US 9172270

I'll search for authoritative information on this specific patent number and any related 2026 litigation activity.

Now let me check for any litigation or CAFC 2026 docket activity involving this patent.

Let me check whether this specific patent is asserted in any pending litigation and confirm the assignee details.

US 9,172,270 B2 — Verification Summary

Important caveat on the scope of my search: I searched for the exact identifier 9172270 (and "9,172,270") and did not find it in any Court of Appeals for the Federal Circuit 2026 docket, nor in any of the CyboEnergy district-court complaints surfaced in these results. The CyboEnergy cases I found assert different patents (referred to in the pleadings as the '133, '488, and '489 patents). I could not confirm from these results which specific patent numbers those shorthand references map to. Treat the "no litigation found" conclusion as absence of evidence in the sources I could reach, not as proof that none exists — PACER-level dockets and PTAB filings are not fully indexed by general web search.

Bibliographic data (from the authoritative patent text)

Note on term: a 2013-03-18 filing under a 20-year term would nominally expire 2033-03-18; the listed 2034-03-01 adjusted expiration implies patent term adjustment of roughly one year.

Abstract (verbatim)

"A method and apparatus is disclosed for solar power generation when irradiance is low and unstable due to sunrise, sunset, clouding, partial shading, warped solar module surfaces, moving solar modules, and other low or varying irradiance conditions. A multi-channel solar power inverter connected to multiple solar modules can work in a 'Lunar Power Mode', inverting DC power induced from the sky, street lights, or surrounding environment to AC power by using a unique rotating power pulling technology. This allows the inverter to dynamically adapt to irradiance variations and generate AC power under complex irradiance conditions. A UPS (uninterruptible power supply) works with a DC power supply to provide DC power to the internal electronic circuits of the inverter allowing it to run in normal power, low power, or lunar power modes."

Key terminology point: the specification expressly states that "lunar power does not mean the power has to come from the moon. It simply means that the irradiance is low and unstable." The "lunar" label is a metaphor, not a claim of moon-based generation.

Plain-language overview of each independent claim

Claim 1 — Multi-channel grid-tied inverter (apparatus). A DC-to-AC inverter with at least two DC input ports, one DC-DC boost converter per port, a DC power combiner that parallels the boost outputs (summing currents), a DC-AC inverter stage, and an internal AC powerline feeding the grid. It requires (a) a load interface circuit that filters high-frequency components out of the inverter's AC output; (b) a UPS plus a DC power supply that jointly power the inverter's internal electronics, sourced either from solar module(s) or from the UPS; (c) a digital microcontroller that measures input voltage/current, computes per-channel DC power, runs MPPT, performs inversion/synchronization/current control, handles powerline communications and AC line switching, and — importantly — runs the inverter in normal, low-power, or lunar power mode based on calculated DC input power and selects the DC supply or UPS accordingly; (d) a powerline modem for two-way data; (e) a line sensing circuit detecting grid phase and zero-crossing; and (f) a solid-state switch that isolates the internal AC powerline from the grid during non-generation times. This is the broadest multi-channel, grid-interactive claim and explicitly recites the three-mode architecture.

Claim 4 — Single-channel grid-tied inverter (apparatus). The same enumerated architecture as claim 1, but with only one DC input port, one boost converter, and no DC power combiner. Critically, the microcontroller is limited to running normal or low-power mode — the lunar power mode is omitted. This is consistent with the specification's statement that a single-channel inverter "will not work in the lunar power mode" because rotating power pulling requires multiple channels.

Claim 6 — Method of powering the inverter's internal electronics. A three-step method: (a) enter and exit a normal power mode and a low power mode based on measured DC input power; (b) in normal mode, supply the internal electronics through a DC power supply fed from one or more solar modules; (c) in low power mode, supply them through the UPS; and (d) charge the UPS's rechargeable batteries during normal mode. This claim is method-focused and does not itself require the full inverter architecture (dependent claim 9 adds that architecture).

Claim 10 — m-channel solar power inverter (apparatus). At least two DC input channels, each comprising a DC-DC boost converter plus measurement circuits, supporting circuits, cables and connectors; an AC output port to grid or load; a DC power combiner; a UPS plus DC power supply for internal electronics; and a digital microcontroller that measures input voltage/current to calculate per-channel DC input power and is constructed to run normal, low-power, or lunar power mode based on that calculated power. Compared with claim 1, this claim omits the line sensing circuit, powerline modem, solid-state switch, and load interface circuit requirements — making it the more architecturally permissive of the two multi-channel apparatus claims.

Claim 14 — Multi-channel off-grid "AC Master" inverter (apparatus). Claim 1-type architecture but aimed at an AC load rather than the grid, and adding: (a) a load detector that checks the impedance of the connected AC load; and (b) a microcontroller tasked with checking that impedance against specifications, initially energizing a dead AC powerline, continually delivering AC power so that other off-grid inverters on the same line can synchronize to it, determining whether the load is too large or too small, and shutting off with an error signal if so. Note that the claim's mode-control language is internally redundant — it recites "running the power inverter in normal or low power mode... based on calculated DC input power, running the power inverter in a normal mode, a low power mode, or a lunar power mode..." — which reads as a drafting artifact rather than a distinct limitation.

Claim 16 — Method of rotating power pulling. A three-step method of pulling power from input channels of a multi-channel solar inverter "in rotation": (a) check energy availability for each input channel; (b) define a power pulling time window based on AC signal frequency; and (c) pull power only from the channels that have sufficient stored energy during that window. Dependent claim 17 specifies the window is based on the half-cycle or full-cycle of the grid AC sinewave; claim 18 specifies ~10 or 20 ms for 50 Hz and ~8.33 or 16.66 ms for 60 Hz. This is the core "rotating power pulling" claim — the mechanism by which the inverter is described as dynamically skipping under-charged channels to avoid the sagging "energy starving" waveform shown in FIG. 3.

Practical summary of the invention

The patent addresses a known failure mode: when irradiance collapses, pulling current from a PV module drags its voltage down along the I-V curve, starving the inverter's own internal 15 V/3.3 V/5 V supplies and forcing a shutdown — which under UL 1741/IEEE 1547 triggers a mandatory five-minute reconnect wait. The two-part solution disclosed is (1) an internal UPS with rechargeable batteries that powers the inverter's control electronics during low/low-and-unstable irradiance so the inverter stays connected and synchronized, and (2) "rotating power pulling," in which the microcontroller polls each channel's boost-converter storage-capacitor charge at the start of each AC half-cycle and draws power only from channels that are sufficiently charged. The specification reports tested operation with a channel delivering as little as 1.5 W DC (14 V, 0.11 A), roughly 0.6% of a 250 W module's rating. It also contemplates scaling to 100+ channels if boost converters are integrated into ICs, and offers a night-time soldier-clothing application using 50 thin-film modules.

Uncertainty disclosures

• Federal Circuit / 2026 dockets: No CAFC 2026 appeal involving US 9,172,270 was found. I cannot rule out pending or unindexed proceedings.
• Litigation: CyboEnergy, Inc. is an active plaintiff (cases against, among others, Northern Electric Power Technology, Home Depot, Duracell Power Center, Yotta Energy, Deye ESS Technology, and a 2026 case against Generac Power Systems, 4:26-cv-07159). Those complaints identify the asserted patents as the '133, '488, and '489 patents — not the '270 patent — based on the excerpts available to me. I could not verify the '133/'488/'489 numbers against the register from these results.
• Status data: The "Active / expires 2034-03-01" information comes from Google Patents' legal-status field, which that page itself labels an assumption and not a legal conclusion. Confirm maintenance-fee and term status on USPTO Patent Center before relying on it.
• Date basis: The docket information retrieved reflects page-capture timestamps up to early 2026; more recent developments may exist.

I'll research litigation involving US Patent 9,172,270. Let me run several targeted searches.

Initial results are mostly noise. Let me search more specifically for the assignee and patent in litigation databases.

Let me dig deeper to see whether the '270 patent specifically was asserted, and identify all CyboEnergy cases.

Answer: No litigation asserting US Patent 9,172,270 was found

Bottom line: Based on my searches of the available litigation sources (CourtListener/RECAP, PACER-derived dockets, Justia, Docket Alarm/Unified-Patents-style aggregators, and the Ramey LLP firm's own announcements), I could not identify any lawsuit, ITC investigation, PTAB proceeding, or CAFC appeal in which US 9,172,270 ("Smart and scalable lunar power inverters") was asserted, challenged, or otherwise at issue. To the extent US 9,172,270 has been litigated, I found no evidence of it.

I want to be explicit about a limitation: mid-research I hit a tool/search cap, so my last four planned queries did not run. This is therefore a negative finding based on the searches I completed, not a certified exhaustive clearance search. Given the strict-ID instruction, I have not substituted results for similar patent numbers (see caution below).

---

Why a "no results" answer is notable here (not a search failure)

US 9,172,270 is owned by CyboEnergy, Inc. (application 13/846,708; filed 2013-03-18; granted 2015-10-27; adjusted expiration 2034-03-01). CyboEnergy is an active and repeat patent enforcer, so the absence of a '270 suit is a real data point rather than a sign I searched the wrong party. What CyboEnergy has actually asserted in court is a different set of patents — specifically the '133, '488, and '489 patents — which are separate, related-company patents, not US 9,172,270.

CyboEnergy enforcement activity I did find (none asserting the '270 patent)

None of cases 1–6 involve US 9,172,270.

I also saw reference (in a Chinese-language trade report on the Deye suit) to CyboEnergy having previously sued Hoymiles (禾迈) and APsystems/Altenergy (昱能) — the latter is case #1 above. I could not obtain the docket for a Hoymiles case before my search cap, and I found no indication that the '270 patent was part of any of those matters.

---

Important caution against number-confusion (per your strict-ID rule)

My searches repeatedly surfaced unrelated uses of the string "9172270" that are not this patent and should be excluded:

• A Czech insolvency/credit-agreement number ("úvěrová smlouva číslo 9172270," KSHK 41 INS 15359/2021) — a loan contract number, not a patent.
• A Brazilian court document ID "ID 9172270" (TRF-3).
• A Delhi University B.Com. exam roll number "9172270."

Similarly, do not conflate the target with its CyboEnergy siblings that are being litigated: US 8,786,133, US 9,331,488, and US 9,331,489. These are distinct patents with distinct numbers; only the '133/'488/'489 have appeared in the dockets above.

---

Foreign counterparts of US 9,172,270

The '270 patent has family counterparts:

• EP 2,976,833 (EP2976833A1 published 2016-01-27; EP2976833A4 published 2017-01-11) — status: withdrawn / not active
• WO 2014/153301 A1 (PCT/US2014/030888) — status: ceased / not active

I found no opposition, cancellation, or nullity proceeding involving these counterparts. The inactivity of the EP/WO members is consistent with the absence of litigation on this specific patent family.

---

What I could not verify (honest caveats)

• I did not complete exhaustive queries against Unified Patents' litigation portal, Docket Navigator, or full PACER, so I cannot rule out a short-lived or obscure filing that never populated the free aggregators.
• I could not confirm the defendant name in case #2 (E.D. Tex. 2:23-cv-00311-JRG).
• I found no PTAB (IPR/PGR/CBM), ITC Section 337, or CAFC activity tied to the '270 patent. (Note: PTAB hits for other "'227"/"'270" patents surfaced in search but are different patents and are not attributable to US 9,172,270.)

Recommended next sources to close the gap

1. Unified Patents litigation portal — portal.unifiedpatents.com/litigation/caselist, filter by patent 9,172,270.
2. USPTO Patent Trial and Appeal Board — search "9,172,270" in PTAB decisions.
3. PACER / Docket Navigator — full-text search "9,172,270" across all district courts and the Court of Federal Claims.
4. USITC EDIS — search the patent number for any Section 337 complaint.

Conclusion: As of this research, there is no known litigation involving US Patent 9,172,270. The patent's owner, CyboEnergy, is an active litigant, but it has asserted its related patents ('133/'488/'489) rather than the '270 patent.

I'll verify the ODP default against live sources before finalizing.

Let me run a few more targeted checks before concluding.

One more confirmation pass on the specific patent number.

Proceedings overview

Zero AIA trial proceedings on file for US 9,172,270. The structured "PTAB proceedings on file" block — sourced from the USPTO Open Data Portal and treated here as canonical — reports no IPR, PGR, or CBM naming this patent, and independent web checks (Google Patents family/litigation tab, PTAB docket aggregators, Unified Patents PATROLL contest pages, and CyboEnergy's own district-court dockets) surfaced no proceeding number to add. Breakdown by status is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied.

I want to be explicit about what I did not find, because the absence is the headline: no petition has ever been filed against US 9,172,270. To be clear about my confidence level — I can confirm no proceedings appear in ODP or in the public PTAB/FWD/CAFC material I could reach; I cannot rule out a very recently filed petition that has not yet been publicly posted, and I found no Federal Circuit appeal touching this patent either.

Bottom-line defensive posture: this is not a "patent that has survived IPRs and is hardened" in the sense of a tested-and-upheld validity record — it is an untested patent. All 19 claims (1–19) stand exactly as issued on 2015-10-27, unamended, with no adverse PTAB record, no statutory disclaimer, and no certificate of correction narrowing them. For a defendant, that cuts both ways: there is no FWD to hide behind, but there is also a completely clean slate for a first-filed IPR with no § 325(d) "previously presented art" problem, no § 315(e)(2) estoppel, and no Patent Owner track record of defending these claims at the Board.

---

Per-proceeding detail

None — no proceeding exists for US 9,172,270

There is no proceeding section to write. I am not supplying placeholder docket numbers, petitioner names, panels, or FWD outcomes, because any such content would be fabricated. Per the operating instruction, the default holds: no PTAB activity on file.

---

Adjacent-proceeding context (different patents — do not conflate)

The reason the answer is worth double-checking is that patent owner CyboEnergy, Inc. is a highly active and repeatedly-challenged assertor — just not of this patent. The AIA activity and the litigation are concentrated on three sibling patents in the same inverter family:

What is publicly established about the siblings:

• IPR2024-01106 — Hoymiles USA, Inc. et al. v. CyboEnergy, Inc. Filed 2024-06-27; challenged claims 15–24 of US 8,786,133; Terminated–Settled per the board's 2024-10-28 decision. The parties filed a joint motion to terminate with a confidential license/settlement agreement (Exhibit 1011) filed under seal as business confidential information under 35 U.S.C. § 317(b) and 37 C.F.R. § 42.74(c). Institution decision date is shown as "-", i.e., the case settled before institution was decided. (case summary)
• IPR2024-00362 — Altenergy Power System USA, Inc. v. CyboEnergy, Inc. Filed January 2024 against US 8,786,133 (Paper 5, Notice of Accepting Corrected Petition, mailed 2024-01-22). Petitioner counsel Quarles & Brady; Patent Owner counsel DLA Piper. (PTAB Paper 5)
• IPR2024-00360 — Altenergy Power System USA, Inc. v. CyboEnergy, Inc. Same petitioner, filed contemporaneously; an exhibit lodged in that proceeding is the claim-construction decision from CyboEnergy, Inc. v. Altenergy Power Sys. USA, Inc., No. 6:22-cv-01136-KC (W.D. Tex. 2023-12-20), which construed claim 14 of the '489 patent. I have not independently confirmed the challenged patent/claims for -00360, so treat the '489 attribution as probable but unverified.
• Unified Patents (PATROLL) ran separate crowdsourced prior-art contests on US 9,331,488 (claim 16) and US 9,331,489 (claim 14), each with a $2,000 prize and a Sep–Oct 2026 contest window — i.e., a defensive aggregator is actively hunting art against the sibling patents right now. (’488 contest; ’489 contest)

Critical distinction for the reader: Unified Patents' contests and the Hoymiles/Altenergy IPRs target '133, '488, and '489 — not '270. There is no defensive-aggregator contest, no IPR, and no PGR against US 9,172,270 as of the last accessible data.

---

Strategic summary

Claim status of US 9,172,270: everything is UNTESTED. No claim has been canceled, disclaimed, confirmed, or construed by the PTAB. The patent issued 2015-10-27 with 19 claims, comprising: claims 1–3 (multi-channel grid-interactive inverter, incl. MFA-controller/MFA-optimizer limitations in claim 3), claims 4–5 (single-channel grid-interactive inverter), claims 6–9 (method of supplying DC power to internal electronics via solar modules in normal mode / UPS in low-power mode), claims 10–13 (m-channel inverter, with claim 13 reciting a rotating power pulling routine), claims 14–15 (multi-channel off-grid "AC Master" inverter with load detector), and claims 16–19 (method of rotating power pulling, with claim 17 reciting half-cycle or full-cycle windows and claim 18 reciting the 10/20 ms and 8.33/16.66 ms windows). Legal status is Active with an adjusted expiration of 2034-03-01 (patent term adjusted from the nominal 2033-03-18), so the assertion window is long. The EP counterpart EP2976833 was withdrawn and the PCT WO2014153301A1 went not_active/ceased, which means US-only enforcement — there is no European validity ruling to borrow and no parallel nullity result to cite.

Estoppel landscape is a blank sheet. Because no IPR/PGR was ever filed against '270, § 315(e)(2) estoppel does not attach and there are no barred grounds. A defendant today can raise any prior-art combination and any § 102/§ 103 theory without a § 325(d) "same or substantially the same art previously presented to the Office" discretionary-denial argument, and without the risk that a prior petitioner's privies already burned a reference. The entire universe of '133/'489 art asserted elsewhere (including art developed in the Unified PATROLL contests and the Altenergy/Hoymiles petitions on the siblings) remains freely available against '270, subject to the obvious caveat that claim language differs materially between '270 and its siblings and art must be mapped to '270's own limitations — notably the UPS-based internal power architecture (claims 1(h), 4(g), 10(d), 14(i)) and the rotating power pulling technique (claims 13, 16–19), which are '270-specific hooks.

Pattern signals. CyboEnergy is a serial assertor (Sensata, Altenergy, Hoymiles, Aptos Solar, Duracell, Home Depot, Yotta Energy, Deye, Generac) and has been litigating since at least 2021. It has proven willing to settle early and quietly — Hoymiles terminated IPR2024-01106 pre-institution via sealed license, and Altenergy (华州昱能) paid a disclosed $300,000 settlement with roughly $5.52M in its own legal fees. CyboEnergy's litigations have been weakly pleaded on direct infringement: courts in W.D. Tex. and E.D. Tex. repeatedly dismissed direct-infringement theories of the '133 patent because CyboEnergy accused individual inverters rather than the claimed multi-inverter systems (CyboEnergy v. Hoymiles, 2:23-cv-00311-JRG, Dkt. 25). Notably, '270 has never been asserted in any of the located suits — the assertion portfolio has been '133, '488, and '489. There is no Federal Circuit appeal by CyboEnergy on any '270 issue, because there has been no '270 Board decision to appeal.

---

Recommended next steps

If you are a defendant facing assertion of US 9,172,270. My honest framing: there is no FWD to link to and no canceled claims to quote, and I will not invent one. The defensive opportunity is the inverse of the usual story — you would be the first petitioner. Concretely:

1. Run a § 102/§ 103 search now, before the § 315(b) one-year clock (service of the complaint) runs out. With no prior petition, you avoid § 325(d) and § 315(e)(2) entirely and can pick the best art and the best expert unencumbered. Focus the search on (a) UPS-backed internal supply rails in a multi-channel PV inverter and (b) per-half-cycle channel rotation / capacitor-charge-based channel selection — the two features most distinctive to '270 over the '488/'489 siblings already under attack.
2. Mine the sibling proceedings' art. The Unified PATROLL contests on '488/'489 and the Altenergy/Hoymiles petition papers on '133 are public or discoverable and are near-certainly relevant to '270's overlapping disclosure. Cross-map that art to '270's claims rather than re-using it verbatim — the claim language does not match.
3. Check the specification/prosecution history for § 112 issues. Claim 14 as issued contains an internally inconsistent double-recitation of the microcontroller's modes ("…running the power inverter in normal or low power mode based on calculated DC input power, running the power inverter in a normal mode, a low power mode, or a lunar power mode…"), which is worth an indefiniteness look — though note the Board's institution standard requires a how-the-claim-is-to-be-construed showing, and 35 U.S.C. § 112 challenges in IPR are limited to § 112(b) indefiniteness against a proposed construction.
4. Mind the litigation posture. CyboEnergy's pleadings have been vulnerable to 12(b)(6) attacks on the '133 system claims (accusing individual inverters, not the claimed system). If '270's multi-channel claims are asserted the same way, the Hoymiles/Altenergy dismissal reasoning is directly reusable — that is a cheaper first move than an IPR.
5. If a proceeding appears after today's date, re-check ODP/PTAB E2E before filing: a later-filed petition on '270 by another party can trigger § 315(e)(2) estoppel against you as a privy if there is common ownership or control, and can change the § 325(d) landscape.

Trial-stage milestones: not applicable — with zero proceedings, there is no institution-decision deadline, no oral hearing date, and no statutory 1-year FWD due date to track. If you file, the clock is: 3 months to a Patent Owner preliminary response (or waiver), and a § 314(b) institution decision within 6 months of the petition, with the FWD due within 1 year of institution.

The signal itself. Well-asserted patents eventually attract IPRs — and CyboEnergy's other patents certainly have. The fact that '270 has attracted none across a decade of active enforcement is meaningfully ambiguous: it may mean '270 is being held back as a licensing/blocking patent rather than an assertion patent, or that competitors have simply not had it asserted against them. It should not be read as a validity endorsement. Note that '270's closest sibling, the '489 patent, is under an active PATROLL prior-art hunt with a contest window closing 2026-10-17 — art developed there may transfer to '270, and that is the single fastest-moving development to monitor.

I'll research the assignment/ownership record for US 9,172,270 across the USPTO Assignment Center, Google Patents legal events, and asserter directories.

Let me pull the Google Patents legal-events/assignment data and look for the recorded assignment details.

Let me search for the specific assignment record details and any security agreements or transfers.

Assignment & Ownership Analysis — US 9,172,270 B2 ("Smart and scalable lunar power inverters")

Scope caveat stated up front: I attempted to pull the record copy from the USPTO Assignment Center (assignmentcenter.uspto.gov) and the legacy Assignment Search (assignment.uspto.gov), and those targeted queries returned no retrievable record text before I hit my tool-step cap. What follows is therefore reconstructed from (a) the authoritative patent text, (b) the Google Patents legal-events block for this patent, and (c) CyboEnergy corporate/press materials. I did not retrieve a reel/frame number, and I will not invent one. Where a field is unfilled, I say so rather than filling it.

---

Inventors

Employer determination. Both inventors are CyboSoft/CyboEnergy personnel. CyboEnergy publicly describes itself as "a subsidiary of CyboSoft, General Cybernation Group Inc." and its CEO is Dr. George Cheng (CyboSoft release, July 2, 2013; CyboEnergy support note, Jan 16, 2026).

Unusual-pattern check — NEGATIVE (no departing-inventor signal). The classic fire-sale precursor (all inventors leaving the original assignee within 12 months of filing) is not present. Both inventors continued to co-invent on CyboEnergy filings for a decade after the 2013-03-18 filing — e.g., Cheng + Mulkey + Chow on applications filed 2015-02-18 (US 9,871,379), 2018-02-15 (US 2018/0191167), 2019-08-22 (US 11,258,267), and 2022-01-13. Continuity, not exodus.

---

Original assignee

CyboEnergy, Inc. — Rancho Cordova, California (US). An operating company, not a holding vehicle.

• Product embodying the claims: Yes. CyboEnergy ships the CyboInverter line of solar power mini-inverters — e.g., the 4-channel grid-interactive CI-Mini-1000A (240 V, 60 Hz, 960 W) and the off-grid CI-Mini-1000T, later the CIM-1200N/N1a/N2a/H/Ya/Qa/Sa/Ta/Wa series. These are UL 1741 / IEEE 1547 certified (E113426), CSA 107.1, FCC Part 15 Class A, NEMA 6 (IP67), made in the USA, with per-channel MPPT — i.e., they practice the multi-channel architecture and low-power-mode concepts of this family (CyboEnergy Residential Solar System Design Guide, May 2024).
• Primary line of business: Development, manufacturing, marketing and servicing of grid-interactive and off-grid solar power inverters and microgrid/EV-charging systems. Parent/affiliate CyboSoft, General Cybernation Group, Inc. (founded 1994) is a control-technology company (Model-Free Adaptive control).
• Current status: Operating. Still publishing product support and new-product announcements as of 2026 (SunBandit replacement hardware note, Jan 16, 2026; battery-less off-grid EV charging, Jan 2, 2023; 2024 design guide). Privately held — no SEC 10-K/8-K filings (not a reporting company). No bankruptcy, receivership, or dissolution found. Two Frost & Sullivan awards (2013, 2017).

---

Assignment timeline

Only one assignment is recorded in the sources reachable to me. Google Patents' legal-events block for US 9,172,270 shows exactly one reassignment event:

• 2013-03-18 (executed) / recorded 2013-03-18 — Reel [NOT RETRIEVED] / Frame [NOT RETRIEVED]
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: CHENG, GEORGE SHU-XING; MULKEY, STEVEN L.
  • Assignee: CYBOENERGY, INC. (Rancho Cordova, CA)
  • Correspondent: Not retrieved. I could not read the record copy to capture the attorney/firm of record for the recording. I will not name one speculatively. (For context only: later CyboEnergy applications list Stout, Uxa & Buyan, LLP as agent of record, per PatentGuru — but that is not confirmation of the correspondent on this 2013 recording.)
  • Context: Original assignment of the inventors' rights to their employer, executed on the application filing date.

No post-issuance assignment, security agreement, merger, change-of-name, license, release, or correction is recorded or reported for this patent. Google Patents lists the current assignee as CyboEnergy, Inc. — the same entity that received the 2013 assignment. The EP (EP 2 976 833, withdrawn) and PCT (WO 2014/153301, ceased) family members likewise show no transfer out.

If you need the reel/frame and correspondent, run the exact-number search at USPTO Assignment Center (or the legacy Assignment Search) for 9172270 / application 13/846,708. The record copy will show the reel/frame and the recording correspondent. My queries on those pages returned no retrievable text in this session — that is an access failure, not evidence of absence.

Note on the corporate relationship. CyboEnergy is described as a subsidiary/affiliate of CyboSoft, General Cybernation Group, Inc., but no assignment to General Cybernation Group or CyboSoft is of record for this patent. If an intra-group transfer exists, it is not reflected in the assignment data I could reach.

---

Timeline diagram

timeline
    title Ownership of US 9172270
    2013 : Inventors assign to CyboEnergy Inc
         : Application filed on same date
    2014 : PCT and EP family members filed
    2015 : Patent issued to CyboEnergy Inc
    2017 : EP counterpart withdrawn
    2022 : CyboEnergy asserts sibling patents
    2026 : CyboEnergy still operating and selling CyboInverters

---

NPE / troll-pattern signals

---

Verdict

Operating-company assertion.

The entire chain is one link — inventors Cheng and Mulkey assigning to their employer CyboEnergy, Inc. on 2013-03-18, the application filing date — and CyboEnergy remains the assignee of record to this day. CyboEnergy is a genuine operating company that manufactures and sells the UL 1741–certified CyboInverter product line practicing this family's multi-channel architecture, and it has asserted its sibling patents in its own name against competitors (Altenergy, Hoymiles, Deye, Yotta, Generac) — a non-NPE posture. None of the eight NPE signals is present, and there are no cascading transfers, shell entities, or pre-litigation assignments to weigh against that.

One important qualification: this verdict describes the owner's character, not this patent's litigation history. US 9,172,270 itself has never been asserted; the enforcement campaign runs on the '133 / '488 / '489 siblings. The patent has simply never left the hands of the operating inventor-assignee.

Verify at: USPTO Assignment Center — search 9172270 · Legacy Patent Assignment Search

---

Uncertainty disclosures

• Reel/frame number: NOT retrieved. I could not read the recorded assignment document. Do not treat any reel/frame as confirmed; the field above is explicitly blank.
• Correspondent of record: NOT retrieved. No attorney/firm is stated for the 2013 recording. Any name I might have supplied would be fabricated.
• Negative findings are search-limited. "No security agreement / merger / change-of-name / license" is a negative finding from the sources I could reach, not a certified clearance of the Assignment Center docket. The Assignment Center and legacy search pages returned no retrievable text in this session.
• Corporate-status data comes from CyboEnergy/CyboSoft self-published materials (2023–2026) — reliable for "the company says it is operating and selling product," but not an independent registry check. Privately held, so no SEC cross-reference is available.
• No reconciliation needed with prior sections. This analysis is consistent with the earlier Litigation, PTAB, and Prior-Art sections: same assignee, same family mappings ('133 = US 8,786,133; '488 = US 9,331,488; '489 = US 9,331,489), and the same conclusion that '270 is unasserted and un-challenged.

I'll verify the citations against live sources before analyzing them. Let me run targeted searches.

Prior-Art Analysis — US 9,172,270 B2 ("Smart and scalable lunar power inverters")

1. Verification of the target patent

I confirmed US 9,172,270 B2 on the authoritative sources for the literal identifier 9172270 (no similar-number substitution):

• Google Patents: https://patents.google.com/patent/[US9172270B2](/patent/US9172270B2)/en — issued 2015‑10‑27, app. 13/846,708, filed 2013‑03‑18, CyboEnergy, Inc., inventors Cheng & Mulkey.
• USPTO-style PDF of the granted document: https://patentimages.storage.googleapis.com/cd/d3/4e/7be86175356175/US9172270.pdf
• Pre-grant publication US 2014/0265595 A1.

The ten backward references below are the "References Cited" printed on the face of the '270 patent, plus one family-cited Korean document and one non‑patent citation (the PCT search report). These are the citations to analyze. They are not forward citations (documents citing the '270), which are not prior art.

Search-completeness caveat: I independently re-verified five of these references on live pages (JP2001008383A, US20090174259A1, US20100231045A1, US20110012430A1, plus the PCT/EP family). My tool budget ran out before I could re-open individual pages for the remaining five (US20110148195A1, US20120212065A1, US20120313443A1, US20130002031A1, US20140252862A1, US20140265584A1, KR101097260B1). For those, the descriptions below are anchored to the authoritative patent text (the '270's own incorporation-by-reference paragraphs and the citation table) and to each document's title/assignee/date as printed in that table. Treat the unverified descriptions as lower-confidence.

2. The § 102 framework that applies

US 9,172,270 has an effective filing date of 2013‑03‑18 and claims no earlier priority. Because that date is on/after the AIA's 2013‑03‑16 cutoff, the AIA version of § 102 governs:

• § 102(a)(1): prior art if patented/published/on-sale/publicly used before 2013‑03‑18.
• § 102(a)(2): a U.S. patent or published application is prior art if it was effectively filed before 2013‑03‑18 — even if it published afterward.
• § 102(b) exceptions then remove certain disclosures, including, critically, § 102(b)(2)(C): commonly owned subject matter. This matters enormously here because most of the listed references are CyboEnergy's/Cheng's own applications.

A § 102 anticipation requires one reference disclosing every element of a claim, arranged as claimed. Nothing I reviewed meets that bar. The references are best characterized as § 103 (obviousness) material, with two narrow exceptions worth flagging (discussed in § 5).

3. Summary table

4. Reference-by-reference

(1) JP 2001‑008383 A — Nissin Electric Co., Ltd., "Photovoltaic power generating set"

• Full citation: JP 2001‑008383 A; JP app. 11‑173464 (filed 1999‑06‑21); published 2001‑01‑12. https://patents.google.com/patent/JP2001008383A/ja
• Verified description (translated abstract): A PV array (solar cell 15) feeds a first inverter (17); a rectifier (19) plus storage battery (20) feed a second inverter (22); an input bridging switch (23) and a common AC output circuit (24) supply a critical load (28) without interruption. Control circuit (30) has two modes: normal operation — open switch 23, run inverter 17 at MPPT of the solar cell, run inverter 22 in CVCF control; and shortage operation — when CVCF supply is insufficient, switch so that solar cell + rectifier + battery become a common DC source driving both inverters.
• Claim mapping / § 102: Relevant to the concept of a battery-backed, uninterruptible PV inverter with two operating regimes. But the claim‑6 method is expressly about powering the inverter's own internal electronic circuits — this reference powers an external critical load, not the inverter's control electronics. It also lacks the multi‑channel boost/combiner architecture and rotating power pulling. No anticipation. Best used as § 103 art for the "battery backup + mode change on insufficient PV" idea behind claims 1(h)/(i) and 6.

(2) US 2009/0174259 A1 — Shan‑Cheng Lin (Tranergy Technologies), "Natural Energy Power Supplying Apparatus"

• Full citation: US 11/971,864, filed 2008‑01‑09, published 2009‑07‑09; abandoned. https://patents.google.com/patent/US20090174259
• Verified description: A natural‑energy source (solar module or wind generator) feeds an inverter with an MPPT device; a UPS control circuit with an AC input, AC output and DC output is paralleled with the utility; a deep‑cycle battery is charged via a charging control circuit; an electromagnetic switch reconfigures the circuit depending on whether the utility is present. MPPT may be implemented in a DSP.
• Claim mapping / § 102: Directly discloses a UPS + battery charged from a renewable source and MPPT, i.e., the genus of claim 6(c)/(d) and claim 10(d)/12. However, it has no multi‑channel DC combiners, no per‑channel boost converters, no load interface circuit filtering, no powerline modem, and nothing resembling lunar mode or rotating power pulling. No anticipation of any independent claim; a supporting § 103 reference for the UPS/battery‑charging limitation.

(3) US 2010/0231045 A1 — First Solar, Inc., "Photovoltaic Power Plant Output"

• Full citation: US 12/706,081, priority 2009‑02‑13, published 2010‑09‑16; later granted US 8,492,926. https://patents.google.com/patent/US20100231045
• Verified description: Utility‑scale PV plant with an auxiliary power source (battery/energy storage, fuel cell, generator) controlled to limit the plant's rate of change of output power (ramp rate) caused by cloud passage; measures irradiance/PV power and adjusts the auxiliary source. Explicitly frames the problem as "short‑term variability" from clouds.
• Claim mapping / § 102: Relevant only as evidence that smoothing PV output under variable irradiance via storage was known. It addresses plant‑level output smoothing, not powering a micro‑inverter's internal electronics, and contains no multi‑channel combiner or rotating power pulling. No anticipation; weak § 103 relevance at best (possibly background for claim 1(i)'s "monitoring ... for generated power amount and status").

(4) US 2011/0012430 A1 — General Cybernation Group, Inc. (Cheng & Mulkey), "Smart and scalable power inverters" — ⚠️ the closest structural reference

• Full citation: US 12/837,162, filed 2010‑07‑15, claiming benefit of provisional 61/226,141 (2009‑07‑16); published 2011‑01‑20; granted as US 8,786,133 B2 on 2014‑07‑22. This is the family member the '270 itself incorporates by reference and is the subject of the CyboEnergy enforcement campaign (see the Litigation summary). https://patents.google.com/patent/US20110012430A1/en
• Verified description: Smart single-, dual-, triple-, quad- and multi‑input inverters that "daisy chain," each inverting DC from PV modules to single‑ or three‑phase AC. The figures/description include a DC‑DC boost converter, a DC power combiner, a DC‑AC inverter, a load interface circuit, a solid‑state switch, an MFA microcontroller (with MPPT), a powerline modem, and an interface circuit, plus AC line sensing/synchronization.
• Claim mapping / § 102: This reference discloses essentially all of the hardware architecture of claim 1 (a–g, j, k, l) and claim 10 (a–c) — multi‑channel inputs, per‑channel boost converters, a paralleling DC combiner, DC‑AC inversion, load interface filtering, solid‑state isolation switch, and a microcontroller performing MPPT and powerline communications. What it does not disclose is the '270's point of novelty: (h) the UPS + DC power supply that jointly power the inverter's internal electronics (the '133 instead derives internal supplies from the AC line, as the '270 itself acknowledges at ¶[0026]), and (i) the normal/low/lunar three‑mode operation with supply selection. No full § 102 anticipation, but this is the primary § 103 base reference — combined with a UPS/battery reference (JP '383, Lin '259, or Samsung '195), it renders the apparatus claims' architectures obvious; the only remaining question is whether the lunar-mode/rotating-power-pulling limitations (claims 10, 13, 16–19) survive, since the '133 does not disclose them.

(5) US 2011/0148195 A1 — Samsung SDI Co., Ltd., "Energy storage system and method of controlling the same"

• Full citation: priority 2009‑12‑23, published 2011‑06‑23. (Listed as a reference of record on the '270 face; not independently re-opened this session.)
• Description (from title/assignee/date + family context): A grid‑connected energy‑storage system with a battery, a power‑conversion system, and a controller that transitions between grid‑connected and standalone/discharge operation, with the battery charged from the grid/PV and discharged to loads.
• Claim mapping / § 102: § 103 reference for the "UPS/battery + DC supply, selected by operating mode" concept in claims 1(h)/(i), 6, and 10(d)/(e). No multi‑channel architecture, no rotating power pulling. No anticipation.

(6) US 2012/0212065 A1 — George Shu‑Xing Cheng, "Scalable and redundant mini‑inverters"

• Full citation: US 13/397,402, priority 2011‑02‑15, published 2012‑08‑23; granted as US 9,093,902 B2 (2015‑07‑28). (Listed in the '270 citation table.)
• Description (from the '270's incorporation‑by‑reference paragraph and the citation table): Mini‑inverters with double/triple/quadruple redundancy, one‑to‑multiple input channels, PV modules, daisy‑chaining to the grid.
• Claim mapping / § 102: Relevant § 103 art for claim 1(c–g, k, l) and claim 10(a–c) — multi‑channel/mini‑inverter hardware, boost conversion, isolation switching. Adds nothing on the UPS‑for‑internal‑electronics or lunar‑mode features. No anticipation.

(7) US 2012/0313443 A1 — CyboEnergy, Inc., "Smart and scalable off‑grid mini‑inverters" — ⚠️ most relevant to claim 14

• Full citation: US 13/493,622, priority 2011‑06‑10, published 2012‑12‑13; granted as US 8,994,218 B2 (2015‑03‑31). (Listed in the '270 citation table and incorporated by reference by the '270 at ¶[0004]/¶[0056].)
• Description: Off‑grid mini‑inverters that supply AC loads (motors, pumps, fans, lights, appliances, homes) and support the "AC Master" behavior — energizing a dead AC powerline and permitting other inverters on the line to synchronize.
• Claim mapping / § 102: This is the closest prior art for claim 14 (multi‑channel off‑grid AC Master inverter): it discloses the off‑grid mission, load‑handling, and AC‑master/synchronization behavior. But claim 14 additionally requires the load‑detector‑by‑impedance element and the UPS + three‑mode elements, and the '443 does not disclose the lunar mode or the UPS‑powered internal‑electronics architecture. No anticipation; strong § 103 base for claim 14 (combine with a UPS/battery reference).

(8) US 2013/0002031 A1 — CyboEnergy, Inc., "Enclosure and Message System of Smart and Scalable Power Inverters"

• Full citation: US 13/537,206, priority 2011‑06‑30, published 2013‑01‑03; granted as US 9,331,488 B2 (2016‑05‑03).
• Description: Enclosure and LED messaging/status system for multi‑channel mini‑inverters.
• Claim mapping / § 102: Discloses no element of any '270 claim beyond generic inverter housing/status hardware. No anticipation and negligible § 103 relevance.

(9) US 2014/0252862 A1 — CyboEnergy, Inc., "Maximizing Power Production at Low Sunlight by Solar Power Mini‑Inverters" — ⚠️ most relevant to claim 6 (subject to common‑ownership)

• Full citation: US 13/789,637, filed 2013‑03‑07, published 2014‑09‑11; granted as US 9,331,489 B2 (2016‑05‑03). (Incorporated by reference by the '270 at ¶[0006].)
• Description: A multi‑channel mini‑inverter that, in a low‑power mode under low sunlight, takes power from one solar module to supply DC power to its internal electronic circuits while inverting the remaining modules' DC to AC — i.e., exactly the low‑power‑mode feature the '270 builds on.
• Claim mapping / § 102: Substantively this is the closest reference to claim 6 (method of powering the inverter's internal electronics with mode‑based supply selection) and to the claim 10(e)/claim 1(i) low‑power‑mode limitation. However, it is a § 102(a)(2) reference only, and it is commonly owned by CyboEnergy with the '270 — so AIA § 102(b)(2)(C) likely disqualifies it as prior art. Flag this as a contradiction‑risk item rather than a threat: on the face of the citation table it looks like anticipatory art for claim 6, but the common‑ownership exception probably removes it. It remains highly relevant as context and as an obviousness‑type "state of the applicant's own art" reference.

(10) US 2014/0265584 A1 — CyboEnergy, Inc., "Online Surveillance System to Protect Solar Power Plants"

• Full citation: US 13/844,484, filed 2013‑03‑15, published 2014‑09‑18 (as US 2014/0265584 A1 per the citation table); granted as US 9,240,112 B2 (2016‑01‑19).
• Description: Real‑time day/night monitoring of solar inverters via the AC powerline, generating surveillance alarms if an inverter is disconnected — i.e., powerline‑communication/monitoring functionality.
• Claim mapping / § 102: Bears only on claim 1(i)/(j) (powerline communications, monitoring status). Same commonly-owned / § 102(b)(2)(C) problem as item (9). No anticipation.

(F) KR 10‑1097260 B1 — Samsung SDI, "Grid‑connected energy storage system and method for controlling grid‑connected energy storage system"

• Full citation: priority 2009‑12‑15, published 2011‑12‑22. Listed in the "Family Cites Families" row of the '270 page.
• Description: Grid‑connected ESS with battery and mode control (grid‑tied vs. discharge/backup).
• Claim mapping / § 102: § 103 support for the battery/UPS‑plus‑mode‑selection concept (claims 1(h)/(i), 6, 10(d)). No anticipation.

(NPL) PCT International Search Report dated Aug 20, 2014 in PCT/US2014/030888

• This is the search report for the '270's own PCT application (WO 2014/153301 A1). It is not prior art; it is only evidence of the art the PCT examiner considered.

5. Bottom line — which references matter, and against which claims

No listed reference anticipates any claim of US 9,172,270 under § 102. The two structural pillars of the claims are (i) the UPS + DC power supply co‑powering the inverter's internal electronics, and (ii) the normal/low/lunar three‑mode operation with rotating power pulling — and none of the ten references discloses both pillars in a single disclosure, and most disclose neither.

Ranked by relevance/risk:

1. US 2011/0012430 A1 → US 8,786,133 (item 4) — highest risk, § 103. Discloses the complete multi‑channel inverter hardware of claims 1(a–g, j–l) and 10(a–c). Combine with any battery/UPS reference (JP '383, Lin '259, or Samsung '195) and the apparatus claims' architecture becomes a § 103 question, with novelty resting on the lunar‑mode/rotating‑power‑pulling limitations (claims 10(e), 13, 16–19).
2. US 2014/0252862 A1 → US 9,331,489 (item 9) — highest § 102 exposure on paper, but likely excepted. It targets claim 6 and the low‑power‑mode element most directly; but it is a commonly owned § 102(a)(2) reference and probably disqualified by § 102(b)(2)(C). Flag as a contradiction to resolve against the Patent Center/assignment records.
3. US 2012/0313443 A1 → US 8,994,218 (item 7) — best § 103 base against the off‑grid claim 14.
4. JP 2001‑008383 A (item 1) and US 2009/0174259 A1 (item 2) — the two best independent, third‑party (§ 102(a)(1)) references for the battery‑backed/UPS PV‑inverter concept behind claims 1(h), 6, 10(d) and 12.
5. US 2012/0212065 A1 (item 6) and Samsung SDI '195 / KR '260 (items 5, F) — secondary § 103 art for multi‑channel mini‑inverter hardware and ESS mode control, respectively.
6. First Solar '045 (item 3) and CyboEnergy enclosure '031 (item 8) — negligible.
7. PCT ISR (NPL) — not prior art.

Per-claim "most likely citation": Claim 6 → US 2014/0252862 (commonly-owned, likely excepted) and, if excepted, JP 2001‑008383/Lin '259 as § 103; Claims 1, 4, 10 → US 2011/0012430 in view of a UPS reference; Claim 14 → US 2012/0313443 in view of a UPS reference; Claims 16–19 (rotating power pulling) → no close art in the listed citations — the half‑cycle/full‑cycle window and channel‑polling routine appear unaddressed by all ten references.

6. Uncertainty disclosures

• Tool‑cap limitation: I verified ~half the references on live pages and hit my step limit before re‑opening items 5, 6, 7, 8, 9, 10, F. Descriptions for those are drawn from the authoritative '270 text (incorporation‑by‑reference paragraphs and the printed citation table) and title/assignee/date metadata, not from a fresh read of each specification.
• Correction I am making to the earlier "Litigation summary": that section listed the commonly‑asserted CyboEnergy patents as "'133, '488, '489" and treated them solely as siblings. This analysis confirms the mapping: '133 = US 8,786,133 = US 2011/0012430 A1 (item 4); '488 = US 9,331,488 = US 2013/0002031 A1 (item 8); '489 = US 9,331,489 = US 2014/0252862 A1 (item 9). Two of the three patents CyboEnergy is asserting in court are therefore also cited references on the '270 face — an important strategic fact, though the '270 itself remains (per the earlier finding) unasserted.
• § 102(b)(2)(C) call: My conclusion that items 9 and 10 are likely excepted as prior art rests on the shared CyboEnergy/Cheng ownership shown in the family data. I did not pull the current assignment chain or any terminal disclaimer to confirm perfect common ownership at the relevant date. Confirm before relying on it.
• Statutory‑category caution: The '270 is an AIA‑era filing; I applied AIA § 102(a)(1)/(a)(2). If any pre‑AIA priority were later established, the § 102(e) analysis would change and items 9/10 would need re‑evaluation.
• No similar‑number substitution: consistent with your strict‑ID rule, none of the above results belong to a different "'227"/"270"/"'133" patent unless explicitly identified as such.

I'll verify the load-bearing facts about the primary references before building the § 103 combinations.

Obviousness Analysis — US 9,172,270 B2 under 35 U.S.C. § 103

Patent: US 9,172,270 B2, "Smart and scalable lunar power inverters"
Effective filing date: 2013-03-18 (AIA § 102/§ 103 applies)
Claims analyzed: 1–19 (independents: 1, 4, 6, 10, 14, 16)
Art base: the ten references of record from the Prior Art section, plus the applicant-admitted prior art in the '270 specification itself
Sources for the key references: US 8,786,133 · US 2014/0252862 A1 / US 9,331,489 · US 2014/0252862 (FPO) · JP 2001-008383 A · US 2009/0174259 A1 · US 2011/0012430 A1

---

1. Legal framework and framing decisions

1.1 Standard

• Governing law: AIA § 103. The '270 claims no pre-2013 priority, so its 2013-03-18 filing date falls after the 2013-03-16 AIA cutoff. The Graham factors apply: scope/content of the prior art, differences between art and claims, PHOSITA level, and secondary considerations.
• Test: whether the subject matter as a whole would have been obvious to a PHOSITA at the time of filing. Under KSR Int'l v. Teleflex, a combination is obvious where the elements are known, the combination is of a type "obvious to try," there is a "design need or market pressure to solve a problem," or there is a "finite number of identified, predictable solutions." A rigid teaching–suggestion–motivation test is not required, though an articulated reasoning with rational underpinning remains necessary.
• Burden: preponderance of the evidence in an IPR; clear and convincing evidence at district court.

1.2 One exclusion that materially reshapes the grounds

Two references on the '270 face are the applicant's own co-pending filings and are § 102(a)(2)-only art:

• US 2014/0252862 A1 → US 9,331,489 (filed 2013-03-07, i.e., 11 days before the '270)
• US 2014/0265584 A1 (filed 2013-03-15, i.e., 3 days before the '270)

Both were owned by CyboEnergy, Inc. at the relevant time and are incorporated by reference into the '270. Under § 102(b)(2)(C) (which by its terms applies to § 103 as well), commonly owned § 102(a)(2) subject matter is disqualified as prior art. I therefore treat '489 and '584 as presumptively unavailable in a § 103 rejection — but expressly available as (a) evidence of the state of the art, and (b) a source of the applicant's own admissions where the same matter appears in the '270's own Background.

Important survival point: the content of the '489 disclosure is not needed to establish the problem or the two known options, because the '270 and '489 specifications contain verbatim parallel Background passages. Those passages are applicant admissions about the prior art ("AAPA") and remain fully usable regardless of the common-ownership exclusion. ('489 Background, as filed) This is the single most important framing move in the analysis below.

1.3 Proposed PHOSITA

A person having ordinary skill in the art at 2013-03-18: a B.S. in electrical engineering (or equivalent) with 2–4 years of experience in power electronics for PV/grid-tie applications (or an M.S. with 1–2 years). Such a person would be familiar with: boost-converter topologies and their input capacitors; MPPT algorithms; grid-synchronization/zero-crossing detection; UL 1741 and IEEE 1547 (including the five-minute reconnect requirement); microcontroller-based inverter control; powerline communications; and — critically for this patent — conventional battery/UPS backup of control electronics, a technique that long predates 2013.

Note that the PHOSITA is a power electronics engineer, not a battery specialist. This matters: the UPS-backup-of-control-rails element is a routine application of an established art to a familiar circuit.

---

2. Element-level mapping of the claimed novelty

Stripping the '270 claims to their substantive limitations, four elements carry the weight:

Key structural observation: the '489 claim 1 — the sibling closest to '270 — recites the same multi-channel architecture and the same "normal or low power mode based on calculated DC input power," but reaches the low-power mode without a UPS (it dedicates one input channel and its solar panel to the internal DC supply). The '270's delta over its own sibling is therefore essentially element B alone — the substitution of a rechargeable-battery UPS for the dedicated-channel solar feed. Element B is precisely the element that is thoroughly disclosed in the independent third-party art. That is the core of the obviousness case.

---

3. The motivation to combine — and why it is unusually strong here

Obviousness lives or dies on motivation. This record supplies five independent, mutually reinforcing rationales.

(i) Applicant-admitted problem and applicant-admitted solution space (strongest)

The '270 specification (Background) expressly states that a grid-interactive inverter "can be designed to include a DC power supply unit that can take the AC power from the grid, invert it to DC power, and supply the appropriate DC voltages," and then identifies three shortcomings: added complexity/cost, consumption even in sleep mode, and inoperability off-grid. It then states that the alternative — "take the DC power from the connected solar module(s)" — "has a major limitation," namely that a voltage collapse below ~18 V starves the 15 V rail and forces shutdown.

This is a textbook KSR "design need" and "finite number of identified, predictable solutions" scenario written by the applicant itself. The patent frames the artisan's choice as a two-option problem, critiques both options, and then claims the third option. A PHOSITA reading this problem statement is directed toward a third source of control power. A rechargeable-battery UPS is the canonical third source.

(ii) The third option is old, conventional, and in the record

Each of the following makes element B explicit:

• JP 2001-008383 A (Nissin Electric): PV array → first inverter; rectifier + storage battery (20) → second inverter; a switch (23) reconfigures the circuit. Control circuit (30) operates in two modes — normal operation (inverter 17 at MPPT of the solar cell; inverter 22 in CVCF control) and "shortage operation" (when CVCF supply is insufficient, the solar cell + rectifier + battery become a common DC source driving both inverters). This teaches both the battery-backed architecture and mode-based reconfiguration on PV insufficiency. It does not teach away — it teaches toward switching to stored energy when PV is inadequate.
• US 2009/0174259 A1 (Lin): an inverter with MPPT, a UPS control circuit, and a deep-cycle battery charged via a charging-control circuit from the renewable source, paralleled with the utility. This supplies claim 6(d)/claim 12's "charging the rechargeable batteries of the UPS during the normal power mode."
• US 2011/0148195 A1 (Samsung SDI) and KR 10-1097260 B1: grid-connected energy-storage systems with a battery, a power-conversion system, and a controller transitioning between grid-connected and discharge/backup operation.

(iii) Same field, same problem, overlapping disclosures

All of the above are in PV power conversion / energy storage. The combination is "familiar elements according to known methods… yielding no more than predictable results" (KSR). No change in principle of operation is required: the claim requires only that a battery-backed UPS be available to feed the internal DC supply, with switching logic. Nothing about that is technically difficult in 2013.

(iv) The applicant was already attacking the same problem in the same architecture

The '489 (same inventors, same architecture) discloses exactly the low-power mode and the channel-dedication logic for feeding internal electronics — and its own flow chart (Blocks 318–322) sets a "System Power State" to "Low Power," selects the channel with the lowest input voltage, and puts it in "DC Supply" mode. A PHOSITA designing the next iteration — asked to survive further collapse of the input voltage — would naturally ask "what if even the dedicated channel collapses?" The answer writ large in the art is: add storage. This is the classic "improvement of the applicant's own prior work" motivation and is entirely legitimate, especially where the improvement substitutes a known element from third-party art.

(v) External pressure from the regulatory regime

The '270 and '489 both recite the UL 1741 / IEEE 1547 five-minute mandatory reconnect wait after disconnection. That is a concrete, quantified design incentive to never trip off — i.e., to guarantee control-rail hold-up. KSR recognizes market/regulatory pressure as a valid motivation. The UPS is the accepted solution.

---

4. Ground-by-ground claim analysis

Ground 1 — Claims 1, 2, 3, 9: US 2011/0012430 A1 ('133) in view of JP 2001-008383 A (and/or Lin '259)

Predicted outcome: Claims 1, 2, 9 likely unpatentable under § 103. Claim 3 is high-confidence if MFA appears in the '133 specification; otherwise it needs pending verification.

---

Ground 2 — Claims 4, 5: US 2011/0012430 A1 ('133, single-input embodiment) in view of JP '383 and/or Lin '259

Claim 4 is claim 1's architecture with one input port, one boost converter, and no combiner, limited to normal or low-power mode (no lunar mode). The '133 expressly discloses "smart single-input" inverters (FIGS. 1, 2, 8, 11). The UPS element again comes from JP '383/Lin '259.

Simplification that strengthens this ground: because claim 4 omits the lunar mode, it omits the only limitation that is not squarely in the third-party art. Claim 4 is therefore more vulnerable than claim 1, and its dependent claim 5 (single/three-phase) adds nothing.

Predicted outcome: Claims 4 and 5 likely unpatentable.

---

Ground 3 — Claims 6, 7, 8: JP 2001-008383 A in view of US 2009/0174259 A1 (Lin) [Samsung '195 / KR '260 as alternatives]

Claim 6 is a pure method claim with four steps: (a) enter/exit normal and low-power modes based on measured DC input power; (b) normal mode → DC supply from solar module(s); (c) low-power mode → UPS; (d) charge the UPS batteries during normal mode.

Motivation: both references address the identical problem (PV output insufficiency) with the identical architecture class (PV + inverter + battery). Combining them is the combination of two references from the same field addressing the same problem, with predictable results.

Caveat and honesty flag: the best single reference for claim 6 is the commonly owned '489, which discloses mode selection based on calculated DC input power for the identical architecture (its Block 320/322 logic). If a petitioner can establish that the '489 is not commonly owned as of the effective filing date (e.g., a different obligee), the § 102(b)(2)(C) exception fails and '489 becomes the primary reference for claim 6 — a materially stronger ground. This is the highest-value verification item in the whole analysis.

Predicted outcome: Claim 6 likely unpatentable over JP '383 + Lin '259; stronger if '489 is available.

---

Ground 4 — Claims 10, 11, 12, 13: US 2011/0012430 A1 in view of JP '383/Lin '259

Claim 10 is the most permissive independent apparatus claim in the patent: it requires only (a) ≥2 input channels each with boost converter, measurement circuits, supporting circuits, cables/connectors; (b) an AC output port to grid or load; (c) a DC combiner; (d) UPS + DC power supply for internal electronics; (e) a microcontroller measuring V/I, computing per-channel power, and running normal/low/lunar modes.

Critically, claim 10 omits the line-sensing circuit, the powerline modem, the solid-state isolation switch, and the load interface filter. Each omission removes an element that had to be sourced. Claim 10 should be treated as the broadest and most exposed independent claim.

---

Ground 5 — Claims 14, 15: US 2012/0313443 A1 ('443 → US 8,994,218, off-grid AC Master mini-inverters) in view of JP '383/Lin '259 (and '489 for mode logic)

Claim 14 is the off-grid "AC Master" variant. Prior Art item 7 ('443) is the closest art for:

• the off-grid mission (powering AC loads rather than the grid);
• the AC-Master behavior (initially energizing a dead AC powerline; continually delivering AC power so other inverters on the line can synchronize); and
• load handling/specification checking.

The '270 specification itself states that the m-channel off-grid inverter "can be designed as an AC Master off-grid inverter or a regular off-grid inverter as disclosed in U.S. patent application Ser. No. 13/493,622" — an admission that the off-grid/AC-Master functionality is the '443's contribution, with the '270 adding the UPS and lunar mode on top.

Remaining elements: (i) the load detector detecting load impedance — verify how thoroughly '443 describes impedance-based load detection; and (ii) the UPS/two/three-mode selection — from JP '383/Lin '259/'489.

Drafting defect worth noting: claim 14's microcontroller paragraph contains an internally redundant double recitation — "running the power inverter in normal or low power mode based on calculated DC input power, running the power inverter in a normal mode, a low power mode, or a lunar power mode based on calculated DC input power." Under Nautilus/Williamson, this is a § 112(b) indefiniteness candidate. It is not a § 103 argument (IPR § 112 review is limited to indefiniteness against a proposed construction), but it is a serious parallel vulnerability and it weakens any infringement case built on claim 14.

Predicted outcome: Claims 14–15 likely unpatentable over '443 + JP '383/Lin '259.

---

Ground 6 — Claims 16, 17, 18, 19: rotating power pulling — no supporting art in the record

Claim 16: (a) check energy availability for each input channel; (b) define a power-pulling time window based on AC signal frequency; (c) pull power only from channels with sufficient stored energy during that window.
Claim 17: window = half-cycle or full-cycle of the grid sinewave.
Claim 18: ~10 or 20 ms for 50 Hz; ~8.33 or 16.66 ms for 60 Hz.
Claim 19: claim 16 + claim 10's architecture.

I cannot make an honest § 103 case on this record. None of the ten references of record — not Nissin, not Lin, not First Solar, not the '133, not the '443, not the '488, not the '489, not the '584, not Samsung/KR — discloses or suggests per-half-cycle polling of channel storage-capacitor charge with channel deselection. The '489 gets closest (it checks channel voltage and dedicates the lowest-voltage channel), but it selects one channel to feed the supply, rather than rotating power extraction among charged channels within the AC window. Those are different operations solving different problems.

Two partial arguments exist, and I want to be precise about their strength:

1. Claim 18 is nearly free. Once the concept of the claim 16 window is assumed, claim 18's numbers are arithmetic derived from 50/60 Hz (1/(2·50)=10 ms; 1/(50)=20 ms; 1/(2·60)=8.33 ms; 1/(60)=16.66 ms). Claim 18 adds no technical content beyond claim 16. Similarly, claim 17's disjunctive "half cycle or the full cycle" is a two-option alternative that the specification itself says produces different output uniformity. If claim 16 falls, 17–19 fall with it. But that does not help unless claim 16 falls.

2. A generic obviousness theory for claim 16 would rest on analogous prior art in (a) interleaved/rotating multiphase DC-DC converters (round-robin phase enabling, ripple cancellation), (b) sequential/gated channel activation in multi-input power supplies, (c) load-sharing and current-sharing controllers that skip under-capable sources, and (d) burst-mode / skip-mode switching in low-power converters. The specification's own framing — that the problem is "energy starving" and that the solution is to pull only from sufficiently charged channels to avoid harmonic distortion (UL 1741 limits) — makes the rotational selection an intuitive, problem-driven design choice. But I have not verified any specific reference in these categories, and I will not name one I cannot support. This ground requires a new prior-art search.

Recommended search targets (categories, not citations): interleaved multiphase boost converters with phase shedding/phase rotation; multi-input DC-DC converters with round-robin input sampling; MPPT systems that sequence or skip inputs based on available power; inverter control schemes keyed to the AC zero-crossing for per-half-cycle power draw; "input capacitor charge threshold" gating of converter enable signals.

---

5. Summary ground chart

---

6. Anticipated patent-owner rebuttals and their likely weight

---

7. Refinements and contradictions relative to earlier sections

1. Refinement to the Prior Art section, item 9 ('489): that section correctly identified the '489 as the closest reference to claim 6. Confirmed on re-verification, and sharpened: the '489 claim 1 contains no UPS at all. Its low-power mode dedicates one input channel and its solar panel to the DC supply ("DC Supply" mode, Blocks 320/322). Therefore the '270's novelty over its own sibling reduces almost entirely to the UPS substitution — the element most thoroughly covered by independent third-party art (JP '383, Lin '259, Samsung '195, KR '260). This strengthens the Ground 3 analysis. (US 2014/0252862 A1 · FPO text)
2. Confirmation of the common-ownership problem: the '489 Background contains the same "shortcomings of the AC-derived supply / limitation of the solar-derived supply" passages that appear in the '270. This confirms the applicant-admitted-prior-art argument is available even if '489 is excluded as prior art under § 102(b)(2)(C). No contradiction with the earlier section — but the earlier section's framing of '489 as "highest § 102 exposure on paper, likely excepted" should be adjusted: it is not the best prior art, but it is the best admission.
3. No contradiction found with the Litigation, PTAB, or Assignment sections. The absence of any IPR on the '270 means: no § 325(d) bar, no § 315(e)(2) estoppel, and the art developed in the sibling proceedings (IPR2024-00360/-00362 on the '133 and the PATROLL contests on '488/'489) is unburdened for use against the '270 — subject to the art having to be mapped to '270's own claim language, especially elements B and D.

---

8. Bottom line

US 9,172,270 is a $103-vulnerable patent across most of its claims, with one well-defined shelter.

• Highly vulnerable (§ 103): claims 1, 2, 4, 5, 9, 10, 11, 12, 14, 15, and claim 6–8 on a two-reference ground (JP '383 + Lin '259). The combination is US 2011/0012430 A1 ('133) as the architectural base + JP 2001-008383 A and/or US 2009/0174259 A1 as the UPS/control-power teaching, motivated by the applicant's own admitted defects in both prior alternatives, by the regulatory pressure of the UL 1741 five-minute reconnect penalty, and by the fact that a battery-backed auxiliary supply was a decades-old, field-conventional technique.
• Verify before relying on: claim 3 (MFA limitation — check whether the '133 specification contains the MFA boilerplate; also check Cheng's earlier MFA controller patents); and whether the '489 can be disqualified from (or, conversely, stripped of) the § 102(b)(2)(C) common-ownership exception.
• Not vulnerable on this record: claims 13, 16, 17, 18, 19 — the rotating power pulling family. The listed references do not disclose per-half-cycle, charge-gated channel rotation. Claims 17 and 18 add only frequency arithmetic on top of claim 16 and cannot survive independently, but claim 16 itself requires new analogous art — interleaved multiphase converters, round-robin/phase-shedding channel control, multi-input channel sequencing, or zero-crossing-keyed power extraction. This is the one gap that a § 103 challenge must close, and it is also the patent's only real point of novelty.
• Parallel non-§ 103 attack worth preserving: claim 14's internally inconsistent double recitation of the microcontroller's operating modes is a § 112(b) indefiniteness candidate and a live vulnerability in any enforcement scenario.

Uncertainty disclosures

• Reference verification is incomplete. Live pages were re-opened for US 8,786,133, US 2014/0252862 A1/US 9,331,489, JP 2001-008383 A, US 2009/0174259 A1 and US 2011/0012430 A1 (PCT/WO family) in this pass and in the prior-art pass. US 2011/0148195 A1, US 2012/0212065 A1, US 2012/0313443 A1, US 2013/0002031 A1, US 2014/0265584 A1 and KR 10-1097260 B1 remain described from the citation table and incorporation-by-reference paragraphs, not a fresh read. Elements sourced from those documents are medium-confidence.
• Claim 3 / MFA: I did not confirm the MFA limitation in the '133 specification. I am not naming any specific MFA controller patent number because I have not verified one. If claim 3 must be attacked, that verification is a prerequisite.
• Load detector / claim 14: I did not verify whether '443 (US 8,994,218) discloses impedance-based load detection specifically (as opposed to load presence/size checking). If it does not, claim 14's load-detector element needs separate art.
• Rotating power pulling: I am stating plainly that this record contains no supporting reference. I have not searched the analogous-art categories named in § 4/Ground 6 to the point of identifying a specific citable document, and I will not invent one. Claims 13 and 16–19 require a dedicated search before any § 103 position is taken.
• Common ownership: the § 102(b)(2)(C) call for '489 and '584 rests on the family/assignee data, not on a pulled assignment record at the effective filing date. Confirm on USPTO Assignment Center before relying on the exclusion — and note that the analysis is deliberately built so it does not depend on that exclusion.