IPWR reported Aug 13 — this analysis reviews the prior quarter.

Ideal Power Inc. (IPWR) | The Buildout — AI Infrastructure

Mkt cap · 52-wk · YTD · delayed
Updated Jul 11, 2026Q1 FY2026 reviewed
Ideal Power develops B‑TRAN bidirectional solid‑state power switches enabling solid‑state circuit breakers for 800‑V DC AI data centers.
Funnel >$300M
Sales opportunity funnel surged >50% in ~2.5 months vs Feb 2026.
Hyperscaler LOI signed
Co‑developing SSCB for NVIDIA Rubin Ultra, prototype end‑Q4 2026.
Cash burn below guide
Q1 2026 cash burn $2.3M, below guided $2.6–2.8M.
Pre‑revenue, $0 revenue
No meaningful revenue timeline; cash burn guided $10–10.5M FY2026.
The Buildout Takeaway
Ideal Power’s commercial traction accelerated sharply, with a surging sales funnel and a signed LOI linking B‑TRAN to a U.S. hyperscaler’s NVIDIA Rubin Ultra data center. Yet the company has yet to generate revenue and holds no firm purchase orders; the entire investment case rests on converting a now‑>$300M pipeline into binding contracts in the next 3–6 months.
10 analysts·7 Buy3 Hold0 Sell
Coverage is thin — no price estimates on file, so no target is shown

FY2026 cash burn guided ~$10.0–10.5M; no revenue guidance provided.
Important: The Buildout is a data analytics platform. Content is generated by algorithms and AI agents using public filings, earnings transcripts, and market data. This is not personalized investment advice.
Our View

The Verdict

Ideal Power is an early‑stage fabless semiconductor company that designs a novel bidirectional solid‑state power switch called B‑TRAN. Conventional power switches based on silicon carbide typically need back‑to‑back configurations to handle current in both directions, adding size, cost, and complexity. B‑TRAN’s inherent bidirectionality and microsecond fault‑clearing capability make it a natural fit for solid‑state circuit breakers — a technology that the 800‑volt DC power architecture emerging in AI data centers depends on to safely distribute massive electrical loads. By targeting the solid‑state circuit breaker socket on the NVIDIA Rubin Ultra platform, IPWR aims to solve a critical protection problem that mechanical breakers cannot address in next‑generation GPU racks.

Market Cap
Revenue (TTM)$0M
Net Cash$16M
Earnings Beats3 of 7
P/E (TTM)
EV/EBITDA (TTM)

What We Like

  • Sales opportunity funnel exceeded $300M as of May 2026, up >50% in under three months, with a 50/50 split between automotive and data‑center/industrial applications.
  • Signed LOI with an industry partner to co‑develop a B‑TRAN‑enabled intelligent SSCB prototype for a U.S. hyperscaler, explicitly tied to the NVIDIA Rubin Ultra 800‑V DC AI data center — prototype delivery expected end‑Q4 2026 with a purchase order.
  • Lead Asia customer (one of the largest circuit protection equipment manufacturers in Asia) expanded from one to three active SSCB projects; prototypes and initial low‑volume orders on track for Q4 2026.
  • First solid‑state transformer (SST) customer engagements initiated, opening a new application vertical for B‑TRAN in AI data centers and grid infrastructure.
  • 103 issued patents plus a trade‑secret double‑sided wafer process create a defensible intellectual property moat.

What We’re Watching

  • The company has yet to generate revenue; purchase orders expected in the next 3–6 months (by ~November 2026) will test whether the funnel converts.
  • Automotive reliability testing completion (summer 2026) — a binary catalyst to unlock the EV vertical with Stellantis and other automakers.
  • Post‑quarter $30M equity raise on 18 May 2026 strengthened the balance sheet but diluted existing holders by ~44% relative to the prior fully diluted count.
  • Infineon’s CoolSiC JFET and other incumbent power‑semiconductor solutions could erode B‑TRAN’s technical differentiation if they deliver comparable solid‑state circuit protection.
Bottom Line

The thesis that B‑TRAN can become the standard power switch for solid‑state circuit breakers in 800‑V DC AI data centers is strengthening. The sales funnel grew dramatically, a U.S. hyperscaler signed an LOI tied to NVIDIA’s Rubin Ultra, and the lead Asia customer expanded to three projects — all pointing to real customer pull. However, the company has yet to generate revenue, and no purchase orders have materialized yet. The key test is whether management can deliver on prototype timelines and begin converting the opportunity pipeline into production orders in the coming quarters.

Next upNear term, automotive reliability testing completion (summer 2026) could unlock the EV vertical. The first major test of AI data‑center traction arrives in Q4 2026 with prototype deliveries to the lead Asia customer and the U.S. hyperscaler, each accompanied by an expected purchase order.
Last Quarter — Q1 FY2026

Earnings

Ideal Power reported zero product revenue in Q1 2026, as expected. Operating expenses rose to $3.7M from $2.8M a year earlier, driven by CEO‑transition stock‑based compensation and new sales hires. Net loss widened to $3.6M from $2.7M in Q1 2025. Operating cash burn (operating & investing) came in at $2.3M, below the guided $2.6–2.8M, and quarter‑end cash stood at $16.4M with no debt.

MetricQ1 FY2026Q4 FY2025Q1 FY2025YoY
Revenue$0M$0M$0M
Gross margin-157.1%
EBITDA−$4M−$2M−$3M+33.3%
EPS$-0.32$-0.21$-0.30+8.3%
the power systems delivering electricity to operate the next generation of those chips are now the limiting factor, and the industry is mobilizing to address it.— David Somo, President and CEO, 14 May 2026

Management tone: Management’s tone on the Q1 2026 call was confident and execution‑oriented under new CEO David Somo, who emphasized operational discipline, biweekly funnel reviews, and the priority of converting the sales pipeline into orders.

Management Guidance

Management guided Q2 2026 cash burn of $2.5–2.7M and full‑year 2026 cash burn of approximately $10.0–10.5M, reflecting additional sales and engineering hires. No revenue guidance was issued; when asked about a meaningful revenue timeline, management referred to the prior business‑update call and said 'no change in our expectation' without providing a specific date.

Business Trajectory

Trajectory

Revenue has remained at zero as Ideal Power is still in the prototyping phase. Operating expenses rose from $2.8M in Q1 2025 to $3.7M in Q1 2026, largely reflecting stock‑based compensation tied to the CEO transition and headcount additions. Quarterly operating cash burn increased slightly to $2.3M from $2.1M a year earlier, but came in below the guided range. The company’s no‑revenue trajectory will only change when initial evaluation orders begin to convert to production revenue — something management expects in the coming quarters.

Revenue & Margin Trajectory
RevenueGross margin$0$0$1$0M$0M$0M$0M$0M$0M$0M$0M-1213%0%Q2'24Q3Q4Q1'25Q2Q3Q4Q1'26
RevenueGross margin$0$0$1$0M$0M$0M$0M$0M$0M$0M$0M-1213%0%Q2'24Q3Q4Q1'25Q2Q3Q4Q1'26
Gross margin as reported.
Share Price — 12 Months
$2$5$8$052-wk high $9Aug '25OctJan '26AprAug '26
52-week range $3–$9.
Share Price — 12 Months
$2$5$8$052-wk high $9Aug '25OctJan '26AprAug '26
52-week range $3–$9.
The Numbers

The Model

The model projects FY+1 revenue of $0.2M and an EBITDA loss of $14M (‑6,400% margin), followed by FY+2 revenue of $2.0M and an EBITDA loss of $16M (‑860% margin). The near‑term projection is anchored by the expectation that initial low‑volume orders from prototypes will begin to trickle in during FY+1 but remain immaterial, while FY+2 assumes some design wins progressing to small production volumes. EBITDA losses remain large as R&D and sales investments outpace modest revenue.

Revenue & EBITDA Projections
REVENUE$0M$0M$2MFY25FY+1 (E)FY+2 (E)EBITDA & MARGIN−$10M−$14M−$16MFY25FY+1 (E)FY+2 (E)
REVENUE$0M$0M$2MFY25FY+1 (E)FY+2 (E)EBITDA & MARGIN−$10M−$14M−$16MFY25FY+1 (E)FY+2 (E)
Solid bars are reported actuals; outlined bars are model projections — not company guidance.
MetricFY2025Next FY (E)Following FY (E)
Revenue$0M$0M$2M
YoY Growth+900.0%
EBITDA−$10M−$14M−$16M
EBITDA Margin-6400.0%-860.0%

Projections are the median of 5 independent model runs.

Management guided Q2 2026 cash burn of $2.5–2.7M and full‑year 2026 cash burn of approximately $10.0–10.5M, reflecting additional sales and engineering hires. No revenue guidance was issued; when asked about a meaningful revenue timeline, management referred to the prior business‑update call and said 'no change in our expectation' without providing a specific date.

What Could Go Right — and Wrong

What good looks like
  • The U.S. hyperscaler prototype evaluation succeeds, leading to a production design win for the NVIDIA Rubin Ultra platform and triggering purchase orders from multiple hyperscalers.
  • The lead Asia customer’s three SSCB projects advance from prototypes to volume production, generating sustained revenue from a top‑tier circuit protection manufacturer.
  • Automotive qualification completes on time and Stellantis transitions from development to field trials and a new production purchase order, opening the EV drivetrain vertical.
  • Additional strategic partnerships or investments from major power‑solution or hyperscale companies validate B‑TRAN and accelerate adoption.
  • The 800‑V DC transition in AI data centers accelerates, pulling in broad demand for B‑TRAN‑based solid‑state circuit breakers and solid‑state transformers.
What could go wrong
  • The sales funnel fails to convert into binding purchase orders; prototypes are delayed or underperform, and customers opt for alternative SiC‑based solutions from Infineon or other incumbents.
  • Automotive reliability testing is delayed or fails, preventing B‑TRAN from entering the EV market and undermining a full half of the pipeline.
  • Cash burn persists without meaningful revenue; despite the $30M raise, further dilution is required before commercial traction materializes, eroding shareholder value.
  • A large incumbent introduces a competitive solid‑state circuit breaker product that obviates B‑TRAN’s differentiation, shutting IPWR out of the market.
  • The 800‑V DC transition proceeds more slowly than anticipated, or hyperscalers adopt alternative power architectures that do not require bidirectional solid‑state breakers.
What’s Next

Looking Ahead

The next twelve months will be defined by the company’s ability to demonstrate commercial traction. Completion of automotive reliability testing in summer 2026 is the nearest binary catalyst, followed by prototype deliveries and expected purchase orders from the lead Asia customer and the U.S. hyperscaler in Q4 2026. The newly initiated solid‑state transformer engagements could broaden the addressable market, while ongoing strategic investment discussions may bring a validating partner. Failure to secure firm orders within the indicated 3–6 month window would raise serious questions about market adoption.

Catalysts
  • Summer 2026Automotive reliability testing completion — Removes a key gate for EV production deployment; enables Stellantis field trials.
  • Mid‑2026Stellantis remaining deliverables — Completion of current PO could trigger next‑phase discussions or a new PO.
  • ~Nov 2026Expect first purchase orders — Management anticipates commercial announcements in the next 3–6 months.
  • Q4 2026Lead Asia SSCB prototype delivery — Low‑current SSCB prototype and initial low‑volume orders; first step toward revenue.
  • End‑Q4 2026U.S. hyperscaler SSCB prototype — Co‑developed intelligent SSCB prototype delivery for NVIDIA Rubin Ultra; expected PO.
Numbers

Financials

Annual Summary

MetricFY2025TTM
Revenue$0M$0M
Gross Margin-112.1%
EBITDA−$10M−$22M
EBITDA Margin
Net Income−$10M−$11M
Free Cash Flow−$9M−$19M
Net Cash

Key Ratios (Trailing)

Valuation
  • P/E TTM
  • EV/EBITDA TTM
  • EV/Revenue TTM
  • Price/FCF TTM
Profitability
  • ROIC-513.5%
Reference

The Company

Ideal Power is a fabless semiconductor company developing a novel solid‑state bidirectional power switch called B‑TRAN. The device uses a patented bipolar junction transistor design that can conduct current in both directions with very low losses, unlike conventional power switches. This makes B‑TRAN particularly well suited for solid‑state circuit breakers that must interrupt faults in microseconds while handling high power — a requirement emerging in next‑generation 800‑volt DC power distribution systems for AI data centers, electric vehicles, and grid infrastructure.

The company operates on a fabless model, outsourcing all wafer fabrication and packaging to undisclosed third‑party foundries and packaging firms. Its sole physical facility is a 5,775‑square‑foot leased office and laboratory in Austin, Texas. The double‑sided wafer process that produces B‑TRAN is treated as a trade secret, and the company maintains a portfolio of 103 patents. Ideal Power has no debt and funds its operations through equity sales, with a recent $30M registered direct offering in May 2026 strengthening its cash position.

Business Segments

Discrete B‑TRAN
Management’s first commercialized device
A single‑die packaged bidirectional switch designed for low‑conduction‑loss applications in solid‑state circuit protection and EV contactors.
Growth driver: Adoption in 800‑V DC SSCBs for AI data centers and EV systems.
SymCool Power Module
Multi‑die power module
A multi‑die module packaging multiple B‑TRAN dies to handle higher currents and power levels, targeting solid‑state circuit breakers and EV traction inverters.
Growth driver: Demand from hyperscalers and automakers for high‑power, bidirectional

Competitive Landscape

Ideal Power competes with major power‑semiconductor incumbents, including Infineon, Texas Instruments, onsemi, STMicroelectronics, and others, that offer established silicon and silicon carbide devices for power management. On the Q1 2026 call, management specifically addressed Infineon’s CoolSiC JFET as a competing technology for solid‑state circuit breakers, arguing that B‑TRAN’s inherent bidirectionality allows a single device to replace back‑to‑back SiC configurations, reducing component count and cost. The competitive dynamic hinges on whether B‑TRAN’s technical differentiation translates into compelling system‑level advantages that outweigh incumbents’ deep customer relationships and manufacturing scale.

  • Infineon
    Direct competitor discussed on the Q1 2026 call; its CoolSiC JFET is a competing technology for SSCBs. Management argues B‑TRAN’s single‑device bidirectionality offers a system‑level advantage.
  • Texas Instruments
    Broad incumbent in power management; data‑center revenue grew ~90% YoY, but TI does not yet offer a direct bidirectional SSCB competitor. Its scale and customer reach pose a long‑term threat.
  • onsemi
    Named in 10‑K as a competitor; not discussed in detail on the call.
  • STMicroelectronics
    Named in 10‑K as a competitor; not discussed in detail on the call.
  • Mitsubishi Electric
    Named in 10‑K as a competitor; not discussed in detail on the call.
Competitor names derived from the company’s 10‑K and Q1 2026 call transcript.

Supply Chain

Ideal Power occupies a component supplier position in the power‑system value chain, providing B‑TRAN switches to circuit‑breaker and power‑system manufacturers who in turn serve hyperscalers, automakers, and grid operators. No neighbor in our supply‑chain scans mentioned Ideal Power by name.

Supplier
Foundry partners (undisclosed)
Wafer fabrication for double‑sided B‑TRAN process
Supplier
Packaging partners (undisclosed)
Assembly and test of discrete devices and modules
Inherent bidirectionality in a single device
IPWR
Designs and tests B‑TRAN devices in‑house; outsources all wafer fabrication and assembly.
Stellantis
1 active purchase order
Custom B‑TRAN power module for next‑gen EV drivetrain inverter; top‑10 global automaker.
Lead Asia customer (unnamed)
3 active projects
One of the largest circuit protection equipment manufacturers in Asia; first SSCB design win, now expanded to medium‑current and smart‑building projects.
U.S. hyperscaler (via industry partner LOI)
1 signed LOI
Co‑development of intelligent SSCB for NVIDIA Rubin Ultra 800‑V DC data center; prototype expected end‑Q4 2026.
Asia‑based SST suppliers
First engagements
Exploring B‑TRAN for solid‑state transformers in AI data centers and grid applications.

Analysis updated Jul 11, 2026, reviewing Q1 FY2026. Prices delayed. Built with The Buildout’s published methodology. Not investment advice. No positions held. © The Buildout 2026.