Japanese Power Semiconductor Trio's Integration Progress

Is the "Three-Company Consolidation" Truly Underway? The Current Status of Rohm, TDK, and Toshiba's Alliance

The "consolidation of three Japanese power semiconductor companies" has been repeatedly reported since 2025. Garnering attention as a state-backed reorganization supported by the Ministry of Economy, Trade and Industry (METI), the contours of this integration remain ambiguous when examining the actions of the companies involved. What has been decided, and what remains undecided? This article aims to clarify the current situation.

Based on news reports, the framework appears to be the preparation for the establishment of a power semiconductor joint venture between Rohm and TDK, with discussions underway for the integration of Toshiba Device & Storage's business into this entity. However, as detailed information regarding this integration project is not available in the fact cards provided at the time of writing, this analysis will be based on publicly available information.

The Reasons for the Three Companies Joining Forces Go Beyond "SiC"

In the world of power semiconductors, a rapid transition to SiC (silicon carbide) is currently underway. SiC MOSFETs offer lower switching losses compared to silicon (Si) IGBTs, making them a crucial material for enhancing the efficiency of electric vehicle inverters and industrial power supplies, prompting accelerated investment from various companies. However, the situation regarding the transition to SiC is not monolithic across Japanese companies.

Rohm is one of Japan's leading suppliers of vertical SiC devices, mass-producing its fourth-generation SiC MOSFETs and positioning the balance between low on-resistance and high short-circuit withstand capability as a key differentiator. TDK, while strong in passive components and sensors, has synergies in power modules with power semiconductors. Toshiba Device & Storage offers a product line that combines Si and SiC, and has continued to deepen its Si technology, announcing its triple-gate IGBT in 2021, which achieved up to 40.5% loss reduction.

While the strengths of the three companies may appear to overlap, they are in fact complementary. By combining Rohm's SiC material technology and manufacturing expertise with TDK's and Toshiba's customer touchpoints and power system knowledge, a structure emerges that enables integrated proposals from individual devices to power system modules.

This diagram illustrates that the three companies' relationship aims not at "eliminating competition" but at "rebuilding the value chain." However, complementarity does not guarantee a smooth integration.

Short-Circuit Withstand Capability: A Subtle But Crucial Point

The "short-circuit withstand capability" of SiC MOSFETs emerges as a technical backdrop to this integration. Short-circuit withstand time (SCWT) refers to the duration a device can withstand a short circuit before failure, serving as a critical indicator for the grace period before a protection circuit is triggered.

Due to their smaller die size and higher current density compared to silicon, SiC devices experience faster temperature rises during a short circuit. For Microchip's SiC MOSFETs (700V/1200V rated), the datasheet lists a typical value of 3μs, which serves as a benchmark for designing protection circuit timelines.

The common countermeasure to this "rapid temperature rise" is the DESAT (desaturation) protection function. This mechanism constantly monitors the drain-source voltage (VDS) in the on-state and turns off the gate the moment overcurrent is detected, playing a role in protecting the system within the constraints of short-circuit withstand capability.

Crucially, short-circuit withstand capability is not solely a matter of the individual device. It varies with operating conditions such as applied voltage, gate voltage, and junction temperature, and greater margin is gained by easing these conditions. In other words, unless the device, protection circuit, and thermal design are optimized in an integrated manner, the rated withstand capability cannot be fully utilized.

Viewed from the perspective of this trinity of device, protection circuit, and thermal design, the significance of the three-company integration changes. If a single group can cover everything from device manufacturing to system integration, it is possible to shorten the optimization loop.

Competitors Are Already Pursuing "Full-Stack" Strategies

The momentum for the three-company integration has been fueled by the moves of global competitors. Infineon offers solutions that include gate drivers centered around CoolSiC, while onsemi boasts a full portfolio of SiC MOSFETs, diodes, and modules from 650V to 1700V. STMicroelectronics has entered into an agreement with Ampere for the supply of SiC modules for EVs, commencing in 2026.

The characteristic of overseas players is a clear shift from "selling devices" to "selling power conversion solutions." This strategy aims to reduce the customer's system design workload while increasing the stickiness of their own products (switching costs), creating a structure that is less susceptible to price competition.

When the integration of the three Japanese companies is discussed in this context, the focus is not solely on device competitiveness. The extent to which they can provide added value "outside the device," such as system design support, integration with gate drivers, and thermal design expertise, becomes a key factor in adoption decisions. Even if Rohm possesses manufacturing technology that places it among the top global market share holders, when it comes to competitiveness as a system solution, the combination with TDK's power module expertise and Toshiba's customer touchpoints is a logical complementary move.

Conditions for the Integration Not to Become a "Pie in the Sky"

On the other hand, even if integration discussions materialize, actual technical integration will take time. Integrating design processes, quality assurance systems, and ERP systems will take several years. Furthermore, Toshiba Device & Storage faces the variable of its parent company's strategic direction.

From a technical standpoint, fundamental challenges like the trade-off between short-circuit withstand capability and on-resistance will not be solved by organizational integration. Individual structural improvements, such as Mitsubishi Electric's introduction of a p-type protective layer in its trench-type SiC-MOSFET to enhance withstand capability, will remain at the core of product competitiveness. Integration can create a "vehicle for multiplying comparative advantages," but the content within that vehicle must be built through each company's research and development.

Several aspects can be considered as practical indicators for judging the progress of the integration. The announcement of a joint product roadmap, the transfer of product lines to the joint venture, and the scope of guaranteed operation for combinations of gate drivers and SiC MOSFETs—when such concrete actions emerge, it is useful to view them as turning points where the integration shifts from being an "organizational matter" to a "market matter."

What Will Change in Selection, Procurement, and Investment Decisions?

How will the progress of this integration concept impact practical decision-making?

In terms of design, there will be no significant changes in the short term. Rohm's SiC MOSFETs, Toshiba's IGBT and SiC products, and TDK's power modules continue to be supplied as independent products. In device selection, rather than the risk of part numbers changing or supply systems being altered due to integration, the enhancement of design support resulting from integration may offer benefits sooner.

In procurement, while the consolidation of points of contact after integration could lead to more efficient quoting and negotiation, the risk of alternative sourcing being concentrated within a single group needs to be considered. At present, integration is not complete, and it is a phase of observing the situation while maintaining existing supplier relationships.

From the perspective of investment and business development, understanding "why the government is supporting this at this particular time" is more valuable as a decision-making factor than the success or failure of the integration itself. If the trigger for integration is the recognition that Japanese suppliers' market share is relatively declining in the main battlegrounds for SiC—EVs, renewable energy, and industrial automation—then the direction of domestic investment in this field may continue regardless of the integration's outcome.

This graph illustrates that all major players cover voltage ratings up to 1700V, making the differentiation solely based on voltage range increasingly difficult. As the axis of competition shifts from "what voltage can be achieved" to "how much can customer costs be reduced through system integration," the significance of the three-company integration aiming for full-stack capability becomes apparent. In selecting SiC MOSFETs, a situation is emerging where composite judgments are required, not only on device specifications but also on compatibility with gate drivers, depth of support for short-circuit protection design, and supplier supply continuity. The progress of the three-company integration is a trend directly linked to "supplier strength and integrated proposal capability," and is worth continuing to track.