DB HiTek achève la qualification de fiabilité du procédé de fabrication des MOSFET SiC de 8 pouces et 1 200 V

DB HiTek completed reliability qualification for its 1,200V 8-inch SiC MOSFET process, which it describes as the world's first end-to-end 8-inch SiC foundry platform, with volume production targeted for 2027. Its second-generation process achieves specific on-resistance of 2.5 mΩ·cm² or below, while a third-generation PDK due in November targets 2.3 mΩ·cm² or below and short-circuit withstand time of at least 2.5 μs. The PDKs are intended to shorten customer product-development cycles by more than one year, supporting customer acquisition in EV, renewable-energy and industrial power applications.

Analysis

The strategic read-through is less about near-term SiC demand and more about foundry market structure: an independent 200mm platform lowers the barrier for power-semiconductor designers that do not want to fund captive capacity. That could pressure the scarcity premium embedded in vertically integrated SiC suppliers—Wolfspeed (WOLF), onsemi (ON), STMicroelectronics (STM) and Infineon (IFNNY)—but only if DB HiTek converts PDK engagement into qualified customer designs and acceptable wafer yields. The more immediate beneficiary is likely the outsourced-design ecosystem, as lower prototyping friction expands the number of automotive, industrial and energy customers testing SiC architectures.

The key economic uncertainty is yield, not the stated device specifications. Moving from a qualified process to commercially repeatable 200mm SiC production requires defect-density control, substrate availability, cycle-time stability and automotive-grade customer qualification; any one of these can defer meaningful revenue beyond 2027. For WOLF, additional foundry supply is directionally negative for its long-dated capacity-utilization narrative, although DB HiTek's ramp could simultaneously validate 200mm SiC adoption and improve confidence in the broader addressable market.

Consensus may overstate the near-term competitive disruption. Automotive inverter design cycles and qualification windows are typically multi-year, so the first financial signal should be design wins and wafer commitments over the next 3-9 months, rather than immediate share loss at incumbents. A credible third-party 200mm alternative nonetheless raises the risk that future SiC pricing tracks manufacturing learning curves more rapidly, limiting gross-margin upside for merchant-device suppliers during 2028-29.

Key Decisions for Investors

• No immediate directional trade in DB HiTek absent a liquid identified listing, disclosed SiC capex, customer commitments and yield data; set an alert for binding wafer agreements or volume-production guidance, which would make the competitive signal investable.
• Maintain a 6-12 month relative-value watch: short WOLF versus long ON or STM if WOLF continues to price in rapid 200mm utilization while third-party foundry capacity expands. Thesis is most vulnerable if WOLF discloses contracted utilization, improving gross margin, or material substrate-cost reductions.
• For existing long ON/STM exposure, treat this as a 2028-29 margin-risk flag rather than a near-term sell catalyst; reduce if management begins guiding SiC pricing concessions or if EV/industrial SiC inventory digestion persists into 2027.
• Monitor SiC substrate suppliers and equipment exposure—Coherent (COHR), AXT (AXTI), and silicon-carbide processing tool vendors—for evidence that 200mm qualification translates into incremental substrate orders. Do not add solely on this announcement; purchase-order visibility and capex disclosures are required confirmation.