en.Wedoany.com Reported - Navitas Semiconductor demonstrated an 800V-to-6V DC-DC power board for the NVIDIA MGX platform at Computex 2026. Utilizing GaNFast technology, this product eliminates the traditional 48V intermediate bus converter stage within server trays, enhancing system efficiency, reliability, and space savings.
The power board employs 16 GaNFast FETs rated at 650V with an on-resistance of 11 milliohms, housed in the latest DFN8×8 dual-cooling package. Targeting a peak efficiency of 97.5% and operating at a frequency of 1 MHz, it achieves a power density of 2100 watts per cubic inch. Its ultra-slim profile, approximately 20% thinner than a mobile phone, enables close integration with GPU boards to optimize transient performance and improve power distribution efficiency.
Chris Allexandre, President and CEO of Navitas Semiconductor, stated that as AI workloads continue to expand and drive unprecedented computing demands, power delivery has become a critical challenge for next-generation gigawatt-scale AI smart factories. By collaborating with the NVIDIA MGX ecosystem, Navitas is delivering gallium nitride (GaN) and silicon carbide (SiC) power technologies that enable megawatt-class AI server racks with higher power density, smaller system footprints, and improved thermal performance, accelerating the transition to more efficient and scalable AI infrastructure.
Navitas Semiconductor offers a comprehensive portfolio of wide-bandgap power technologies, laying the foundation for next-generation AI factory infrastructure. Its GeneSiC silicon carbide solutions support efficient power delivery from the grid to computing racks, covering key applications such as solid-state transformers. Products include ultra-high-voltage 2300V and 3300V SiC power modules, as well as high-power three-phase power supply units based on 1200V SiC MOSFETs using the latest fifth-generation technology. Together, these technologies help data centers improve efficiency, power density, and system reliability.
Navitas' GaNFast technology enables high-frequency, high-efficiency DC-DC power conversion to support the rapidly growing power demands of AI GPUs. Leveraging GaN's superior switching performance, the solution operates at megahertz-level frequencies, delivering higher power density and faster transient response, allowing power to be efficiently transferred directly from the rack level to the GPU.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
