Renesas Electronics Corporation (TSE:6723) announced that it is supporting efficient power conversion and distribution for the 800 Volt Direct Current (800VDC) architecture introduced by NVIDIA, designed to accommodate next-generation AI workloads.
As data center power demands rise into multi-hundred-megawatt levels due to GPU-driven AI applications, operators are seeking architectures that are both energy-efficient and scalable. Wide bandgap semiconductors, including gallium-nitride (GaN) FET switches, are increasingly adopted for their faster switching, lower energy losses, and improved thermal performance. GaN devices also enable 800VDC rack designs that reduce distribution losses and the size of bus bars while allowing continued use of 48V components via DC/DC step-down conversion.
Renesas’ GaN-based solutions are designed for dense and efficient DC/DC conversion, supporting operating voltages from 48V up to 400V, with the option to stack to 800V. Using the LLC Direct Current Transformer (LLC DCX) topology, these converters can reach efficiencies of up to 98 percent. For AC/DC front-end applications, bi-directional GaN switches simplify rectifier design and improve power density. The company’s REXFET MOSFETs, drivers, and controllers complement these systems, streamlining design and optimizing performance for 800VDC architectures.
“AI is transforming industries rapidly, and power infrastructure must keep pace with growing demands,” said Zaher Baidas, Senior Vice President and General Manager of Power at Renesas. “Our portfolio of GaN FETs, MOSFETs, controllers, and drivers delivers high-density, scalable solutions for modern AI data centers, combining performance and efficiency.”
Renesas’ solutions aim to provide the scalable, energy-efficient power infrastructure required to support next-generation AI data centers and the adoption of 800VDC rack architectures.
