{"version":1,"type":"story","url":"https://digestai.news/story/infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers","json":"https://digestai.news/story/infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers.json","markdown":"https://digestai.news/story/infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers.md","slug":"infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers","headline":"Infineon and NVIDIA advance 800 VDC power systems for AI data centers","summary":"Infineon Technologies and NVIDIA have announced an expansion of their collaboration on 800 Volt direct current (VDC) power architectures for AI data centers, timed with NVIDIA's GTC Berlin conference. The focus is on high-voltage intermediate bus converters (IBC), a critical stage in power delivery that steps down voltage for AI accelerator cards. Infineon is positioning its gallium nitride (GaN) technology as the preferred solution for this stage, citing its ability to switch at frequencies up to 1 MHz with lower losses than silicon.\n\nThis work builds on joint efforts announced in May and October 2025, which focused on enabling hot-swappable 800 VDC server racks. The new system solutions include GaN power switches, PSoC control C3, and EiceDRIVER product families. These components support conversion ratios from 800 V to 50 V, 12 V, and 6 V, allowing data center operators to optimize rack architectures for higher power density and efficiency. The partnership also covers power path protection solutions, including hot-swap controllers, to ensure safety and serviceability in high-density compute environments.\n\nJochen Hanebeck, CEO of Infineon, stated that semiconductors are central to solving the power infrastructure challenges posed by AI. The collaboration aims to address the extreme power draw of AI workloads, which are pushing rack power levels toward the megawatt range. By reducing distribution losses and shrinking magnetic components, the GaN-based systems help manage thermal limits and board area constraints in modern data centers.","keyPoints":["Infineon and NVIDIA are advancing 800 VDC power architectures for AI data centers, focusing on intermediate bus converters.","Gallium nitride technology enables switching frequencies up to 1 MHz, improving power density and efficiency over silicon.","The collaboration includes full system solutions with PSoC control C3 and EiceDRIVER, supporting 800 V to 50 V, 12 V, and 6 V conversions."],"whyItMatters":"This partnership addresses the critical power infrastructure bottleneck for AI scaling. By enabling higher power density and efficiency in 800 VDC racks, it supports the deployment of denser, more energy-efficient AI data centers, which is essential for handling increasing computational demands.","category":{"slug":"hardware","name":"Hardware & Compute","url":"https://digestai.news/category/hardware"},"entities":{"companies":["Infineon Technologies AG","NVIDIA"],"models":[],"people":["Jochen Hanebeck"]},"firstPublishedAt":"2026-10-21T05:29:00Z","updatedAt":"2026-10-21T05:29:00Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"circuitdigest.com","title":"Infineon collaborates with NVIDIA on gallium nitride-based power conversion systems for 800 VDC AI data centers","url":"https://circuitdigest.com/news/infineon-collaborates-with-nvidia-on-gallium-nitride-based-power-conversion-systems-for-800-vdc-ai-data-centers","publishedAt":"2026-10-21T05:29:00Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Infineon and NVIDIA advance 800 VDC power systems for AI data centers\", 21 October 2026, https://digestai.news/story/infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers","publisher":"Digest AI","title":"Infineon and NVIDIA advance 800 VDC power systems for AI data centers","datePublished":"2026-10-21T05:29:00Z","url":"https://digestai.news/story/infineon-and-nvidia-advance-800-vdc-power-systems-for-ai-data-centers"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}