STMicroelectronics Ampere SiC Power Modules for EV 2026

STMicroelectronics and Ampere Forge Ahead with SiC Power Modules for EVs, Aiming for 2026 Commercialization

STMicroelectronics and Ampere (a distinct EV powertrain design company, not the Arm-based server designer formerly under Oracle) are reportedly collaborating to develop SiC power modules for next-generation EV traction inverters, targeting commercialization in 2026. While the announcement is straightforward, it reflects the industry consolidation driven by the "full-scale adoption phase of SiC."

Why is this significant? Traction inverters for EVs represent the largest demand source for SiC devices. Securing adoption here will directly translate into economies of scale, driving down device costs. For STMicroelectronics, this partnership is not only a platform to demonstrate its technological superiority but also a crucial step in securing order backlog amidst the intensifying SiC supply competition expected over the next two to three years.

The Significance of Choosing "Modules"

The decision to opt for power modules rather than discrete components offers insights from both design and manufacturing perspectives.

Power modules integrate SiC chips, gate drivers, heat sinks, and encapsulation resin into a single product, significantly reducing the workload for system designers. EV powertrains face stringent demands for power density, where cooling design, layout optimization, and EMC countermeasures critically influence inverter performance. By adopting pre-verified modules instead of assembling from individual chips, vehicle manufacturers can shorten their development timelines.

Conversely, modules exhibit higher technological dependence on suppliers compared to discrete components. Encapsulation structures, thermal designs, and internal wiring specifications tend to be vendor-specific, making changes costly once a decision is made. From a procurement standpoint, this presents a trade-off between long-term sourcing stability and vendor lock-in.

STMicroelectronics' proposal in module form suggests that Ampere's requirements extend beyond "chip performance" to encompass "ease of system integration." This direction aligns with the specifications sought by other Tier 1 suppliers and ODMs, reinforcing the view that the SiC module market will expand at a pace exceeding that of the discrete market.

The Unseen Hurdle of "Short-Circuit Withstand Time" Supporting SiC MOSFETs

When using SiC MOSFETs in EV traction inverters, short-circuit withstand time (SCWT) is as crucial a design consideration as efficiency and voltage rating. It indicates the duration a device can withstand a load short circuit before failure, providing a "grace period" for protective circuits to operate.

This is where SiC devices present a challenge. Due to their smaller die size and higher current density, they experience faster temperature rises compared to Si devices, necessitating shorter operating times for protective circuits. In other words, the overall response speed of the protection system is directly linked to device reliability.

The widely used protection method is DESAT (desaturation) detection. This mechanism monitors the drain-source voltage (VDS) in the on-state and turns off the gate upon detecting overcurrent. It is integrated into major gate driver ICs from companies like Infineon and Texas Instruments. Key design parameters include the DESAT trigger threshold (VDESAT), DESAT current (IDESAT), and blanking time.

Short-circuit withstand time varies with operating conditions. It tends to become stricter with higher drain-source voltage and lower junction temperature. For Microchip's 700V/1200V devices, the typical value stated in the datasheet is 3μs, serving as a benchmark for protective circuit response speed design.

Ron vs. Short-Circuit Withstand Time: Where the Difference Lies

When comparing the device performance of SiC MOSFETs, on-resistance (Ron) and short-circuit withstand time are in a trade-off relationship. Reducing Ron requires structural changes such as widening the channel or increasing current density, which in turn increases heat generation during short circuits.

Manufacturers are addressing this challenge through structural improvements. Mitsubishi Electric has introduced a p-type protection layer in its trench SiC-MOSFETs, significantly enhancing short-circuit withstand time. Rohm aims to achieve both low RonA and high short-circuit withstand time with its proprietary device structure in its 4th-generation SiC MOSFETs. STMicroelectronics also claims to possess its unique TRIACMTM structure, and the performance level achieved in this EV module development will be a focal point of evaluation.

Major players are each tackling the Ron vs. SCWT trade-off with their unique design philosophies. The positioning of STMicroelectronics' EV modules will ultimately be determined by comparing datasheets and operating conditions.

Towards 2026 Commercialization: Where Will the Competitive Edge Lie?

The competition in SiC power modules for EVs is not limited to STMicroelectronics. Infineon CoolSiC, onsemi EliteSiC, Rohm, and Mitsubishi Electric are all deploying products in the same market, and evaluation will proceed along three axes: supply capacity, quality track record, and support systems.

onsemi offers a broad portfolio, including SiC MOSFETs, SiC diodes, and SiC modules from 650V to 1700V, and is reported to have secured adoption in multiple vehicle projects. STMicroelectronics is leveraging its strength in vertical integration of SiC wafers (in-house wafer manufacturing) as a strategy to emphasize supply stability.

Voltage range alone does not determine superiority. What is truly critical for EV applications is compatibility with 800V architectures (primarily 750V-1200V devices), the thermal resistance of modules and flexibility in cooling design, and production yields and supply responsibility during mass production ramp-up. The 2026 commercialization timeline, when calculated backward from vehicle development cycles, implies that device finalization and evaluation must be completed by this year. From both design and procurement perspectives, this is effectively the decision window for supplier selection.

What to Watch for with the "2026" Timeline

The moves by STMicroelectronics and Ampere can be interpreted as evidence that the EV SiC module market is transitioning from the "technology evaluation phase" to the "mass production procurement phase."

Technically, the initial points to confirm are the module's SCWT specifications and their alignment with protective circuits. As mentioned, SiC demands stricter protective response speeds than Si, and the overall system reliability is determined by the combination with the chosen gate driver IC. Carefully reviewing datasheet conditions—drain voltage, gate voltage, and temperature—will provide crucial information to prevent post-production issues.

From a business and procurement standpoint, the focus will be on how STMicroelectronics' vertical integration strategy contributes to supply stability. The procurement risk for SiC wafers is directly linked to a device manufacturer's in-house wafer ratio; suppliers with high external procurement reliance face higher risks during supply shortages. While the extent to which the partnership with Ampere constitutes a "long-term agreement with mass production commitment" is undisclosed, in similar cases, the presence or absence of a Long-Term Agreement (LTA) often serves as a de facto dividing line for procurement risk.

The performance competition in SiC MOSFETs is shifting from "which specification is superior" to "which manufacturer can demonstrate reliability in mass production." Whether the 2026 commercialization by STMicroelectronics and Ampere becomes one such demonstration depends on the mass production verification phase over the next one to two years.