Renesas Electronics Corporation announced that it is supporting efficient power conversion and distribution for the 800 Volt Direct Current power architecture announced by NVIDIA, helping fuel the next wave of smarter, faster AI infrastructure.
As GPU-driven AI workloads intensify and data center power consumption scales into multi-hundred megawatt territory, modern data centers must adopt power architectures that are both energy optimized and scalable. Wide bandgap semiconductors such as GaN FET switches are quickly emerging as a key solution thanks to their faster switching, lower energy losses, and superior thermal management. Moreover, GaN power devices will enable the development of 800V direct current buses within racks to significantly reduce distribution losses and the need for large bus bars, while still supporting reuse of 48V components via DC/DC step-down converters.
Renesas’ GaN based power solutions are especially suited for the task, supporting efficient and dense DC/DC power conversion with operating voltages of 48V to as high as 400V, with the option to stack up to 800V. Based on the LLC Direct Current Transformer (LLC DCX) topology, these converters achieve up to 98 percent efficiency. For the AC/DC front-end, Renesas uses bi-directional GaN switches to simplify rectifier designs and increase power density. Renesas REXFET MOSFETs, drivers and controllers complement the BOM of the new DC/DC converters. Visit renesas.com/power for more details.
“AI is transforming industries at an unprecedented pace, and the power infrastructure must evolve just as quickly to meet the explosive power demands,” said Zaher Baidas, Senior Vice President and General Manager of Power at Renesas. “Renesas is helping power the future of AI with high-density energy solutions built for scale, supported by our full portfolio of GaN FETs, MOSFETs, controllers and drivers. These innovations will deliver performance and efficiency, with the scalability required for future growth.”
Renesas has published a white paper that explores the topology of its devices supporting 800V power distribution in AI infrastructure.
