§ docs  ·  NVTS  ·  Supply chain
ticker
NVTS
position
long
conviction
2 / 5
analyst
supply-chain-analyst
company
Navitas Semiconductor
generated
2026-05-04

Supply Chain Analysis — Navitas Semiconductor (NVTS)

§ 01Executive View

Navitas's input chain is the most fragile in the deep-dive cohort on a per-revenue basis and the most exposed of any deep-dive name to a single-counterparty event: TSMC's announced exit from GaN foundry by end-July 2027 has forced a multi-foundry transition (PSMC for 200mm in Taiwan, GlobalFoundries for 200mm in Vermont, X-Fab for SiC in Texas) that must be qualified, ramped, and yielded into Rubin Ultra / Kyber 2027 reference-design windows simultaneously with the company's mobile-to-AI revenue pivot. The bull case is that a successful ex-TSMC ramp gives Navitas the cohort's most defensible US-domiciled GaN footprint at exactly the moment the synthesis's "supply chain chose 800V" thesis pulls demand through. The bear case is that the company is mid-cycle on a foundry move, mid-cycle on a process portability transition, and mid-cycle on revenue model — any one of those slipping by 6 months wipes out the 2027 inflection. Conviction is 2/5 on the supply-chain dimension specifically: structural fragility dominates structural opportunity until at least the 100V family ramps at PSMC in 1H26 and a credible GF Burlington tape-out is demonstrated.

§ 02Chain Map (Epi → Customer) — NVTS Exposure at Each Node

Node Layer NVTS Exposure Source
Epitaxy (GaN-on-Si) MOCVD reactor — Aixtron / Veeco Indirect; the foundry (TSMC, PSMC, GF) chooses the epi tool. PSMC Fab 8B and GF Burlington are both 200mm CMOS-style fabs that will use Aixtron G5+/G10 or Veeco Propel reactors for GaN epi growth on silicon substrates. [corpus silent on which specific reactor — verify in PSMC/GF capex disclosures] Synthesis L8d; cohort corpus
GaN substrate 200mm silicon wafers (Si substrate, not freestanding GaN) NVTS's GaN is GaN-on-Si, not GaN-on-SiC or GaN-on-GaN — silicon substrates are commodity (Shin-Etsu, SUMCO, GlobalWafers); not a chokepoint. The epi layer is what matters Navitas-PSMC press release; Navitas 2025 10-K
SiC substrate (GeneSiC) Boules, wafers — Wolfspeed, Coherent (II-VI), SK Siltron CSS, ROHM, Chinese (TanKeBlue, SICC) [corpus silent — 10-K does not name SiC substrate suppliers; X-Fab Texas is the device fab but the boule/substrate vendor upstream of X-Fab is not disclosed] GeneSiC traditionally sourced 150mm wafers via merchant market; 200mm transition unannounced NVTS 10-K (Feb 2026) — substrate vendor unnamed; verify in 10-K Item 1
Wafer fab — GaN TSMC (650V, exiting Jul 2027); PSMC Fab 8B Zhunan, Taiwan (200mm, 180nm CMOS, 100V Q1H26 / 650V transitioning over 12-24mo); GlobalFoundries Burlington VT (200mm, dev early 2026 / production late 2026) The single most concentrated risk in the entire memo. ~50% of TSMC's GaN output was Navitas; transition required by 2027. Triple-foundry path is the mitigation, but adds qualification cost, yield variance, and process portability risk TrendForce Nov 2025; Digitimes Aug 2025; GF Nov 2025 PR; NVTS 10-K
Wafer fab — SiC X-Fab Lubbock TX on 150mm wafers (GeneSiC products) Single-source for SiC device fab. X-Fab has competing customer demand from auto OBC and industrial SiC; Navitas has no second SiC foundry named in 10-K NVTS 2024 10-K; cohort customer.md prior research
Die test / wafer probe Taiwan-based subcontractors (unnamed in 10-K) "Test subcontractors, also primarily in Asia" — not named, but Taiwan tail-aligned NVTS 2025 10-K
Packaging / OSAT Amkor (announced 2017 as packaging partner); QFN, dual-side-cooled, GaNSafe / GaNFast IC packaging in Asia The IC-style GaN integration (driver + FET + protection in one package) creates packaging-process-IP that is partially co-developed with Amkor. Geographic detail not disclosed; Amkor's GaN-relevant sites are likely Philippines / Korea / Taiwan rather than Vietnam (Vietnam is logic ATP-focused) Navitas-Amkor 2017 PR; NVTS 10-K
Module / discrete Direct sales (GaNSafe, GaNFast IC modules; GeneSiC SiCPAK 1200V modules); customers integrate into PSU / DC-DC / VRM boards NVTS sells both discretes and integrated ICs. SiCPAK 1200V module launched Apr 2025 as a packaged subassembly Navitas product announcements; Q3 2025 earnings
Reference design NVIDIA 800VDC silicon partner (1 of 14 — see below); Navitas demos 10kW DC-DC at 98.5%, 12kW with GeneSiC + GaNSafe + Intelliweave; "800V-to-6V single-stage" demonstrated at GTC 2026 NVIDIA partner status is not exclusivity — 14-vendor silicon list (AOS, ADI, EPC, Infineon, Innoscience, MPS, Navitas, onsemi, Power Integrations, Renesas, Richtek, ROHM, ST, TI). NVTS competes with TI's March 2026 800V-to-6V at 97.6% efficiency directly NVIDIA Dev Blog; cohort customer.md
Box-builder / module partner Vertiv, Eaton, Schneider, Delta named in NVIDIA 800V box-builder ecosystem; Murata named by NVTS in past as ref-design partner [corpus silent on BOM-level NVTS-Vertiv or NVTS-Delta lock-in] — synthesis Section 6 contested claim 14 explicitly flags Vertiv/Schneider as the named box-builder set, but no main-text BOM-level confirmation that Navitas silicon is inside a Vertiv 800V PSU. NVIDIA reference plumbing is the closest credible "lock-in" but is participation, not exclusivity NVIDIA Dev Blog; cohort corpus; cohort customer.md
End customer Hyperscalers (via Vertiv/Delta/Schneider PSU + via NVIDIA Kyber rack); AI server ODMs (SuperMicro DCBBS path); EV OEMs via GeneSiC Customer concentration disclosed in 10-K: "limited number of customers account for a significant portion of revenue"; two distributors >10% in FY24; FY25 mobile <25% of Q4 revenue (pivot in execution) NVTS 10-K; cohort customer.md

§ 03Input Map

Tier 1

Input Supplier(s) Concentration Geography Substitution Notes
GaN-on-Si wafer fab (650V) TSMC (until July 2027); transitioning to PSMC Fab 8B + GlobalFoundries Burlington Sole until 2024; tri-source by end-2027 (target) Taiwan (TSMC, PSMC) + Vermont (GF) Active transition — 100V at PSMC 1H26; 650V at PSMC 12-24mo; GF dev early 2026 / production late 2026 The most acute single-supplier risk in the cohort. Navitas was ~50% of TSMC's GaN output; TSMC ending GaN by 2027 forces simultaneous process portability + foundry qualification
GaN-on-Si wafer fab (100V) PSMC Fab 8B (lead) Single-source (initial ramp) Taiwan TSMC capable but exiting; GF on roadmap 100V is critical for the GPU board (Point-of-Load); first volume at PSMC 1H26
SiC device fab (GeneSiC) X-FAB Lubbock TX Single-source US (Texas) None named 150mm wafers per prior 10-K disclosures; 200mm transition not announced
SiC substrate / wafers [corpus silent — 10-K refers generally to "silicon substrates from multiple qualified suppliers"] Industry default for non-IDM SiC: Wolfspeed, Coherent (II-VI), SK Siltron CSS, ROHM, TanKeBlue, SICC [corpus silent — verify in filings] Likely US (Wolfspeed/Coherent) + Korea (SK) + China (TanKeBlue, SICC) blend Multi-source feasible; substrate market in oversupply per WOLF analysis Navitas does not own SiC substrate capacity. Per WOLF/supply-chain.md, the 150mm/200mm SiC substrate market is at oversupply with ASPs falling ~10-20%/yr — net input-cost tailwind to NVTS GeneSiC margins
MOCVD epi reactors Aixtron (DE) primary, Veeco (US) secondary — at the foundry tier Foundry-managed Germany (Aixtron Herzogenrath); US (Veeco) Indirect exposure — NVTS's foundries (TSMC, PSMC, GF) make the choice Aixtron G5+/G10 is the de-facto Western 200mm GaN tool. Same Aixtron lock-in WOLF has on SiC
OSAT / packaging Amkor (announced 2017); other Asian OSATs implied Concentration unclear; Amkor named publicly Asia (Philippines/Korea/Taiwan likely) ASE is the natural second-source but not disclosed GaNSafe / GaNFast IC packaging is partially co-developed; switching costs higher than commodity QFN
Probe / final test "Test subcontractors, primarily in Asia" — unnamed Multi-vendor implied Asia High Standard subcontracting; not a chokepoint
Silicon wafers (substrate for GaN epi) Shin-Etsu, SUMCO, GlobalWafers — managed by foundry Foundry-managed Japan / Taiwan Multi-source Not the binding constraint for GaN-on-Si
Specialty gases (NH3, TMGa, TMAl, hydrogen) Linde, Air Liquide, Taiyo Nippon Sanso; trimethyl-aluminum / -gallium specialty Foundry-managed; gallium-precursor concentrated Mostly Asia Multi-source for bulk; gallium chain China-concentrated Gallium chokepoint: China holds ~80%+ refined gallium; Aug 2023 export controls active. Indirect exposure since foundry buys precursors, but a gallium ban would price-shock the entire GaN industry
Wet chemistry, photoresist, etch chemistry Standard (JSR, TOK, Sumitomo, Merck) Foundry-managed Mostly Japan Multi-source Not a Navitas-specific risk

Tier 2 chokepoints

The hidden dependencies — and where NVTS's input-side picture is genuinely worse than the tier-1 table suggests:

  1. Process portability between TSMC, PSMC and GF. Navitas has spent ~10 years tuning device performance on TSMC's specific 650V GaN process at TSMC's specific 200mm fab. Moving the same product to PSMC's 200mm 180nm CMOS process and to GF Burlington's 200mm GaN line (licensed from TSMC, not natively developed) requires PDK re-qualification, parametric re-binning, ESD/reliability re-qualification (HTRB, HTOL, H3TRB — typically 1,000h+ per process), and customer re-qualification at every design-in. The corpus is silent on whether GaNSafe ICs (the integrated driver+FET+protection family) port cleanly to PSMC's 180nm CMOS or whether the integrated function requires substantial design rework. This is the single most underpriced risk in the chain.
  2. Aixtron G10-SiC / Planetary GaN reactor backlog. Same Aixtron concentration the WOLF memo flags for SiC applies in mirror to GaN epi tools. Every Western GaN ramp (Infineon Villach 300mm, TI Aizu/Dallas, Innoscience Suzhou 8", Navitas-PSMC, Navitas-GF) is queueing on Aixtron capacity. PSMC and GF each need to install / re-tool reactors during 2025-2026 to support Navitas's volume — PSMC Fab 8B's GaN tool installation cadence is the gating step Navitas does not control.
  3. GaNSafe IC packaging IP / Amkor relationship. The integrated-IC GaN architecture (the "density" thesis) needs a packaging partner that has worked through dual-side cooling, isolation, and high-frequency-switching board layout co-design. Amkor has been Navitas's named partner since 2017. A loss of access to Amkor's specific packaging line (industrial fire, capacity reallocation, M&A) is non-trivial and is not addressed by adding GF or PSMC at the wafer-fab tier. Also, since GaN ICs switch at 1 MHz+, parasitic-inductance-aware packaging matters more than for silicon power semis — switching cost to a different OSAT is real.
  4. Renesas-GaN counterparty exposure (indirect). Renesas has its own GaN LLC DCX program (synthesis L8b) and has a strategic interest in domestic Japanese GaN. If Renesas leans into the cohort competitively, the question of Japanese OSAT capacity at Amkor/ASE becomes contested.
  5. Gallium itself (tier-3 chokepoint). Gallium concentration in China is ~80%+ refined; the Aug 2023 export-control regime is precedent. NVTS's foundries are price-takers for trimethyl-gallium precursor — a counter-sanction event would compress GaN industry margins broadly. Synthesis Section 5 calls this out as a NVTS/TXN/Infineon shared headwind. Cohort note: "Supra" (Crucible private portfolio) is the user's onshore-gallium-recovery thesis explicitly tracked as the response.
  6. GeneSiC SiC substrate vendor unidentified in public filings. GeneSiC is the only deep-dive name where a meaningful product line's substrate-tier supplier is genuinely unknown to me from public sources. WOLF's substrate share has fallen from 60% to 34% (synthesis), so default exposure is to a Wolfspeed-anchored merchant chain; if NVTS sources from Coherent or SK Siltron CSS instead, the merchant resilience picture shifts. [corpus silent — flag for follow-up: 10-K Item 1 "Manufacturing" subsection, plus 10-Q raw-materials disclosures, are the next disclosure to read in full PDF form.]
  7. GF Burlington as a second-source is structurally less proven. GlobalFoundries has licensed TSMC's 80V/650V GaN process (per Electronic Design / GF press release Nov 2025). A licensed process at a different fab is not the same as a tape-out-validated process. GF's track record on power-semi process portability is short (Burlington was historically RF/SOI for IBM). Production "later 2026" is aggressive; slip risk is elevated.

§ 04Risk Scoring

Risk vector Score (1-5; 5 = severe) Why
Single-source exposure 5 TSMC is the historical sole-source GaN foundry, exiting July 2027. PSMC and GF are unproven replacements still in qualification. X-Fab is sole SiC fab. Worst single-source profile in the cohort
Geographic concentration of inputs 4 Tier-1 wafer fab is Taiwan (TSMC + PSMC) + Vermont (GF). OSAT in Asia. The mitigation (GF Burlington) is real but late and small in 2026-27
Geopolitical exposure 4 PSMC + Amkor + test all sit inside the Taiwan tail per synthesis Open Question #3. GF Burlington is the only US-domiciled GaN escape valve. China gallium counter-sanction risk is industry-wide tailwind for non-China names but a price-shock event regardless
Capacity tightness 3 PSMC Fab 8B has stated capacity for the 100V family in 1H26; 650V transition is 12-24 months — manageable if no slips. Aixtron tool backlog applies, but NVTS's volumes are smaller than IFX's so allocation pressure is lower per-fab. Capacity is sufficient if and only if execution holds
Inventory cushion 3 Q4 2025 earnings call disclosed "stockpiling buffer inventory" against the TSMC exit. Specific days-of-supply not disclosed. Cash position $236.9M (YE'25) leaves room for inventory build, and FY25 capex was only $1.5M (fab-lite). [Verify: Item 7 MD&A inventory days disclosure]
Pass-through power 4 (= weak, high risk) Trapped between (a) foundry pricing power increasing during transition (PSMC and GF can charge re-qualification economics), and (b) hyperscaler customer demanding 800V $/W cost-down via NVIDIA reference designs. Q3 2025 non-GAAP GM 38.7% (vs Q3 2024 40.1%); $3.2M China SiC inventory reserve in 9M 2025 — early evidence of margin compression. Compare TSMC 58% GM, IFX 43%, MPWR 55%, ON 38% — NVTS is at the low end of the cohort
Counterparty concentration on the foundry side 5 Adding to single-source: TSMC explicitly exiting the relationship is unique among the deep-dive cohort. INTC, TSM, NVDA, AVGO all have increasing foundry capacity commitments; NVTS is the only deep-dive name losing one

Synthesis. The pattern: every classical supply-chain vector except capacity is at 4-5. The mitigation (PSMC + GF + buffer inventory + $230M cash) is real but sequenced wrong relative to the demand-pull window. The synthesis frames 2027 (Kyber rack, Rubin Ultra) as the inflection — the same year TSMC GaN goes to zero. The tightrope is the GF Burlington production ramp closing exactly as TSMC closes, and PSMC's 650V transition completing 12-24 months from July 2025 (i.e., July 2026 - July 2027). There is no slack in this calendar.

§ 05Pass-Through Power

Navitas's pass-through is structurally weak and currently deteriorating. Three pieces of evidence:

  1. Non-GAAP gross margin trend, 2024 → 2025: Q3 2024 40.1% → Q3 2025 38.7%; full-year FY24 34.0% → FY25 31.0% (per StockAnalysis aggregation). The trajectory is the wrong direction during a foundry transition.
  2. No long-term wafer-supply agreement disclosed with TSMC, PSMC, or GF. Compare to Wolfspeed-Renesas $2B 10-year LTSA, or Apple's per-node prepay for TSMC. NVTS does not have a contractual cost-pass-through mechanism in either direction.
  3. NVIDIA "ecosystem partner" status is participation, not exclusivity. The 14-vendor 800V silicon partner list (cohort competitor.md id=38) means NVIDIA can substitute Innoscience, EPC, Infineon, Power Integrations, or TI for any specific socket if NVTS's pricing is uncompetitive. TI's March 2026 800V-to-6V at 97.6% efficiency / >2000W/in³ is the live test — TI's vertical-integration cost structure plus 4× internal GaN expansion (Aizu + Dallas) directly attacks Navitas's flagship spec.

The structural reason pass-through fails: NVTS sits between two more-concentrated counterparties (foundry; hyperscaler-backed reference design) and is the price-taker on both sides. This is the inverse of the TSMC pass-through profile — TSMC is the chokepoint, NVTS is the squeezed middle.

That said, two partial offsets: (1) the SiC substrate ASP collapse (synthesis 3.3, WOLF supply-chain.md) is a tailwind to GeneSiC product-line cost, and (2) the gallium / TMG precursor market is broadly stable in the absence of a Chinese counter-sanction. Net: score 4 — weak pass-through, with the GaN axis as the binding constraint.

§ 06Stress Scenarios

Scenario 1: GF Burlington production slip from late-2026 to 2H-2027 or beyond

Probability: mid-high, 40-55% over 24 months. Vermont is a fab that has historically run RF/SOI processes; GaN power devices are a different discipline. The GF press release timing ("development early 2026, production later 2026") is aggressive even on greenfield logic terms. PSMC carrying both 100V and 650V volume during a slip is plausible technically but stresses Taiwan-tail concentration.

Mechanism. TSMC GaN goes to zero July 2027. If GF is not in volume by then and PSMC 650V transition is also delayed, NVTS's 650V revenue collapses for 1-3 quarters until either fab catches up. The 100V/100A AI server P3 socket (the new flagship) requires PSMC qualification on a 180nm CMOS process that did not previously host GaN — process portability is the binding step.

Financial impact. 12-24 month revenue gap of $20-50M in 2027-28 (estimate); margin compression to mid-20s GM during the gap. Cash-burn extends from "Q4 2025 trough then growth" to "another year of losses." With $237M cash and $15M/quarter opex, NVTS can absorb 2 years of $40-50M cash burn — the runway exists, but the equity multiple does not survive a second consecutive year of revenue compression.

Response options. (1) Lean harder on PSMC, accept higher Taiwan concentration. (2) Add a third foundry (rumored X-Fab GaN line in Erfurt — unconfirmed; Tower Semi possible). (3) Sacrifice 650V design wins to preserve 100V Point-of-Load wins. (4) Pricing concession to NVIDIA / Vertiv to retain reference-design slot. None is a thesis-saver alone.

For the long thesis. This is the dominant operational tail risk. Disclosure that would change view: GF Burlington tape-out date with disclosed process-corner data, plus PSMC 650V engineering-sample qualification status by mid-2026.

Scenario 2: Taiwan tail event (PSMC Fab 8B + Amkor disrupted)

Probability: low-mid, 5-10% over 5 years for a meaningfully disruptive event short of full kinetic conflict (per cohort synthesis Open Question #3). Earthquake, blockade, customs harassment are all live vectors; Zhunan Science Park is in Hsinchu County, central Taiwan.

Mechanism. PSMC Fab 8B unavailable for shipment; Amkor's Taiwan-resident GaN packaging line concurrently exposed. GF Burlington becomes the only GaN supply path. NVTS's quarterly revenue is ~$8-12M in this regime; a 6-month PSMC outage compresses 2027 revenue 30-50% even with GF ramping.

Financial impact. Equity is binary in the kinetic case (Taiwan tail closes the entire foundry sub-stack — TSMC, PSMC, UMC, VIS, ASE, etc.; cohort-wide). In the blockade case, NVTS relatively benefits from having GF Burlington, but absolute revenue still compresses. The cohort impact is asymmetric — Innoscience (China-domiciled, captive 200mm) is helped by Taiwan disruption; NVTS is hurt less than pure-Taiwan competitors.

Response options. Accelerate GF Burlington (at maximum still 6-12 months from full qualification on a per-product basis). License a third foundry. None is fast enough to fill a 6-month PSMC outage.

For the long thesis. Within-cohort, NVTS has better Taiwan-mitigation than INTC's foreign customer flows or TSM's own production base. The relative position improves vs. Taiwan-only GaN peers under stress — but absolute equity falls in any scenario where Taiwan-tail risk repriced. Position sizing should respect this.

Scenario 3: Chinese gallium / TMG export counter-sanction

Probability: mid, 15-25% over 3 years. China imposed gallium export controls in Aug 2023 with measured effect; further escalation is plausible if US tightens HBM/DUV/EUV controls further. China holds ~80% of refined gallium globally.

Mechanism. TMG (trimethyl-gallium) precursor prices spike 50-200%; foundry wafer cost rises; entire GaN industry sees input-cost compression. NVTS, as a fabless price-taker, absorbs the foundry surcharge.

Financial impact. GM compression of 200-500 bps (from already-thin 31-39% range). The shock is symmetric across NVTS, Infineon, TI, EPC, Power Integrations. Innoscience is asymmetrically advantaged as a Chinese IDM with assured domestic gallium access — this is the single most material China-vs-West asymmetry in GaN.

Response options. (1) Pass cost to NVIDIA / Vertiv / customer (limited per-pass-through analysis above). (2) Pivot product mix to higher-margin SiC during the shock. (3) Wait it out; export controls historically loosen within 12-18 months. (4) Cohort-side bet: be long Supra / Crucible-onshored gallium.

For the long thesis. This is the kind of event where Navitas's SiC product mix (GeneSiC) is structurally hedge-providing — 30-40% of revenue not exposed to gallium directly. Modestly thesis-supportive over a 3-year horizon if the relative-positioning frame matters more than the absolute earnings hit.

Scenario 4: Innoscience / TI cost curve dominates before Navitas density-spec wins

Probability: high, 50-65% over 24 months. This is the synthesis Open Question #3 (300mm cost collapse) operationalized. Innoscience runs 8" (200mm) GaN-on-Si IDM-style at scale (cohort competitor.md id=42), 12-24 months ahead of NVTS-PSMC on wafer economics. TI has 4× internal GaN capacity expansion at 200mm Aizu + Dallas with 300mm pilot complete. If both companies achieve ~20-30% cost-per-die advantage versus NVTS by 2027, NVTS's "density" reference designs lose at the design-in moment.

Mechanism. This is not a supply-chain disruption — it is a competitive supply-chain event where peers' supply chain matures faster. NVIDIA's 14-vendor partner list lets the customer chose on cost; NVTS's value-add is density and integrated-IC architecture, but if the cost gap exceeds 20-25%, integrated IP cannot recover the BOM.

Financial impact. Slow-burn rather than acute: 2027-28 design-in losses translate to 2028-29 revenue compression of 30-50% relative to bull case. Cash position absorbs this for 2 years before a second equity raise.

Response options. (1) Concede the 100V volume socket to Innoscience/TI; defend the 650V/integrated-IC mid-volt position. (2) Accept gross-margin compression to maintain volume. (3) Consider a strategic combination (X-Fab? GF? Renesas? small IDM combination).

For the long thesis. This is the biggest existential risk and the one the synthesis explicitly flags. Probability is high; severity is moderate. Disclosure that would change view: any concrete NVTS-PSMC die-cost number relative to Innoscience's published 8" cost curve.

§ 07Bull Points

  • GF Burlington is the cohort's only meaningfully US-domiciled GaN fab. Even if it ramps slow, it is a structural CHIPS-Act-aligned asset that Innoscience (China) and Infineon (Germany) cannot match. National-security buyers (DoD, Five Eyes hyperscalers) may pay a premium specifically for US-made GaN — NVTS is the only deep-dive name with a credible 2026-27 path to that.
  • The TSMC exit is structurally permanent — Navitas has been forced to diversify. Most cohort names that look "single-sourced" still are. NVTS will, by end-2027, be the most multi-sourced GaN supplier in the West (TSMC tail until Jul 2027; PSMC; GF; Amkor packaging; X-Fab SiC). The fragility was disclosed; the response is in motion.
  • SiC substrate ASP collapse is a tailwind for GeneSiC unit economics. Per WOLF/supply-chain.md, the 150mm/200mm SiC merchant substrate market is at oversupply with mid-teens ASP declines. NVTS, as a non-IDM SiC consumer, captures this margin. This partly hedges the GaN-side margin compression.
  • Capex-light fabless model survives input shocks better than IDMs. FY25 capex was $1.5M against $237M cash — Navitas can re-route, requalify, and rebuild inventory without bearing fixed-cost stranded-asset risk. Compare to Wolfspeed's $47M/qtr underutilization charges.
  • NVIDIA reference-design status, even without exclusivity, is a credibility gate. Of the 14 silicon partners, the integrated-IC cohort (NVTS, EPC, Infineon, MPS) is smaller than the discrete-FET cohort. Reference-design lock-in has switching costs that pure-discrete competitors lack.

§ 08Bear Points

  • The TSMC exit by July 2027 is the most acute foundry-counterparty event in the cohort. No other deep-dive name has a critical foundry actively exiting their product. Conviction of execution dominates the bull case, and execution is binary on PSMC 650V + GF Burlington tape-out by H1 2027.
  • NVTS sits in the squeezed middle of pass-through. Foundry consolidation gives PSMC and GF pricing power during qualification; NVIDIA's 14-vendor partner list gives the customer pricing power on the way out. Q3 2025 GM compression to 38.7% is the early signal.
  • Amkor packaging exposure is structurally Asian and concentrated — NVTS's IC packaging IP is co-developed with one named partner. A single-facility event affects the entire GaNSafe / GaNFast IC line.
  • Process portability across TSMC → PSMC → GF is unproven for the integrated-IC product family. The 100V FET ramp at PSMC in 1H26 is the test case; the GaNSafe IC family port to PSMC's 180nm CMOS is the harder test. [Verify in 10-K Item 1 — process disclosure detail.]
  • GeneSiC SiC sourcing is a black box from public filings. WOLF/supply-chain.md establishes the merchant SiC substrate market is in oversupply, but NVTS's specific substrate vendor is not named — disclosure quality on this dimension is poor relative to peers.
  • Innoscience's IDM 200mm GaN at scale is a 12-24 month structural cost head-start that Navitas cannot match without committing IDM capex, which the cohort financial.md (NVTS file id=37) explicitly rules out.
  • Single-stage 800V-to-6V is "demonstrated" but the synthesis explicitly flags it as "too early to say whether mainstream" (contested claim #2). TI's 800V-to-6V product launch in March 2026 at competitive specs (97.6% efficiency, >2000W/in³) means NVTS's flagship demonstrator faces head-on vertical-integrated competition before achieving design-in scale.

§ 09Conviction (1-5)

2 / 5 on the supply-chain dimension specifically.

The score reflects: (a) acute single-counterparty risk (TSMC exit) that is the worst input-side picture in the cohort; (b) credible but unproven mitigation (PSMC + GF) running on a calendar with no slack; (c) pass-through power is structurally weak and currently deteriorating (38.7% GM trend); (d) the cohort thesis explicitly identifies NVTS as the density bet against Infineon scale and TI vertical integration — and the supply-chain math on density-vs-cost-curve goes against NVTS as 200mm wafer economics mature at Innoscience and TI. Not a 1 because: capex-light fabless model has structural survivability, GF Burlington is a real US-domiciled differentiator, NVIDIA reference-design status is credible, and SiC substrate tailwind partially hedges GaN compression. Not higher because the foundry transition calendar is tight and the synthesis-level open question (density vs. cost curve) cuts against the structure.

§ 10Key Risks to This Read

  • Assumption: TSMC's July 2027 exit holds and PSMC 650V transition completes by mid-2027. If TSMC extends GaN production by 12-24 months (rumored possibility per Mark Lapedus substack), NVTS gains material breathing room. If PSMC slips, the picture worsens materially.
  • Assumption: GeneSiC substrate sourcing is non-Wolfspeed-exclusive. I am defaulting to a merchant-mix assumption (Wolfspeed + Coherent + SK Siltron CSS) given X-Fab's likely standard procurement practice. If GeneSiC is single-sourced from Wolfspeed and Wolfspeed re-restructures, the SiC tailwind reverses. This is the single piece of disclosure I would most want.
  • Assumption: GF Burlington production starts late 2026 with credible yield. Greenfield GaN at a historically RF/SOI fab is an aggressive timeline. Slip-risk is the dominant downside catalyst.
  • Assumption: pass-through power stays at the current weak baseline. A successful design-in at NVIDIA Kyber as the named partner (not just one of 14) could shift this. A pricing-down event from a Vertiv or Delta cost-out program could worsen it.
  • Assumption: Amkor's GaN-IC packaging capacity tracks NVTS's volume ramp. Not disclosed in filings; verify in 10-K Item 1 manufacturing subsection.
  • Disclosure that would most change the read: (1) NVTS-PSMC 650V engineering-sample qualification timeline, (2) GF Burlington tape-out date, (3) GeneSiC substrate vendor name, (4) NVTS days-of-supply for advanced GaN buffer inventory (referenced in Q4'25 call but not quantified), (5) Amkor capacity allocation for GaNSafe IC family.

§ 11Sources

  • Navitas Semiconductor 10-K filed Feb 27, 2026 (extracted via Stocktitan filings summary). https://www.stocktitan.net/sec-filings/NVTS/10-k-navitas-semiconductor-corp-files-annual-report-4a9e4aab1de4.html
  • Navitas Semiconductor 10-K filing direct (Feb 2026). https://ir.navitassemi.com/static-files/9829a89f-e995-47ea-ab3a-a7eb37ef8558
  • Navitas Semiconductor Q3 2025 Financial Results (Nov 3, 2025). https://www.globenewswire.com/news-release/2025/11/03/3179778/0/en/Navitas-Semiconductor-Announces-Third-Quarter-2025-Financial-Results.html
  • Navitas Announces Plans for 200mm GaN Production with PSMC (Jul 1, 2025). https://navitassemi.com/navitas-announces-plans-for-200mm-gan-production-with-psmc/
  • GlobalFoundries and Navitas Semiconductor Partner to Accelerate U.S. GaN Technology and Manufacturing — GF press release (Nov 20, 2025). https://gf.com/gf-press-release/globalfoundries-and-navitas-semiconductor-partner-to-accelerate-u-s-gan-technology-and-manufacturing-for-ai-datacenters-and-critical-power-applications/
  • TrendForce — GlobalFoundries Moves on GaN: TSMC and Navitas Ties Position U.S. as New GaN Production Hub (Nov 27, 2025). https://www.trendforce.com/news/2025/11/27/news-globalfoundries-moves-on-gan-tsmc-and-navitas-ties-position-u-s-as-new-gan-production-hub/
  • Semiconductor Today — TSMC to cease GaN foundry production by end-July 2027 (Jul 3, 2025). https://www.semiconductor-today.com/news_items/2025/jul/tsmc-030725.shtml
  • Digitimes — TSMC GaN exit prompts Navitas to shift orders to PSMC for Nvidia AI servers (Aug 13, 2025). https://www.digitimes.com/news/a20250813PD218/tsmc-gan-wafer-nvidia-psmc.html
  • Navitas Supports 800 VDC Power Architecture for NVIDIA's Next-Generation AI Factory Computing Platforms — press release. https://navitassemi.com/navitas-supports-800-vdc-power-architecture-for-nvidias-next-generation-ai-factory-computing-platforms/
  • NVIDIA Developer Blog — Building the 800 VDC Ecosystem for Efficient, Scalable AI Factories (silicon-partner list). https://developer.nvidia.com/blog/building-the-800-vdc-ecosystem-for-efficient-scalable-ai-factories/
  • NVIDIA 800 V HVDC Architecture Will Power the Next Generation of AI Factories — NVIDIA Developer blog. https://developer.nvidia.com/blog/nvidia-800-v-hvdc-architecture-will-power-the-next-generation-of-ai-factories/
  • Navitas Announces TSMC & Amkor Manufacturing Partnerships (Oct 17, 2017). https://navitassemi.com/navitas-announces-tsmc-amkor-manufacturing-partnerships/
  • Cohort synthesis — Section 2 (chip-to-grid value chain), Section 3.3 (GaN three-way race), Section 5 headwind on single-stage 800V-to-6V, Section 6 contested claim #2 (Navitas mainstream adoption), Open Question #3 (Taiwan-tail), 2026-05-04 refresh. ../synthesis.md
  • Cohort companies.json — NVTS entry (id 12). ../companies.json
  • Cohort corpus — "The AI Power Crisis Part 2" (Crucible / NuttyCLD). ../corpus/corpus.md
  • Cross-reference TSM/supply-chain.md — foundry capacity frame, Aixtron/ZEISS/Inpria tier-2 chokepoints, Taiwan-tail framework. ../TSM/supply-chain.md
  • Cross-reference WOLF/supply-chain.md — SiC substrate market oversupply, ASP collapse mid-teens, Aixtron lock-in, Renesas-LTSA dissolution case. ../WOLF/supply-chain.md
  • TI — Unveils Complete 800 VDC Power Architecture for AI Data Centers with NVIDIA (Mar 16, 2026). https://www.ti.com/about-ti/newsroom/news-releases/2026/2026-03-16-ti-unveils-complete-800-vdc-power-architecture-for-future-generation-ai-data-centers-with-nvidia.html
  • TI — Quadruples internal GaN manufacturing capacity (Aizu + Dallas) (Oct 24, 2024). https://www.ti.com/about-ti/newsroom/news-releases/2024/2024-10-24-texas-instruments-expands-internal-manufacturing-for-gallium-nitride--gan--semiconductors--quadrupling-capacity.html
  • TrendForce — NVIDIA Picks Innoscience as Sole Chinese Supplier for 800 VDC Power (Aug 4, 2025). https://www.trendforce.com/news/2025/08/04/news-nvidia-picks-innoscience-as-sole-chinese-supplier-for-800-vdc-power-unpacking-the-gan-giant/
  • Navitas — Redefining Data Center Power: GaN and SiC Technologies for Next-Gen 800 VDC Infrastructure white paper (Oct 2025). https://navitassemi.com/wp-content/uploads/2025/10/Redefining-Data-Center-Power-GaN-and-SiC-Technologies-for-Next-Gen-800-VDC-Infrastructure.pdf
  • Electronic Design — GlobalFoundries Partners with Navitas to Ramp Up Next Generation of GaN Power ICs. https://www.electronicdesign.com/technologies/embedded/article/55336941/electronic-design-globalfoundries-partners-with-navitas-to-ramp-up-next-generation-of-gan-power-ics
  • Mark Lapedus substack — 5 Reasons Why TSMC Is Exiting the GaN Market. https://marklapedus.substack.com/p/5-reasons-why-tsmc-is-exiting-the

Works cited

  1. Mark Lapedus substack — 5 Reasons Why TSMC Is Exiting the GaN Market
    industry-report marklapedus.substack.com first cited by · competitor-analyst 2026-05-04
    • Independent analysis of TSMC GaN exit rationale
    • TSMC reallocating GaN capacity to higher-margin AI logic
  2. NVIDIA Developer Blog — Building the 800 VDC Ecosystem for Efficient, Scalable AI Factories (silicon-partner list)
    industry-report developer.nvidia.com first cited by · competitor-analyst 2026-05-04
    • NVIDIA 800 VDC silicon-partner list contains 14 vendors: AOS, ADI, EPC, Infineon, Innoscience, MPS, Navitas, onsemi, Power Integrations, Renesas, Richtek, ROHM, ST, TI
    • Demonstrates that 'NVIDIA 800V partner' status is participation, not exclusivity — undercuts the implicit moat framing in NVTS press releases
    • Lists box-builder partners (ABB, Eaton, GE Vernova, Heron, Hitachi, Mitsubishi, Schneider, Siemens, Vertiv) and component partners (Bizlink, Delta, Flex, Lead Wealth, LITEON, Megmeet)
  3. Power Electronics News — APEC 2025 GaN vs SiC competitive boundary
    industry-report powerelectronicsnews.com first cited by · customer-analyst 2026-05-04
    • 650V is the GaN/SiC overlap zone — both technologies contestable; relevant for NVTS's GaN-IC vs GeneSiC SiC product-line strategy
    • GaN encroaches up from low/mid voltage; SiC retains 1200V+ — places NVTS density bet specifically in mid-voltage AI-DC zone where Power Integrations and Innoscience also compete
    • Cross-referenced from WOLF/sources.json id=19
  4. Semiconductor Today — Yole Power GaN device market 42% CAGR to $3bn by 2030 (Oct 2025)
    industry-report semiconductor-today.com first cited by · market-positioning-analyst 2026-05-04
    • GaN revenue of $920M expected in 2026, up 58% on 2025 — calibrates the 2026 ramp slope
    • Confirms 42% CAGR baseline and reinforces Yole 2025 sizing as the cohort's anchor TAM source
  5. TrendForce 2024 GaN power-device market share data (via 36kr summary, 2025)
    industry-report eu.36kr.com first cited by · market-positioning-analyst 2026-05-04
    • 2024 global GaN power-device share: Innoscience 29.9%, Navitas 16.5%, EPC 12.4%, Infineon 10.3%, Power Integrations 9.8%
    • Top-5 concentration ~85% — basis for HHI calculation (~1,490, moderately concentrated)
    • Top-3 share ~57% — used in market-structure table
    • + 1 more
  6. TrendForce — GlobalFoundries Moves on GaN: TSMC and Navitas Ties Position U.S. as New GaN Production Hub (Nov 27, 2025)
    industry-report trendforce.com first cited by · supply-chain-analyst 2026-05-04
    • Navitas held roughly half of TSMC's GaN wafer output prior to the exit announcement — primary independent corroboration of NVTS-TSMC concentration
    • TrendForce confirms Navitas multi-foundry strategy: TSMC (legacy), PSMC (200mm Taiwan), GlobalFoundries (200mm Vermont)
    • Article does not name X-Fab, suggesting X-Fab is SiC-only (not part of GaN strategy)
  7. TrendForce — NVIDIA Picks Innoscience as Sole Chinese Supplier for 800 VDC Power
    industry-report trendforce.com first cited by · competitor-analyst 2026-05-04
    • Innoscience confirmed as sole Chinese partner in NVIDIA 800V silicon list
    • Innoscience runs 8-inch (200mm) in-house GaN at scale — has 12-24 month wafer-economics head start over Navitas's PSMC ramp
  8. Yole Group — 'Power GaN 2025' / 'From chargers to data centers' press release (Oct 2025)
    industry-report yolegroup.com first cited by · market-positioning-analyst 2026-05-04
    • Power GaN device market $355M (2024) growing to ~$3B (2030) at 42% CAGR — primary external TAM anchor
    • Application split by 2030: consumer/mobile >50%, automotive 73% CAGR, telecom/datacenter 53% CAGR to ~$380M
    • First commercial 800V HVDC rollouts anticipated 2027 — corroborates NVTS's stated 2027 inflection
    • + 1 more
  9. Yole Group — Power SiC overcapacity downturn until 2027–2028; $10bn by 2030 (Dec 2025)
    industry-report semiconductor-today.com first cited by · market-positioning-analyst 2026-05-04
    • Power SiC device market projected to reach $10.3B by 2030 at ~20% CAGR
    • SiC market in correction cycle through 2027–2028 due to upstream overcapacity + automotive softness — directly relevant to NVTS GeneSiC near-term cycle position
    • Data center cumulative SiC opportunity ~$200M over next 5 years
    • + 1 more
  10. Bamboo Works — Innoscience makes gains in patent dispute, as growing competition remains bigger threat
    news thebambooworks.com first cited by · competitor-analyst 2026-05-04
    • Bank of America 2025 GaN device share commentary: Innoscience ~30%, Navitas ~17% (#2)
    • Confirms NVTS is not the share leader — corpus three-way race framing is incomplete without Innoscience as #1
  11. Digitimes — TSMC GaN exit prompts Navitas to shift orders to PSMC for Nvidia AI servers
    news digitimes.com first cited by · competitor-analyst 2026-05-04
    • Navitas-PSMC 200mm partnership; 100V volume H1 2026; 650V transitioning over 12-24 months
    • Direct evidence of supply-chain transition timing risk vs Rubin Ultra 2027 ship
  12. Electronic Design — GlobalFoundries Partners with Navitas to Ramp Up Next Generation of GaN Power ICs
    news electronicdesign.com first cited by · competitor-analyst 2026-05-04
    • GF licensed TSMC's 80V/650V GaN process; development work starts early 2026, production later 2026 in Burlington, VT
    • US-domiciled supply optionality differentiator vs Innoscience and Taiwan-only PSMC
  13. Infineon — Completes acquisition of GaN Systems
    news infineon.com first cited by · competitor-analyst 2026-05-04
    • Infineon acquired GaN Systems for $830M, closed Oct 2023
    • Inherited GaN Systems' IP estate, datacenter design wins, and 600/650V HEMT portfolio
    • Establishes Infineon as the deepest GaN patent holder among Western IDMs
  14. Semiconductor Today — Power Integrations 1250V/1700V PowiGaN for 800VDC AI datacenters
    news semiconductor-today.com first cited by · customer-analyst 2026-05-04
    • GaN now extending to 1250V and 1700V (formerly SiC-only territory) — direct adjacency disruption into NVTS's high-voltage GaN-IC roadmap
    • AI datacenter 800V slot increasingly contested by competing GaN suppliers, not just NVTS density bet
    • Cross-referenced from WOLF/sources.json id=20
  15. Semiconductor Today — TSMC to cease GaN foundry production by end-July 2027
    news semiconductor-today.com first cited by · competitor-analyst 2026-05-04
    • TSMC GaN foundry exit confirmed end-July 2027 due to Chinese pricing pressure
    • Forces Navitas mid-cycle process-portability transition during AI datacenter pull window
    • Most material near-term competitive risk to NVTS roadmap on the supply-chain axis
  16. Semiconductor Today — TSMC to cease GaN foundry production by end-July 2027 (Jul 3, 2025)
    news semiconductor-today.com first cited by · supply-chain-analyst 2026-05-04
    • TSMC GaN foundry exit by end-July 2027 confirmed via independent industry trade press
    • Cited rationale: pricing pressure from Chinese GaN rivals (Innoscience and others)
    • Most-cited supply-chain risk for NVTS in deep-dive analyses across cohort
  17. Stocktitan / Power Semiconductors Weekly — Navitas Q4 2025 strategic shift summary
    news powersemiconductorsweekly.com first cited by · market-positioning-analyst 2026-05-04
    • Confirms FY25 SiC pro-forma revenue contribution ~$145M (combining GeneSiC line)
    • GaN/SiC mix reshape underway — supports the 'right tier of bifurcating market' market-structure stance
    • NVTS positioned as one of NVIDIA-named 800V partners for 2027 production
  18. TI — Quadruples internal GaN manufacturing capacity (Aizu + Dallas)
    news ti.com first cited by · competitor-analyst 2026-05-04
    • TI 4× internal GaN capacity expansion (Oct 2024)
    • 200mm production at both Dallas and Aizu Japan; 300mm pilot complete
    • Demonstrates TI vertical-integration depth materially exceeds Navitas fabless model
  19. TI — Unveils Complete 800 VDC Power Architecture for AI Data Centers with NVIDIA (March 2026)
    news ti.com first cited by · competitor-analyst 2026-05-04
    • TI 800V-to-6V bus converter at 97.6% peak efficiency, >2000W/in³ density
    • TI directly competing on Navitas's flagship single-stage spec with vertical-integration cost structure
    • 30 kW 800V AC/DC PSU and 800V capacitor bank product extensions
  20. Tom's Hardware / DCD on GB200 NVL72 rack pricing and configuration
    news tomshardware.com first cited by · market-positioning-analyst 2026-05-04
    • GB200 NVL36 ~$1.8M, NVL72 ~$3M list per HSBC estimates — baseline for content-per-rack triangulation
    • GB200 NVL72 ~120 kW power consumption — current-state rack power baseline
    • Used to anchor the 'today vs 2027–2028 rack content' bands in voltage-stack section (sell-side ranges, not primary tear-down — flagged as directional)
  21. TrendForce — GlobalFoundries-TSMC-Navitas GaN production hub (Nov 2025)
    news trendforce.com first cited by · market-positioning-analyst 2026-05-04
    • Navitas's foundry-light model uses TSMC and now GlobalFoundries — confirms asset-light differentiation
    • Foundry-fabbed GaN broadens NVTS's potential capacity without IDM capex — relevant to barrier-to-entry trend (declining for fabless tier, rising for IC integration)
    • Cross-references Innoscience IDM scale advantage discussion
  22. TrendForce — Innoscience Scores Major Patent Win Against Infineon as ITC Rules No Infringement
    news trendforce.com first cited by · competitor-analyst 2026-05-04
    • Innoscience won ITC ruling vs Infineon Dec 2025 (no infringement of two remaining patents)
    • PTAB invalidation of EPC '294 patent claims (Mar 2025)
    • Demonstrates GaN patent landscape is fluid; Navitas's IP estate has not yet been challenged but is not assured of indefinite protection
  23. NVIDIA Developer Blog — 800 V HVDC Architecture for AI Factories (May 2025)
    web developer.nvidia.com first cited by · customer-analyst 2026-05-04
    • Infineon named as lead partner in NVIDIA 800V HVDC ecosystem
    • Navitas named as ecosystem partner (lower tier than lead) — relevant for 'reference-design durability' assessment
    • Full production with Kyber rack-scale 2027 — calibrates the catalyst window
    • + 1 more
  24. Bloomberg Tax — Navitas Semi Warns of Material Weakness in Internal Controls (2024)
    news-secondary news.bloombergtax.com first cited by · regulatory-analyst 2026-05-04
    • Material weakness disclosed in 2024 over stock compensation and license-agreement accounting
    • Q1 2024 10-Q delayed; 10-K/A amended; no restatement of prior financials required
    • Establishes baseline disclosure-quality risk for FY2025/FY2026 controls assessment
  25. Cohort companies.json — NVTS entry (id 12)
    internal-synthesis first cited by · customer-analyst 2026-05-04
    • NVTS sentiment +1, mentionCount 4 — thinnest deep-dive corpus support after ETN
    • supportingQuotes from 'The AI Power Crisis — Part 2': single-stage 800V-to-6V demos 'especially revealing... too early to say whether that approach becomes mainstream'
    • Risks captured: smaller scale vs Infineon/TI, no comparable 300mm anchor, density advantage erosion if 300mm cost curve dominates, gallium supply concentration
  26. Cohort companies.json — NVTS entry (id=12) and EV/end-market ecosystem entries
    internal-synthesis first cited by · macro-analyst 2026-05-04
    • NVTS catalysts: single-stage 800V-to-6V design wins, OCP / Computex 2026 reference designs, Kyber/Rubin Ultra 2027 timeline
    • NVTS risks: smaller scale vs IFX/TI, no comparable 300mm anchor, single-stage not yet proven mainstream, gallium critical-mineral concentration
    • EV ecosystem entries (Porsche, BYD, Hyundai, Tesla, Mercedes, BMW, VW) as end-market geography proxy for NVTS GeneSiC EV revenue
    • + 1 more
  27. Cohort corpus — 'Building a Datacenter Part II' (Crucible Capital, April 2025)
    internal-corpus first cited by · customer-analyst 2026-05-04
    • NVTS cited in footnotes only — no main-text BOM-level partnership with Vertiv / Schneider / Delta confirmed
    • Schneider 800V real revenue impact framed as 2028-2030 — primary source for the calendar-mismatch risk for NVTS
    • SuperMicro DCBBS framing relevant as a future Navitas reference-design opportunity but not currently confirmed
  28. Cohort corpus — 'The AI Power Crisis — Part 2' (Crucible / NuttyCLD, May 2025)
    internal-corpus first cited by · customer-analyst 2026-05-04
    • Primary source for 'Infineon scale, TI vertical integration, Navitas density' competitive structure
    • 'Navitas's direct 800V-to-low-voltage demonstrations are especially revealing. It is still too early to say whether that approach becomes mainstream. But it clearly shows where the industry wants to go.'
    • Vertiv / Eaton / Schneider / Delta named as the four box-builder partners at the 800V transition layer — none publicly named with Navitas BOM-level reference design in this corpus
    • + 1 more
  29. Cohort cross-reference AVGO/customer.md
    internal-cross-reference first cited by · customer-analyst 2026-05-04
    • Comparison template for what a 'real' multi-year hyperscaler customer relationship looks like — NRE + per-unit royalty + 24-36 month tape-out cycles + foundry-pass-through
    • Used to calibrate that the NVIDIA-Navitas 'collaboration' has none of those contractual features yet
    • Demonstrates that concentration on dramatically stickier customers is the goal NVTS is aspiring to but has not achieved
  30. Cohort cross-reference NVDA/customer.md
    internal-cross-reference first cited by · customer-analyst 2026-05-04
    • Hyperscaler customer frame — top-2 NVDA customers each >10% revenue; 'Customer A/B/C/D' alphabet-coded disclosure
    • Rack-as-product BOM structure — board to Supermicro/Quanta/Foxconn to hyperscaler — relevant for understanding where NVTS could be designed in (BOM level vs. ecosystem partner level)
    • Pull-through vs channel-fill / pre-buy framework
  31. Cohort cross-reference WOLF/customer.md
    internal-cross-reference first cited by · customer-analyst 2026-05-04
    • EV-OEM auto qualification 12-24 month cycle benchmark applicable to NVTS GeneSiC entry
    • Wolfspeed LTSA roster (GM, Mercedes, JLR, Lucid, Toyota OBC, BorgWarner) is the comparison set NVTS GeneSiC has not matched
    • '$5.8B design-in pipeline' framing flagged as suspect-RPO template — directly applicable to NVTS '$2B+ design-in pipeline' management framing
    • + 1 more
  32. Cohort cross-reference — TSM/regulatory.md (Section 232 derivative-product expansion, Taiwan-tail)
    internal-cross-reference first cited by · regulatory-analyst 2026-05-04
    • Section 232 scope-expansion framework directly applicable to GaN/SiC fabbed at TSMC and Powerchip
    • Cross-strait kinetic event framing as macro-owned tail risk
    • FDPR creep to power-semi adjacency framework
  33. Cohort cross-reference — TSM/supply-chain.md (foundry capacity frame, tier-2 chokepoints, Taiwan-tail)
    internal-cross-reference first cited by · supply-chain-analyst 2026-05-04
    • Aixtron/Veeco MOCVD lock-in framework adapted for GaN epi side
    • Taiwan-tail risk frame (~92% of advanced wafer capacity inside one strait) — adapted to PSMC concentration risk for NVTS
    • Pass-through power inversion logic — TSMC pricing power compares to Navitas's squeezed-middle position
    • + 1 more
  34. Cohort cross-reference — WOLF/regulatory.md (CHIPS direct grant template, §48D mechanics)
    internal-cross-reference first cited by · regulatory-analyst 2026-05-04
    • Lutnick Investment Accelerator template (Intel equity-conversion model) as realistic CHIPS direct-grant outcome path
    • Confirms power-semi BIS exposure framing — civil industrial uses outside Section 232 scope at issuance
    • Section 301/232 directional asymmetry framework adapted to NVTS GaN context
  35. Cohort cross-reference — WOLF/supply-chain.md (SiC substrate market dynamics, Aixtron lock-in)
    internal-cross-reference first cited by · supply-chain-analyst 2026-05-04
    • Confirms SiC substrate ASP collapse mid-teens 2024 — tailwind for GeneSiC unit economics as non-IDM consumer
    • Wolfspeed share fall 60%+ → ~34% (2021-2024); Coherent (II-VI), SK Siltron CSS, ROHM, TanKeBlue, SICC as the merchant alternatives
    • Aixtron G10-SiC / Planetary lock-in case study — directly transposable to GaN epi tools at PSMC and GF Burlington
    • + 1 more
  36. Cohort sibling — NVDA/macro.md (Taiwan-tail probability and AI-capex framing reused)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • Taiwan kinetic-event probability ~2-4%/yr, blockade ~5-8%/yr over 2026-2030 — applied to NVTS production base
    • Hyperscaler operating-cash-flow funding model for AI capex (insulating from credit cycle) — applies to NVTS AI-DC revenue ramp
    • USD/TWD wafer-cost dynamic — applies to NVTS TSMC + PSMC GaN cost base
    • + 1 more
  37. Cohort sibling — NVTS/competitor.md (competitor-analyst, 2026-05-04)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • Innoscience #1 GaN power-device share ~30% Yole 2024 vs Navitas ~17% — frames China-domestic-build headwind
    • TSMC GaN foundry exit by July 2027; PSMC + GlobalFoundries transition — frames Taiwan-tail concentration timeline
    • Innoscience as sole Chinese partner on NVIDIA 800V list — frames asymmetric AI-DC competitive risk
    • + 1 more
  38. Cohort sibling — NVTS/customer.md (customer-analyst, 2026-05-04)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • End-market mix FY25 estimates: mobile 60-70%, AI-DC 5-8%, EV 10-15%, solar 10-12%, industrial 5-8% — basis for blended cycle position table
    • Two distributors >10% of revenue, ~17% top customer concentration — frames revenue-side FX exposure via Asian distributors
    • NVIDIA collaboration is non-binding May 2025 — caveat on AI-DC ramp timing
    • + 1 more
  39. Cohort sibling — NVTS/financial.md (financial-analyst, 2026-05-04)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • 83x EV/Sales valuation framing as anchor for terminal-multiple sensitivity to rates
    • FY25 revenue $46M (down 45%); $237M cash post Nov-2025 PIPE; ~7-yr runway at $8M/quarter burn
    • $202M of common stock issuance FY25 — recursive equity-funding-cost channel of rate sensitivity
    • + 2 more
  40. Cohort sibling — NVTS/market.md (market-positioning-analyst, 2026-05-04)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • Yole Power GaN 2025: $355M (2024) → ~$3B (2030) at 42% CAGR; data-center slice ~$380M by 2030 at 53% CAGR
    • SiC market correction through 2027-2028 (Yole) — frames NVTS GeneSiC headwind in 2026 cycle table
    • Datacenter GaN at Y0 of volume ramp — frames calendar-mismatch between 2026-tactical-loser and 2027+-structural-winner
    • + 1 more
  41. Cohort sibling — WOLF/competitor.md (SiC competitive frame Navitas inherits via GeneSiC)
    internal-cross-reference first cited by · competitor-analyst 2026-05-04
    • SiC device share 2024 (TrendForce): STM ~33%, onsemi ~25%, Infineon ~15%, Wolfspeed ~11%, Rohm + others balance
    • Navitas/GeneSiC sits in 'others' — no top-5 SiC device share in any 2024 data; SiC is supplemental optionality not a primary moat
    • Wolfspeed substrate share collapse 60%+→34% (2021-2024) frames how fast structural transitions can punish single-product power-semi pure-plays
  42. Cohort sibling — WOLF/financial.md (cautionary-tale benchmark)
    internal-cross-reference first cited by · financial-analyst 2026-05-04
    • Capex-light (NVTS $1.5M) vs capex-heavy (WOLF $1.27B FY25) contrast — NVTS structurally less catastrophic
    • But: WOLF at 3.5x EV/Sales vs NVTS at 82.7x — multiple risk worse at NVTS
    • Both share negative ROIC ~30%, SBC discipline issues, dependence on hyperscaler design-win timeline
  43. Cohort sibling — WOLF/macro.md (rate / EV-cycle / Taiwan-tail framing reused)
    internal-cross-reference first cited by · macro-analyst 2026-05-04
    • Auto-loan rate environment 2023-2026 suppressing US/EU EV unit demand — applies to NVTS GeneSiC EV revenue
    • Residential solar payback math rate-sensitivity — applies to NVTS Enphase / solar microinverter revenue
    • SiC substrate ASP deflation under Chinese competition — applies to NVTS GeneSiC SiC product line pass-through power
    • + 2 more
  44. Cohort synthesis (semiconductor-industry) — macro lens for NVTS
    internal-synthesis first cited by · macro-analyst 2026-05-04
    • GaN three-way competitive structure: 'Infineon scale, TI vertical, Navitas density' as anchor framing for NVTS strategic position
    • Section 6 contested claim #2 — single-stage 800V-to-low-voltage 'too early to say whether mainstream' is the binding macro-uncertainty for NVTS
    • Open Question #2 — GaN three-way race timing; 300mm cost curve dominance risk before density advantage matters
    • + 3 more
  45. Cohort synthesis.md — semiconductor-industry chip-to-grid value chain (2026-05-04 refresh)
    internal-synthesis first cited by · customer-analyst 2026-05-04
    • L8b value-chain placement: 'high-density GaN conversion; the density-bet pure-play in the 800V transition layer'
    • Theme #3.3 framing: 'Infineon fights with scale, TI with vertical integration, and Navitas with density'
    • Theme #3.2 chip-to-grid pass-through: Vertiv Q4'25 organic orders +152% YoY, $15B backlog, 2.9x book-to-bill; Eaton Q4'25 DC orders +3x YoY
    • + 4 more
  46. Compound Semiconductor — Innoscience files lawsuit against Infineon (Suzhou, Jan 2025)
    news-secondary compoundsemiconductor.net first cited by · regulatory-analyst 2026-05-04
    • Counter-suit in Suzhou Intermediate People's Court (patents 202311774650.7 and 202211387983.X)
    • Signals enforceability of Chinese GaN patents against foreign players — read-through to NVTS IP enforcement strategy in China
  47. DigiTimes — Infineon's GaN patent wall forces global firms to rethink China ties (Dec 5, 2025)
    news-secondary digitimes.com first cited by · regulatory-analyst 2026-05-04
    • Innoscience 30% global GaN share (2024) vs Navitas 17% — direct market-share comparison
    • Patent ecosystem framed as defining factor in China-coupled supply chain decisions
  48. GlobalFoundries press release — GF and Navitas Partner to Accelerate U.S. GaN (Nov 20, 2025)
    company-primary gf.com first cited by · regulatory-analyst 2026-05-04
    • GF Burlington Vermont GaN production for NVTS — development early 2026, production late 2026
    • National-security framing in announcement language
    • Not a CHIPS Act §9902 direct grant; sits indirectly under GF's $1.5B Malta PMT
  49. GlobalFoundries — Long-term Strategic Partnership with Navitas for U.S. GaN (Nov 20, 2025)
    company-press-release gf.com first cited by · supply-chain-analyst 2026-05-04
    • GF licensed TSMC's 80V and 650V GaN power-semi processes
    • Development with Navitas scheduled for early 2026
    • Production expected later 2026 at GF Burlington, Vermont fab
    • + 2 more
  50. Infineon press release — Infineon sues Innoscience for Patent Infringement (March 2024)
    company-primary infineon.com first cited by · regulatory-analyst 2026-05-04
    • Four GaN patent infringement claims filed at ITC and N.D. Cal.
    • Patents in scope: US8686562B2, US9899481B2, US8264003B2, US9070755B2
    • Establishes baseline for ITC 337-TA-1407 docket
  51. Infineon Technologies AG Annual Report FY2025
    company-filings infineon.com first cited by · financial-analyst 2026-05-04
    • IFX FY25 adj. gross margin ~43%; Power & Sensor Systems segment-result margin ~14.9%
    • FY26 adj. gross margin guidance low-40s
    • Anchor for segment-level (not consolidated) comp on the GaN/power axis
  52. Macro background — rates, FX, AI-capex, end-market cycle baselines
    general-knowledge first cited by · macro-analyst 2026-05-04
    • US 10y ~4.0-4.5% / real rates 1.5-2.0% / Fed funds 3.75-4.25% / DXY mid-100s as current-regime baseline
    • Auto-loan rate environment 2023-2026 suppressing US/EU EV unit demand 5-10% per 100bp
    • Residential solar volume rate-sensitivity 10-15% per 100bp over 12-month lag
    • + 4 more
  53. Navitas and NVIDIA Collaborate on 800 V HVDC Power Architecture (Computex 2025)
    company-press-release ir.navitassemi.com first cited by · customer-analyst 2026-05-04
    • Strategic collaboration with NVIDIA on 800V HVDC architecture announced May 21, 2025 at Computex
    • Navitas positioned as ecosystem partner in NVIDIA 800V HVDC AI factory architecture
    • No volume commitment, no LTA, no preferential supply terms disclosed — collaboration only
  54. Navitas Investor Day Presentation (September 2025)
    investor-presentation ir.navitassemi.com first cited by · customer-analyst 2026-05-04
    • Five-end-market framework: mobile/consumer, AI data center, EV, solar/energy storage, motor drive/industrial
    • Reference-design partner roster includes NVIDIA, Murata, Enphase publicly named
    • GeneSiC SiC product-line roadmap — auto OBC and DC fast-charging design-in references
    • + 1 more
  55. Navitas press release — 200mm GaN production with PSMC (July 1, 2025)
    company-primary navitassemi.com first cited by · regulatory-analyst 2026-05-04
    • Powerchip Fab 8B (Zhunan Science Park, Taiwan) qualified for NVTS 100V–650V GaN-on-Si
    • 100V family targets 1H26 production; 650V transitions over 12–24 months
    • Concentrates supply in Taiwan jurisdiction — Section 232 derivative-tariff exposure
  56. Navitas Q4 2025 / Full Year 2025 earnings press release
    company-press-release navitassemi.com first cited by · financial-analyst 2026-05-04
    • FY2025 revenue $45.9M vs FY24 $83.3M (-45% YoY)
    • GAAP net loss $117.0M; non-GAAP net loss $41.6M
    • Q4'25 revenue $7.3M; high-power markets first majority of revenue
    • + 2 more
  57. Navitas Q4 2025 earnings call transcript (Feb 24, 2026)
    earnings-call seekingalpha.com first cited by · financial-analyst 2026-05-04
    • Mobile dropped from majority Q3'25 to <25% of Q4'25 revenue; 'insignificant' by end-2026
    • Q4 framed as the bottom; sequential revenue growth expected through 2026
    • Operating expenses guided flat at ~$15M/quarter through 2026
    • + 1 more
  58. Navitas Semiconductor 10-K filed Feb 27, 2026 — supply-chain disclosures via Stocktitan summary
    sec-filing stocktitan.net first cited by · supply-chain-analyst 2026-05-04
    • Risk language: 'We have historically relied on a single third-party manufacturer (wafer foundry) to fabricate our GaN products, and on a separate, single wafer foundry to fabricate our SiC products.'
    • Single/limited source language extends to key materials and components
    • TSMC GaN production cease by July 2027 disclosed; mitigation via Powerchip and GlobalFoundries with buffer-inventory build
    • + 4 more
  59. Navitas Semiconductor FY2024 10-K (filed March 2025)
    sec-filing sec.gov first cited by · customer-analyst 2026-05-04
    • Customer concentration: 2 distributors >10% of revenue in FY24 (down from 3 in FY23)
    • Top distributor ~17% of revenue (down from ~22% FY23) — concentration easing reflects mobile demand decline rather than diversification
    • No long-term agreements, take-or-pay, or volume guarantees disclosed in filing
    • + 1 more
  60. Navitas Semiconductor FY2025 Form 10-K (filed Feb 27, 2026)
    sec-filing ir.navitassemi.com first cited by · regulatory-analyst 2026-05-04
    • Item 9A controls assessment baseline for any new or unremediated material weakness
    • Item 1A risk factors — TSMC GaN exit by July 2027; PSMC and GF Burlington qualification
    • Item 3 legal proceedings — confirms no active securities class action as of February 2026
  61. Navitas Semiconductor Q1/Q2/Q3 FY25 10-Q filings and earnings calls
    sec-filing ir.navitassemi.com first cited by · customer-analyst 2026-05-04
    • Distributor inventory normalization called out in Q1/Q2 FY25 — channel-fill reversal evidence
    • AI-DC revenue framed as 'small but fastest-growing' — implies <10% of FY25 run-rate
    • Mobile/consumer fast-charge ~60-70% of revenue [inferred from management commentary, not directly disclosed by percentage]
    • + 2 more
  62. Navitas Semiconductor — GeneSiC Acquisition Close (Aug 2022)
    company-press-release ir.navitassemi.com first cited by · customer-analyst 2026-05-04
    • Acquisition closed Aug 2022 — basis for Navitas's SiC product line
    • Provides EV traction-inverter, on-board-charging, DC-fast-charge SiC product entry
    • Acquisition consideration ~$244M; primary footprint of post-deal SiC revenue contribution
  63. Navitas Semiconductor — Q4 and Full Year 2025 Financial Results (Feb 24, 2026)
    company-press-release globenewswire.com first cited by · market-positioning-analyst 2026-05-04
    • FY25 revenue $45.9M vs FY24 $83.3M — concrete revenue decline tied to mobile exit
    • Q4'25 trough revenue ~$7M; designated as expected bottom
    • Mobile <25% of Q4'25 revenue — confirms strategic pivot is in execution, not aspiration
    • + 2 more
  64. Navitas — 800 VDC Power Architecture for NVIDIA AI Factory press release (Oct 2025)
    company-press-release navitassemi.com first cited by · supply-chain-analyst 2026-05-04
    • Navitas spans first stage (grid to 800VDC SST), second stage (800VDC to 54V/12V), and third stage (POL)
    • 100V GaN FETs with dual-sided cooled packages mentioned for GPU power boards
    • 650V GaN FET portfolio + GaNSafe ICs (integrated control/drive/sensing/protection) referenced
    • + 2 more
  65. Navitas — AI Data Center Opportunity investor presentation (August 2025)
    investor-presentation ir.navitassemi.com first cited by · market-positioning-analyst 2026-05-04
    • $2.4B lifetime design-win pipeline disclosure — concrete near-term opportunity anchor (lifetime cumulative, not annual)
    • NVIDIA 800VDC partner status confirmed — basis for reference-design momentum claim
    • End-market framework: AI data center, grid/energy infrastructure, performance computing, industrial electrification, mobile/consumer/appliance (the latter being de-prioritized)
  66. Navitas — Plans for 200mm GaN Production with PSMC (Powerchip) press release (Jul 1, 2025)
    company-press-release navitassemi.com first cited by · supply-chain-analyst 2026-05-04
    • PSMC partnership for 200mm GaN-on-Si in Fab 8B, Zhunan Science Park, Taiwan
    • 180nm CMOS-class process, voltage range 100V to 650V
    • Initial qualification Q4 2025; 100V family volume production at PSMC 1H 2026
    • + 2 more
  67. Navitas — Redefining Data Center Power: GaN and SiC for 800 VDC Infrastructure white paper (Oct 2025)
    company-presentation navitassemi.com first cited by · market-positioning-analyst 2026-05-04
    • 10 kW DC-DC platform at 98.5% peak efficiency, 1 MHz switching — anchors the density narrative
    • 12 kW platform with GeneSiC + GaNSafe + Intelliweave for 500 kW rack power — positions NVTS at the IC tier of the rack power chain
    • GaN wins at 800V→6V and intermediate bus stages (P2/P3 of NVIDIA HVDC architecture) — used in voltage-stack content-per-rack section
  68. Navitas — TSMC & Amkor Manufacturing Partnerships press release (Oct 17, 2017)
    company-press-release navitassemi.com first cited by · supply-chain-analyst 2026-05-04
    • Amkor named as packaging, test and logistics partner — 2017 baseline relationship
    • Amkor provides 'high-volume and low-cost QFN packaging platform' for Navitas GaN
    • TSMC named as wafer foundry — 2017 baseline GaN-on-Si relationship
    • + 1 more
  69. Navitas-GeneSiC acquisition press release (Aug 2022)
    company-press-release navitassemi.com first cited by · market-positioning-analyst 2026-05-04
    • $100M cash + 24.9M shares + earn-out structure — basis for SiC product line acquisition
    • GeneSiC TAM-statement '$20B+ aggregate market opportunity by 2026' explicitly flagged as inflated round-number sizing — used Yole instead
    • Acceleration into EV / solar / energy storage markets by 2-3 years — context for SiC mix reshape
  70. NIST CHIPS for America program portal
    government-primary nist.gov first cited by · regulatory-analyst 2026-05-04
    • NVTS not listed among CHIPS Act Section 9902 direct funding recipients
    • GlobalFoundries Malta NY $1.5B PMT (Feb 2024) covers Burlington VT GaN partnership site indirectly
    • TXN $1.6B and Wolfspeed $750M PMTs cited as cohort comparators
  71. Semiconductor Today — TSMC to cease GaN foundry production by end-July 2027 (July 3, 2025)
    news-secondary semiconductor-today.com first cited by · regulatory-analyst 2026-05-04
    • TSMC publicly announces GaN foundry exit by end of July 2027 due to Chinese price pressure
    • Hard-dated supply-chain cliff for NVTS GaN volume
    • PSMC and GF Burlington qualification must complete before this date
  72. Stanford Securities Class Action Clearinghouse — NVTS docket search (no class action filed as of May 2026)
    legal-database securities.stanford.edu first cited by · regulatory-analyst 2026-05-04
    • No 10b-5 class action against Navitas Semiconductor disclosed through May 2026
    • Notable for a small-cap with prior internal-controls material weakness
  73. StockAnalysis.com — MPWR Statistics (peer comp)
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • MPWR EV/Sales 25.85x, EV/EBITDA 89.39x
    • MPWR gross margin 55.18%, operating margin 27.08%
    • Closest fabless power-management high-multiple comp; even MPWR is 3x cheaper EV/Sales than NVTS despite 8x higher GM
  74. StockAnalysis.com — NVTS Balance Sheet
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • YE'25 cash $236.86M; total debt $6.47M (net cash ~$230M)
    • Goodwill $163.22M (entirely from GeneSiC, unimpaired since Aug 2022 close)
    • Intangible assets $53.26M, amortizing
    • + 1 more
  75. StockAnalysis.com — NVTS Cash Flow Statement
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • FY22-FY25 OCF: -$44.5M / -$41.4M / -$58.8M / -$42.9M
    • FY22-FY25 capex: -$4.6M / -$4.8M / -$6.8M / -$1.5M (fab-lite)
    • FY22-FY25 FCF: -$49.1M / -$46.2M / -$65.6M / -$44.4M
    • + 2 more
  76. StockAnalysis.com — NVTS Income Statement (annual)
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • FY22 revenue $37.94M; FY23 $79.46M; FY24 $83.30M; FY25 $45.92M
    • Gross margins: FY22 31.5%, FY23 39.1%, FY24 34.0%, FY25 31.0%
    • Operating losses: FY22 ($123.6M), FY23 ($118.1M), FY24 ($130.7M), FY25 ($107.8M)
    • + 1 more
  77. StockAnalysis.com — NVTS Statistics & Valuation
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • Market cap $3.87B; Enterprise value $3.80B; price $16.77 (May 4, 2026)
    • Shares outstanding 230.79M (up 12.7% YoY)
    • EV/Sales (TTM) 82.69x; P/S 87.71x
    • + 3 more
  78. StockAnalysis.com — ON Statistics (peer comp)
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • ON EV/Sales 6.94x, EV/EBITDA 23.70x
    • ON gross margin 38.32%, operating margin 17.84%, FCF margin 23.66%
  79. StockAnalysis.com — TXN Statistics (peer comp)
    data-aggregator stockanalysis.com first cited by · financial-analyst 2026-05-04
    • TXN EV/Sales 14.36x, EV/EBITDA 30.55x, P/E 48.03x
    • TXN gross margin 57.32%, operating margin 35.96%, FCF margin 20.18%
  80. TrendForce — Innoscience Scores Major Patent Win Against Infineon as ITC Rules No Infringement (Dec 5, 2025)
    news-secondary trendforce.com first cited by · regulatory-analyst 2026-05-04
    • ALJ initial determination favors Innoscience on remaining two patents (US 9,070,755 and US 9,899,481)
    • ITC Commission final determination scheduled April 2, 2026
    • Sets precedent for US patent enforceability of GaN-Systems-era patents — read-through to NVTS IP moat
  81. USTR Section 301 four-year review and excess-capacity investigation (March 2026)
    government-primary ustr.gov first cited by · regulatory-analyst 2026-05-04
    • March 11, 2026 USTR investigation covers excess capacity in 16 economies including China
    • Semiconductors and EVs explicitly within investigation scope
    • Possible HTS extension to Chinese-origin GaN/SiC discrete devices — potential positive asymmetry for NVTS
  82. White & Case — President Trump orders narrowly targeted 25% Section 232 tariff (January 2026)
    law-firm-analysis whitecase.com first cited by · regulatory-analyst 2026-05-04
    • Confirms narrow scope at issuance — power semiconductors not covered
    • Identifies derivative-product expansion mechanism Commerce can use
    • Frame for assessing NVTS Taiwan-fabbed import exposure
  83. White House Proclamation — Section 232 25% tariff on advanced semiconductors (Jan 14, 2026)
    government-primary whitehouse.gov first cited by · regulatory-analyst 2026-05-04
    • Section 232 25% tariff effective January 15, 2026; scope narrow (H200/MI325X-class)
    • Proclamation contemplates future scope expansion by Commerce — derivative-product expansion risk for GaN/SiC
    • Power semiconductors outside scope at issuance; civil industrial uses exempt