BYD Semiconductor Expands SiC Mass Production

BYD to Produce its Own SiC: Vertical Integration Reshapes the Electrification Supply Chain

In 2024, BYD Semiconductor, BYD's semiconductor subsidiary, announced plans to expand its mass production lines for 1200V SiC MOSFETs, signaling a clear intention to increase the proportion of in-house components for EV inverters. This is not merely about in-house production. The world's largest EV manufacturer is aiming to control upstream power semiconductor components.

SiC devices are key components that influence EV driving range and charging speed, and specialized manufacturers like Infineon, onsemi, and ROHM have dominated the market. BYD's move to enter this space through its own production could shake up the entire industry's supply chain structure, extending beyond procurement strategies.

The True Impact of "In-house Production"

BYD has a long history of developing and producing its own semiconductors. BYD Semiconductor was established in 2004, initially focusing on IGBT development. While their full-scale transition to SiC is a more recent development, they are reportedly already equipping EVs with their second-generation 1200V SiC MOSFETs.

It is important to clarify the scope of "in-house production." A fabless model, where a company designs devices in-house but outsources wafer manufacturing to external foundries, differs significantly from a fully vertically integrated model that covers everything from wafer growth to back-end processes. BYD's current situation appears to involve in-house design and some back-end processes, while relying on external suppliers for wafer substrates. Although not entirely self-sufficient, this internalization of key value-adding processes distinguishes BYD from other EV manufacturers.

While Tesla has fully adopted SiC, it relies on suppliers like onsemi for component procurement. Volkswagen and Stellantis also continue to source power semiconductors externally. In comparison, BYD's move stands out as "the most deeply involved case among major EV manufacturers in semiconductors."

Technical Challenges Specific to SiC and the Hurdles BYD Faces

SiC MOSFETs present several design difficulties distinct from silicon devices. One of these is the short-circuit withstand time (SCWT).

In applications like EV inverters, instantaneous overcurrent can occur due to arm short circuits or malfunctions. The time until the device fails under such conditions is the short-circuit withstand time, which dictates the margin for protective circuits to operate. A shorter SCWT imposes stricter requirements on the response speed of protective circuits.

SiC dies are smaller and have higher current densities, leading to faster temperature increases compared to silicon when the same power is applied. In other words, the time margin for protective circuits to operate is smaller. Microchip's 1200V product lists a typical value of 3μs in its datasheet, which serves as an industry benchmark.

There is a trade-off between short-circuit withstand time and on-resistance (Ron). Efforts to increase efficiency by lowering on-resistance tend to make devices more susceptible to short circuits, and vice versa. Specialized manufacturers are investing significant technical effort to overcome this trade-off. Mitsubishi Electric has improved short-circuit withstand time by introducing a p-type protective layer in trench SiC-MOSFETs, while ROHM's fourth-generation products achieve both low RonA and high short-circuit withstand capability through unique device structures. For BYD to achieve mass production precision, it will need to meet quality standards that can compete in these technical details.

"Conditionality" as a Differentiating Factor from Specialized Manufacturers

Short-circuit withstand time is not a fixed value. It varies depending on operating conditions such as drain applied voltage, gate applied voltage, and junction temperature, tending to increase as conditions become less severe. Furthermore, SiC MOSFETs exhibit a characteristic where on-resistance (RDSon) increases at high temperatures, limiting the saturation current, which in turn improves short-circuit robustness at high temperatures.

Understanding these condition dependencies is crucial for device evaluation. Specialized manufacturers have a track record of systematically disclosing these characteristics through application notes and reference designs. When BYD enters the phase of selling its own devices externally, or when evaluating them internally, the depth of such technical disclosure will be a deciding factor.

The DESAT (de-saturation) function is widely used in designing short-circuit protection circuits. This mechanism monitors the drain-source voltage (VDS) in the on-state and turns off the gate when overcurrent is detected, often incorporated into gate driver ICs. Setting parameters such as the DESAT trigger threshold (VDESAT), DESAT current (IDESAT), and short-circuit blanking time significantly impacts protection performance. BYD's ability to design and manage the entire system, including protective circuits, can be an advantage from this perspective.

Specialized Manufacturers' Reactions: Structural Change Rather Than Competition

How specialized manufacturers perceive BYD's move—whether as "competitor entry" or "customer defection"—varies. In the short term, an increase in BYD's in-house procurement ratio will reduce orders placed with external suppliers. However, whether BYD can cover its entire demand internally is another question, and quality stabilization may take time during the initial phase of mass production ramp-up. Some believe that external procurement will continue to supplement supply during this period.

From a broader perspective, BYD's move may lead other EV manufacturers to recognize that "an era of in-house semiconductor production is possible," potentially prompting similar actions. While automakers' moves into semiconductor design have been discussed in terms of their impact on Renesas and Infineon, the involvement with high-difficulty devices like SiC changes the dimension of the discussion.

Meanwhile, onsemi and Infineon have secured multiple long-term contracts for the EV market and continue to expand their manufacturing capacity. onsemi offers a broad lineup of SiC MOSFETs, SiC diodes, and SiC modules covering voltages from 650V to 1700V, employing a portfolio strategy to reduce dependence on specific OEMs.

This graph illustrates that BYD's current mass-produced products are concentrated in the 1200V range. The 1700V range remains exclusive to specialized manufacturers, indicating a gap for high-voltage systems and industrial applications. Whether BYD will venture into higher voltage products in the future will be a key indicator for changes in the competitive landscape.

Analyzing from Both Technical and Business Perspectives: What to Watch

As BYD's momentum continues, the question of "how far this change will propagate" is of interest to those in technical, procurement, and business development roles.

From both procurement and design perspectives, it is important to observe whether BYD will sell its in-house devices to third parties. Currently, it appears to prioritize its own EVs, but as production scales up and yield stabilizes, the option of selling surplus products or supplying other OEMs may emerge. If this becomes a reality, the market competitive landscape for SiC devices will differ significantly from the present.

From the standpoint of technical reliability, the disclosure of information such as short-circuit withstand capability, dependency on operating conditions, and long-term reliability data will be critical for external evaluation. Whether BYD's device data becomes easily comparable to the accumulated track records of specialized manufacturers will influence the possibility of adopting BYD devices.

From a business development and market analysis viewpoint, the impact of BYD's vertical integration on its cost structure cannot be ignored. The cost of SiC devices is heavily influenced by the price of wafer substrates, but increased in-house production could change the transparency of component costs, potentially affecting the price competitiveness of BYD vehicles. If the cost competition in the EV market and price formation in the power semiconductor market begin to align, it would have cross-industry repercussions.

BYD's move is occurring at a stage where SiC is transitioning from a "specialty component" to a "mass-produced component." The question of who will hold the lead in design and manufacturing at this inflection point is an essential perspective for forecasting the industry structure in the coming years.