Ideal Power's Funnel Surges 50% as Hyperscaler Deal and New Data Center Projects Signal Inflection—But Revenue Remains Elusive
Q1 2026 Earnings Call, May 14, 2026
Ideal Power delivered a quarter heavy on commercial momentum and light on revenue—zero dollars recorded for the third consecutive period—but management used the call to unveil a cluster of genuinely new developments that move the B-TRAN story meaningfully forward. The most significant disclosures were a letter of intent with an unnamed industry partner to co-develop a B-TRAN-enabled intelligent solid-state circuit breaker (SSCB) prototype for evaluation by a U.S. hyperscaler running NVIDIA's Rubin Ultra 800-volt DC AI data center architecture, and a greater than 50% expansion in the company's sales opportunity funnel to over $300 million since the February investor update. Neither of these was previously known to the market.
The Hyperscaler LOI Is the Lead Item
The letter of intent with an undisclosed industry partner targeting a named U.S. hyperscaler for evaluation of B-TRAN in an NVIDIA Rubin Ultra rack environment is the single most commercially credible milestone in the company's history. Prototype delivery is targeted for the end of Q4 2026, and management says a purchase order for prototype units is expected to accompany the engagement. Critically, CEO David Somo indicated the prototype is also planned to be offered to additional U.S. hyperscalers and AI data center operators adopting comparable 800-volt DC architectures, meaning this is not a one-customer experiment but a potential beachhead into the broader hyperscaler ecosystem.
The strategic logic here is tightly tied to NVIDIA's platform roadmap. Somo was explicit: "Legacy 400-volt AC-based data centers were designed for rack power levels of 10 to 50 kilowatts. Next-generation AI workloads are pushing those requirements to 100 to 150 kilowatts in the near term and eventually beyond 500 kilowatts, with some industry planning scenarios targeting 1 megawatt per rack by 2030." The transition to 800-volt DC architectures creates a circuit protection problem that mechanical breakers cannot solve and that silicon carbide MOSFETs handle poorly in bidirectional, high-current applications—precisely where B-TRAN is designed to compete.
Funnel Composition and the 50% Jump
The sales opportunity funnel now stands at over $300 million, up from approximately $200 million at the end of February. Management disclosed for the first time that the funnel is roughly split 50-50 between automotive applications and the combined data center and general industrial segment. The automotive half spans EV contactors, battery disconnect units, and engagements with both automotive OEMs including Stellantis and Tier 1 suppliers. The non-automotive half now includes the company's first solid-state transformer, or SSP, opportunities following Somo's second Asia trip of the year, engagements with several new multinational SSCB customers, and the hyperscaler-related activity described above.
Somo acknowledged the gap between funnel and revenue directly: "While growing funnel is encouraging, converting it into design wins, production orders and revenue is the top priority." The company added that it met with approximately 30 companies at the Advanced Power Electronics Conference in March, some of which have since entered the qualified funnel. Somo personally chairs biweekly funnel review meetings with the sales team, a discipline he described as meaningfully more rigorous than at the start of 2026.
Asia Customer Expands to Three Active Projects
The lead Asia customer, whose identity remains undisclosed, has now spawned three simultaneous development tracks. The original low-current SSCB for 800-volt AI data centers and energy grid applications remains on track for prototype units in Q4 2026, with initial low-volume sales orders expected to support the prototype build. Two new projects were initiated during Somo's most recent Asia trip: a medium-current SSCB targeting 800-volt DC data centers, energy storage, EV charging and industrial microgrids, and a separate low-current SSCB for smart industrial buildings. The breadth of the engagement with a single customer across this many application verticals is a positive sign for eventual design-win depth.
Stellantis on Track, Automotive Qualification This Summer
Initial next-generation B-TRAN custom package samples and development kits have been delivered to Stellantis for EV application evaluation, and management confirmed remaining deliverables under the existing purchase order are on track to be completed by mid-2026. Automotive qualification and reliability testing—a non-negotiable prerequisite for any vehicle deployment—is also expected to be completed this summer. Somo was precise about why this matters: customers can proceed with product development while qualification is underway, "but to put the products into vehicles to begin their field testing on the road and then to release in series models, they need to have the components automotive qualified." Completion of this testing would remove a meaningful gating factor for automotive revenue conversion.
B-TRAN vs. Silicon Carbide: The Competitive Case
During Q&A, Somo addressed a direct comparison to Infineon's CoolSiC JFET—a product that represents the most credible competitive threat from incumbent silicon carbide suppliers. His response crystallized the competitive differentiation: JFETs, like silicon carbide MOSFETs, require a back-to-back configuration of at least two devices to conduct and block in both directions, whereas B-TRAN does this natively with a single device. Beyond component count, Somo noted that customers evaluating silicon carbide MOSFETs for SSCB applications cited difficulty sharing current and controlling turn-on and turn-off in medium and high current scenarios—areas where customer feedback identifies B-TRAN as superior. Reduced component count translates directly into smaller system size, higher power density, and lower total system cost, the three metrics that dominate procurement decisions in data center power infrastructure.
IP Moat and Manufacturing Readiness
The company's issued B-TRAN patent count reached 103, with 50 issued outside the United States across North America, China, Taiwan, Japan, South Korea, India and Europe. Somo flagged an additional layer of protection that patents alone do not provide: the double-sided wafer process flow used to fabricate B-TRAN devices is treated as a trade secret, the identity of fabrication partners is not disclosed, and the company works exclusively with foundries in countries with a history of respecting intellectual property rights. CFO Timothy Burns confirmed that manufacturing relationships with multiple foundries and packaging firms are in place and capable of handling the expected sales ramp over the next several years, with a cost reduction roadmap embedded in the longer-term plan.
Financials: Cash Runway Adequate, Burn Ticking Up
Q1 2026 cash burn from operating and investing activities was $2.3 million, below guidance of $2.6 million to $2.8 million and roughly in line with recent quarters. The company ended March with $16.4 million in cash, no debt, and a clean capital structure. Full-year 2026 cash burn guidance is $10 million to $10.5 million, up from $9.6 million in 2025, driven primarily by new hires in sales and engineering. At the guided burn rate, the company has approximately six quarters of runway without additional capital—a meaningful but not comfortable buffer for a pre-revenue company. Net loss widened to $3.6 million in Q1 2026 from $2.7 million a year ago, with operating expenses rising to $3.7 million from $2.8 million, the increase driven by higher stock-based compensation tied to the CEO transition and new inducement grants. Operating expenses are expected to increase modestly in coming quarters.
Management noted that near-term catalysts over the next three to six months will center on commercial announcements: additional customer agreements, initial purchase orders, continued Stellantis progress, and automotive qualification completion. The revenue timeline was not updated from the February business call, and Burns was deliberate in saying only that there is "no change in our expectation"—leaving the revenue inflection point unspecified but implying it remains a 2026 to 2027 phenomenon contingent on prototype evaluations converting to production orders.
Strategic investment discussions with potential partners who are also prospective customers continue. Somo declined to provide specifics but framed the ideal outcome as an investment that provides both capital validation and commercial validation simultaneously—a structure that would signal product-market fit more convincingly than any press release.
Ideal Power Inc. Deep Dive
Business Model and Monetization Strategy
Ideal Power operates as a fabless semiconductor designer, aiming to commercialize its proprietary bidirectional bipolar junction transistor technology, branded as B-TRAN. The company monetizes its intellectual property by selling discrete devices and multi-die packaged power modules directly to original equipment manufacturers and tier-1 industrial suppliers. Because the company utilizes a fabless model, it outsources capital-intensive wafer manufacturing and packaging to third-party foundries. This structure keeps fixed costs relatively low and theoretically allows the firm to scale production rapidly once commercial validation is achieved. However, the company remains in a highly speculative, pre-revenue phase of its corporate lifecycle. In Q1 2026, Ideal Power recorded $0 in commercial revenue, reflecting a business still fully entrenched in product qualification, prototype delivery, and customer design-in cycles. The ultimate financial viability of the model hinges on translating highly successful bench tests into recurring volume production orders over the coming 12 to 24 months.
End Customers, Key Partners, and Supply Chain
Ideal Power targets a customer base spanning multiple verticals heavily reliant on power conversion: automotive, data center infrastructure, and renewable energy. A flagship engagement involves a multi-phase development program with Stellantis to integrate B-TRAN into electric vehicle contactors and, eventually, drivetrain inverters. In the industrial and data center verticals, Ideal Power recently signed a strategic cooperation agreement with Lazzen, a global circuit protection manufacturer, and secured a letter of intent to co-develop a prototype for a major United States hyperscaler. This latter project specifically targets the NVIDIA Rubin Ultra 800V DC artificial intelligence data center power architecture. On the supply side, the company's fabless operations rely on external foundry partners to manufacture its double-sided wafers. Management strictly guards the identities of these fabrication partners as a core trade secret. By utilizing foundries located exclusively in jurisdictions with robust intellectual property enforcement, the company attempts to mitigate the risk of technological replication. The reliance on external foundries introduces standard supply chain vulnerabilities but allows the firm to sidestep the multi-billion-dollar capital expenditures required to build captive semiconductor fabrication facilities.
Competitive Landscape and Market Dynamics
The power semiconductor industry is dominated by deeply entrenched incumbents such as Infineon, STMicroelectronics, onsemi, and Wolfspeed. These firms command the vast majority of market share through their legacy silicon integrated-gate bipolar transistors and metal-oxide-semiconductor field-effect transistors. Recently, the industry has aggressively pivoted toward wide-bandgap materials, specifically silicon carbide and gallium nitride, to meet the high-efficiency demands of electric vehicles and fast chargers. Ideal Power currently commands 0% market share in this arena. The company operates as an unproven micro-cap challenger attempting to disrupt a conservative hardware industry that prioritizes historical reliability, supply chain security, and massive economies of scale. While new entrants in the wide-bandgap space focus intensely on scaling silicon carbide crystal growth and wafering to drive down costs, Ideal Power sidesteps the raw materials race by fundamentally redesigning the switch topology itself. B-TRAN currently utilizes standard silicon wafers rather than expensive and defect-prone silicon carbide substrates, yet aims to achieve equal or superior performance in specific bidirectional applications. If successful, Ideal Power could carve out a highly lucrative niche within the broader power semiconductor market without engaging in a direct capital expenditure war with legacy incumbents.
The B-TRAN Competitive Advantage
Ideal Power's fundamental competitive advantage lies in the unique architecture of the B-TRAN device. Traditional power converters utilize unidirectional switches, meaning they only control the flow of electricity in 1 direction. To achieve bidirectionality, which is an increasingly common requirement in vehicle-to-grid charging, solid-state circuit breakers, and matrix converters, engineers must traditionally wire 4 separate discrete components: 2 conventional switches and 2 diodes. B-TRAN is engineered as a strictly symmetrical, double-sided device that achieves bidirectionality in 1 monolithic piece of silicon. This 4-to-1 component reduction dramatically simplifies thermal management, shrinks the physical footprint of the system, and significantly lowers overall system cost. Furthermore, B-TRAN yields superior electrical performance profiles. The technology exhibits ultra-low forward voltage drops and conduction losses that are a fraction of those seen in conventional switch setups. This architectural moat is protected by an intellectual property portfolio comprising 103 issued patents globally, with 50 issued outside of the United States. Beyond the patents, the actual fabrication of a double-sided monolithic vertical device requires complex process recipes that the company retains as highly confidential trade secrets, creating an imposing barrier to entry for legacy semiconductor firms attempting to reverse-engineer the switch.
Opportunities and Industry Threats
The most immediate opportunity for Ideal Power is the structural shift toward high-voltage DC power architectures. Legacy AC grids and traditional 480V data center setups are severely strained by the extreme power density requirements of modern artificial intelligence infrastructure. As data centers migrate to 800V DC systems to support next-generation artificial intelligence accelerators, the need for solid-state circuit breakers capable of microsecond fault-response times becomes critical. Traditional mechanical breakers are too slow, and standard silicon carbide-based solid-state breakers suffer from unacceptably high conduction losses. B-TRAN's inherent bidirectionality and low conduction losses perfectly position the technology to capture early market share in this high-growth niche. However, the threats facing the company are existential. The semiconductor industry is notoriously capital intensive and dominated by long design-in cycles that can stretch for years. Legacy automakers and hyperscalers are historically reluctant to single-source mission-critical power components from a pre-revenue supplier. Should a major incumbent successfully develop a competing bidirectional topology or sufficiently drive down the cost of highly efficient silicon carbide modules, Ideal Power's technological advantage could be neutralized. Additionally, scaling novel double-sided wafer manufacturing introduces significant execution risks. Even with qualified foundry partners, achieving automotive-grade reliability and acceptable high-volume manufacturing yields remains an unproven hurdle.
Future Growth Drivers and Emerging Technologies
To transition from a research boutique to a commercial enterprise, Ideal Power developed packaged modules such as the SymCool and the SymCool IQ. The SymCool IQ integrates the bare B-TRAN die with an intelligent driver optimized for bidirectional operation, providing essential protections like overcurrent and undervoltage sensing. By packaging the technology into standard 62 mm industrial formats, the company drastically lowers the integration friction for original equipment manufacturers. Looking further down the product roadmap, the company is actively exploring the integration of B-TRAN into solid-state transformers. The modernization of the electrical grid requires advanced transformers capable of dynamically managing multi-directional power flows from distributed renewable generation and battery storage. Should B-TRAN enable a viable, cost-effective solid-state transformer, it would unlock a massive adjacent total addressable market beyond automotive and data center circuit protection.
Management Track Record
The installation of David Somo as Chief Executive Officer signaled a deliberate strategic pivot from foundational research and development to aggressive commercialization. Under Somo's tenure, the company expanded its global commercial infrastructure, actively recruiting sales and application engineering leadership in key Asian and European markets to interface directly with multinational customers. Financially, management has demonstrated clinical discipline in navigating a protracted pre-revenue phase. The company consistently operates with a lean cost structure, notably posting a Q1 2026 cash burn of $2.3 million, which came in below its anticipated guidance range. Furthermore, management proactively secured the balance sheet by pricing a $30 million direct equity offering in May 2026. While the resulting equity dilution reflects the harsh reality of funding hardware innovation, this capital infusion ensures the necessary liquidity to fund custom development programs, support initial production ramps, and sustain operations through critical upcoming commercial inflection points.
The Scorecard
Ideal Power represents a binary outcome within the global power semiconductor space. Its B-TRAN technology mathematically and empirically solves severe efficiency and packaging bottlenecks inherent to bidirectional power conversion. By replacing 4 conventional components with 1 double-sided silicon switch, the firm offers a highly compelling value proposition to industries wrestling with the extreme power density demands of artificial intelligence data centers and modern electric vehicle architectures. The recent strategic engagements with Stellantis, major United States hyperscalers, and global circuit protection original equipment manufacturers validate the commercial appetite for a structural alternative to legacy components and expensive silicon carbide solutions.
Despite these distinct technological advantages, the thesis is constrained by the brutal realities of hardware commercialization. The company remains pre-revenue and competes in a sector dominated by entrenched global incumbents that possess vastly superior resources and established supply chain hegemony. Ultimate success hinges entirely on management's ability to seamlessly transition from producing highly effective laboratory prototypes to executing flawless high-volume commercial manufacturing via secretive third-party foundries. Consequently, the firm operates as a high-risk call option on the broader electrification and artificial intelligence data center infrastructure build-out, requiring patient capital as design wins slowly translate into recurring commercial revenue.