim verification against the TI and Virginia Tech specifications is recommended before relying on this analysis. (3) All patent numbers and dates are reproduced literally from the sources; where a source disagrees with my prior training, I have followed the source.

I'll ground the analysis by pulling details on the closest cited references before building the § 103 combinations.

Key finding: the TI reference discloses a capacitor common to two RC sense networks for coupled inductors. Let me pin down that structure and the remaining cited references.

Now the two remaining core references and the multiphase/control references.

Obviousness Analysis — U.S. Patent No. 11,402,414 B2 (35 U.S.C. § 103)

Built on the prior sections (bibliography, claim construction, technical substance, litigation check). I do not re-derive the specification or re-summarize the abstract below.

---

1. Analytical frame and level of ordinary skill

Legal standard applied: Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; secondary considerations), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007). The dispositive KSR principles here are: (a) a combination of familiar elements according to known methods is obvious when it yields no more than predictable results; (b) if a technique has been used to improve one device, and a PHOSITA would recognize that it would improve similar devices in the same way, using it is obvious; (c) a finite number of identified, predictable solutions to a known problem makes the solution obvious to try; and (d) a mere change in component value or arrangement, dictated by design incentives, is obvious absent unexpected results.

Level of ordinary skill (proposed): A bachelor's degree in electrical engineering (or equivalent) plus 2–5 years of experience designing switch-mode DC-DC converters and their control loops, or a master's degree plus 1–3 years. Such a person is familiar with: DCR (inductor-resistance) current sensing and its time-constant-matching requirement (τ = L/DCR); multi-phase buck topologies with per-phase high-side/low-side switches; coupled-inductor ("trans-inductance") magnetics and the mutual-inductance term kL·di/dt; and PWM/PFM control with peak-current limiting. All references below are presumed analogous art — every one of them is directed to the same field of endeavor (current sensing in switching voltage regulators).

A threshold observation that drives everything below: the four references flagged as examiner-cited on the face of the patent (* in the prior section) — US 6,424,129 B1 (Semtech), US 2003/0102849 A1 (Analog Devices), US 7,233,132 B1 (Virginia Tech), and US 2016/0187386 A1 (Texas Instruments) — are precisely the four that bear most heavily on the independent claims. Two of them (Virginia Tech and TI) are directed to the exact problem the patent purports to solve: DCR sensing in coupled inductors. That the patent nonetheless issued suggests either that the examiner did not fully appreciate TI's "common capacitor" teaching, or that Applicant made an argument I cannot see (the file wrapper was not retrievable — already flagged). It does not mean the art is weak.

---

2. What each reference contributes (with verification status)

Verified DCR baseline (independent of Semtech): TI '386's own FIG. 1A admits as prior art the time-constant-matched DCR sense network for uncoupled inductors ("described by Tateishi in U.S. Pat. No. 6,469,481"), and Infineon '969 describes the same technique in detail. So the "resistor from switching node to sense node + capacitor at sense node, R·C matched to L/DCR" primitive is established prior art on the verified record without needing the Semtech disclosure.

---

3. The limitation that decides this case

Strip the claims to their structural minimum:

• Claim 1 requires: two switching networks → two magnetically coupled inductors at two switching nodes; a first sensing network coupled to the first inductor at the first switching node; a second sensing network coupled to the second inductor at the second switching node; and "a first capacitor coupled between the first current sensing network and the second current sensing network."
• Claims 4 / 11 / 16 narrow that capacitor limitation to one coupled between the first sense node and the second sense node, where each sense network has a resistor between its switching node and its sense node.
• Claims 6 / 7 / 15 / 18 add only a functional-sizing limitation: the capacitor's value is "based at least in part on a coupling factor" (and/or on the other capacitors' values).
• Nowhere do the claims recite the specification's algebraic design rule k = −C3/(C1+C3) or R_S ≈ R_f[C1+C3] = L. That derivation is spec-only. The claimed scope is therefore the arrangement, plus a broad "sized using the coupling factor" functional wrapper.

This claim-drafting choice is the single most important fact for the § 103 analysis: the patent claims the topology and a result-oriented sizing limitation, not the mathematical relationship that the specification derives.

---

4. Combination A — TI '386 alone renders claim 1 obvious (and borders on § 102)

Element-by-element against claim 1 on TI '386 FIG. 2 (all verified):

Because claim 1 says "between the first current sensing network and the second current sensing network" (not "between the two sense nodes"), TI's shared summing capacitor reads directly on it. TI further discloses the function the patent claims for that capacitor — the voltage across it is proportional to the sum of the coupled-inductor currents (V_CS = DCR/2·(I1+I2)), i.e., exactly the DCR proportionality the patent seeks.

Conclusion on claim 1: obvious over TI '386 alone, and the gap to anticipation is narrow enough that a § 102 challenge should also be evaluated (the only plausible distinction is whether a capacitor "common to" two networks is "coupled between" them — a distinction of wording, not structure). Even under the narrower reading, TI's express purpose ("a voltage proportional to a sum of currents flowing through first and second coupled inductors"), applied by modifying the topology trivially, renders claim 1 obvious.

---

5. Combination B — TI '386 + Virginia Tech '132 → claims 4, 11, 16

Claims 4, 11, and 16 require two distinct sense nodes with the coupling capacitor between them. TI's FIG. 2 uses one shared summing node; Virginia Tech supplies the two-distinct-sense-node structure.

Proposed combination: implement Virgina Tech '132's two-RC-network-per-coupled-inductor arrangement (each network having a resistor from the switching node to a distinct capacitor/sense node, RC ≈ Lk/DCR), and bridge the two sense nodes with TI '386's common/coupling capacitor.

Motivation to combine (specific, articulable, and grounded in the references themselves):

1. Identical problem, identical field, same solution family. Both references are DCR-sensing references for multi-phase/coupled-inductor regulators. VA Tech states the problem in the same words the patent uses ("DCR current sensing is presently incompatible with coupled inductors … the capacitor voltage waveform will have a different shape than the inductor current waveform"); TI states it as "field coupling between shared-core inductors may yield undesirable consequences … such coupling may complicate the sensing of current flowing through each of two shared-core inductors." A PHOSITA confronting that problem would look to both.
2. Both references point at the same mechanism. VA Tech's fix is to combine the two sense-network voltages (voltage adder). TI's fix is to share one capacitor across the two sense networks. Meeting in the middle — two networks plus a bridging capacitor — is the natural intersection of the two teachings, not a leap.
3. Predictable result, no new function. The patent's own equation (V_sns = R_S·i_L) is the textbook DCR result that VA Tech, TI, Infineon, and the admitted prior art all say DCR sensing is supposed to produce. The bridging capacitor adds no function beyond the proportionality the field was already seeking; it merely removes the mutual-coupling error term that both references identify.
4. A finite set of known solutions. Facing the k·di/dt error term, a PHOSITA's options were: (i) time-constant matching (VA Tech), (ii) summing/bridging the sense networks (TI), (iii) digital compensation (Infineon '969). Trying (ii) — a single passive capacitor between two nodes of two RC networks — carried a reasonable expectation of success. KSR, 550 U.S. at 421.
5. No criticality in the claims. The claims do not require the spec's k = −C3/(C1+C3) relation, so the combination need not reproduce the derivation to meet the claim language.

Result: claims 4, 11, and 16 would have been obvious over TI '386 in view of Virginia Tech '132 (further in view of the admitted DCR art of TI FIG. 1A / Infineon '969).

---

6. Combination C — the three-reference route to the same place

If one prefers Semtech '129 as the DCR primitive: Semtech '129 (R–C network across the inductor; capacitor voltage = DCR drop → proportional to inductor current; used in a peak-current loop) + Virginia Tech '132 (duplicate/apply that network to each of two magnetically coupled inductors, with time constants matched to leakage inductance) + TI '386 (bridge the two resulting sense nodes with a common capacitor to cancel the mutual-coupling error). Same motivation analysis as Combination B. Because I could not retrieve Semtech '129's full disclosure, the safer formulation substitutes TI '386's admitted FIG. 1A/Tateishi DCR network or Infineon '969's per-phase RC network as the DCR primitive — that route is fully supported on the verified record.

---

7. Full claim-by-claim mapping

Net: no claim of the patent recites a limitation that is absent from the verified art when the art is properly combined. The only arguably uncovered feature is the specific three-capacitor topology (two shunt capacitors plus a bridging capacitor) of claims 5/14/17 — and even that is reached by adding a bridging capacitor (TI) to a conventional per-inductor DCR network that already has its own capacitor (TI FIG. 1A/Tateishi, Infineon '969, VA Tech '132). Combining two known passive networks into one node set is the paradigm of an obvious arrangement of old elements.

---

8. The strongest non-obviousness arguments, stated candidly

A competent patent owner would press these; each is answerable, but they are the real battleground.

(a) Virginia Tech '132 arguably teaches away from individual per-inductor current sensing. VA Tech says the single-RC capacitor voltage "is not accurately proportional to the current flowing through the first inductor," and that only the sum (via the adder) is accurate. Since claim 11 requires the first voltage to be proportional to the current through the first inductor and the second to the current through the second inductor, Applicant could argue VA Tech disparages the very result claimed.
Answer: teaching away requires criticism, discredit, or discouragement of the claimed solution (In re Fulton; MPEP 2145). VA Tech does the opposite — it motivates the field to find a DCR technique "compatible with coupled inductors" and expressly frames the pre-existing belief ("no known method," "presently incompatible") as a problem to be solved. TI '386 then shows that a shared/bridging capacitor does yield a coupled-inductor sense voltage proportional to the inductor currents, and per-inductor current knowledge was plainly desirable (phase balancing in ADI '849; per-phase current estimates in Infineon '969). The "teaching away" is at most a statement about one pre-2006 configuration, superseded by TI in 2016.

(b) The patent's mechanism is counterintuitive because the cross-capacitor contributes with a negative sign to k. An owner could argue that a PHOSITA would not have expected that adding a capacitor between two sense nodes produces a negative effective coupling term (k ≈ −C3/(C1+C3)).
Answer: the claims don't capture the sign or the formula. The claimed limitation is a capacitor between the sensing networks/net-nodes "based at least in part on a coupling factor." TI '386 already discloses exactly that capacitor and its proportionality result; VA Tech already mandates that the network values be a function of the magnetic parameters (Lk, hence k). The derivation is an explanation of a known topology, not a claimed feature.

(c) Claim 5/14/17's three-capacitor topology specifically. The patent's C1 and C2 are shunt-to-ground capacitors, and it is the interaction of C1 plus C3 that yields the correction. Neither TI (one shared capacitor) nor VA Tech (one capacitor per network, returning to the output node) shows all three.
Answer: this is the closest thing to a genuine point of novelty, but it fails under KSR: a PHOSITA starting from a conventional per-inductor DCR network (which necessarily has a capacitor at the sense node) and applying TI's bridging capacitor obtains the three-capacitor arrangement as a matter of course; the result (summed/bridged sense voltage) is predictable; and the claim adds nothing functional over that combination. Secondary considerations would be needed to rescue it, and none appear in the record.

(d) Secondary considerations. Apple's commercial use of coupled-inductor VRs in its SoCs could theoretically support a nexus argument, but there is no evidence in the record I reviewed of unexpected results, industry skepticism overcome, licensing, or copying tied to this claim set. Without a nexus, secondary considerations carry no weight.

(e) A genuine gap in the disclosure chain. The one substantive weakness in the challenger's case is documentary, not legal: my verified mapping for claim 1 rests on the TI pre-grant publication US 2016/0187386 A1 and its granted counterpart US 9,726,697 B2, and the FIG. 2 description I retrieved is clear but the figure itself was not visually inspected. Before relying on the near-anticipation theory for claim 1, confirm from the granted PDF that CS 235 is drawn between summing node 238 and the output terminal 218 (as the text states) and that RS1/RS2 connect to the switched input terminals 205/208 (as the text states).

---

9. Bottom line

• Claim 1: Obvious over US 2016/0187386 A1 (TI) alone. The TI reference discloses every element, including a capacitor shared between/bridging the two coupled-inductor sense networks, and expresses the same proportionality function the patent claims. Evaluate anticipation as well.
• Claims 11 and 16: Obvious over TI '386 in view of Virginia Tech '132 (optionally with the admitted DCR art of TI FIG. 1A/Tateishi or US 9,627,969 (Infineon)). TI supplies the coupling capacitor and the coupled-inductor current-sensing rationale; VA Tech supplies the two distinct RC sense networks per coupled inductor and the value-matching teaching.
• Claims 2–10, 12–15, 17–20: Obvious over the same combinations plus Infineon '969 (multiphase HS/LS switching networks; per-phase RC sense network; controller and current estimator; OCP) and the peak-current-loop teaching of Semtech '129 / the admitted art. The "based on a coupling factor" limitations (6/7/15/18) are met by VA Tech's mandate that the RC value match Lk/DCR, since leakage inductance is itself a function of the coupling factor (Lk = L(1−k)).
• Weakest link for a challenger: claims 5/14/17's three-capacitor topology — but it fails on KSR grounds (routine combination of two known passive sense networks), not because it is inventive.
• Overall assessment: this is a strong § 103 challenge, concentrated almost entirely in US 2016/0187386 A1 / US 9,726,697 B2 (TI) and US 7,233,132 B1 (Virginia Tech). Both were before the examiner; the patent's survival is more likely explained by how the "first capacitor coupled between the first current sensing network and the second current sensing network" language was read during prosecution (broad claim term, or a narrower reading that distinguished TI's shared summing node) than by the absence of the art. I could not verify the prosecution history, which is where that answer lies.

---

Caveats and flags

1. Date discrepancy (flagged per rules): the task header states "Current Date: April 26, 2026," while my operating header gives today as 2026-09-29. Immaterial to the analysis; noted for accuracy.
2. Unverified references: I did not retrieve the disclosures of US 2003/0102849 A1 (ADI), US 8,228,014 B2 (GM), or US 9,281,748 B2 (Lenovo) this session. They are characterized above only from title/field and are not load-bearing for any conclusion. US 6,424,129 B1 (Semtech) was only partially confirmed (via family member EP 1 309 077 B1); the DCR primitive is independently established by TI '386's admitted FIG. 1A/Tateishi art and Infineon '969, both verified.
3. No prosecution history: office actions, rejections, and Applicant arguments remain unavailable; I cannot say which reference drove allowance or what was distinguished. This is the single most valuable missing evidence.
4. No § 102 or § 112 analysis was performed beyond noting the narrow gap between TI '386 and claim 1; the task was limited to § 103.
5. No verified secondary-considerations evidence exists in my record; conclusions would change only with proof of a nexus between the claimed arrangement and an unexpected result.

Principal sources retrieved this session:

• US 2016/0187386 A1 / US 9,726,697 B2 (TI): https://patents.google.com/patent/US20160187386A1/en ; https://patents.justia.com/patent/[9726697](/patent/9726697)
• US 7,233,132 B1 (Virginia Tech): https://patents.justia.com/patent/[7233132](/patent/7233132) ; https://vtechworks.lib.vt.edu/items/fb01e9b4-6532-4e67-9e7a-24f357fab464
• US 9,627,969 B2 (Infineon): https://patents.google.com/patent/US9627969 ; https://patentimages.storage.googleapis.com/96/d7/51/4a3a75fe25c2b1/US9627969.pdf
• US 6,424,129 B1 (Semtech) family context: https://patentimages.storage.googleapis.com/99/ae/9a/0fb44603fe8bfb/EP1309077B1.pdf
• US 11,402,414 B2 (subject): https://patents.google.com/patent/US11402414/en