Is silicon carbide too expensive to use? Behind Tesla’s announcement of a “75% reduction in SiC usage”

Tesla claims to have figured out a way to reduce the amount of SiC devices used by 75% without compromising the car’s performance or efficiency.

At Tesla’s “investor day” event on Wednesday, a major event broke out, which can be said to shake the entire SiC upstream and downstream.
Colin Campbell, Tesla’s president for powertrain engineering, noted at the event that silicon carbide is an “amazing semiconductor,” but added that silicon carbide is expensive and “really hard” to scale, so reducing the use of 75% of silicon carbide in cars is “a big win for us” (or “crucial,” see original note below).

【Colin Campbell：”Silicon carbide is an amazing semiconductor, but it’s also expensive and it’s really hard to scale. So using less of it is a big win for us.” “Sound is blurry, or a big one for us”]
The stock prices of several SiC OEMs were immediately volatile, including Wolfspeed, ON Semiconductor, STMicroelectronics (ST’s SiC revenue in 2022 is about $700 million), etc.
Campbell also said that Tesla’s new powertrain will use motors that do not contain any rare earth metals.

Cost is key
The cost of the Model 3 has been reduced by 30% in 2022, but Tesla hopes to reduce the production cost of the next generation of cars by another 50%.

The cost of the drive unit is a key part of this.
Truss revealed that it will “reduce the cost of the drive unit by about $1,000” by optimizing SiC devices, batteries and other aspects. Among them, “we figured out a way to reduce the amount of devices used by 75% without compromising the performance or efficiency of the car”.
The new powertrain is also compatible with any battery chemish, which will give Tesla more flexibility in battery sourcing.
Because silicon carbide is not only expensive, but also difficult to scale wafer size, it is worried that the supply of SiC device capacity will be difficult to keep up with the pace of their car sales, so Tesla believes it is necessary to reduce the amount of silicon carbide. They made it clear that the use of SiC devices could be reduced by further optimizing the power module design.
However, Campbell did not say when the company’s next-generation powertrain will be ready for high-volume production and use in the company’s vehicles, nor did it specify how much it is currently spending on these transistors. Executives at the event did not disclose any details about the “next generation” Tesla.
Analysts at New Street Research speculate that “the inverter for the new drivetrain will have a hybrid architecture,” with a mix of silicon and silicon carbide transistors, two types of transistors working together to handle peak loads in Tesla vehicles, primarily during vehicle acceleration. “This hybrid architecture is only suitable for new platforms, i.e. low-cost, small, low-performance cars that will not be adopted by existing models (S, X, 3, Y) or Cybertruck.”
New Street doesn’t expect the low-priced next-generation Tesla vehicles to be “in mass production before 2025 or 2026.”
Industry feedback
Given the impact of Tesla’s move on upstream chip makers, Craig Irwin, an analyst who immediately has a Buy rating on Wolfspeed (WOLF), wrote in an investor note: “In our view, Tesla has a history of exaggeration, so it is likely that this claimed improvement in thermal extraction (packaging technology) has accelerated industry adoption.” ”
Compared to traditional silicon-based semiconductors, silicon carbide semiconductors can operate at higher voltages, temperatures, and frequencies. This makes them a better choice for electric vehicles, solar conversion, 5G wireless, aerospace, and other applications.
However, there is also a voice in the industry that whether it is switching to modules instead of single tubes, or structural processes to reduce the size of SiC dies, it cannot hide the fact that SiC is not cost-effective. The consumer goods attributes of automobiles are becoming more and more obvious, in addition to safety requirements can not be reduced, other performance manufacturers may think that enough. What do you think of the impact of Tesla’s move on the future of SiC?