Enteligent has introduced PowerCore, an 800 VDC rack-to-chip DC/DC conversion stack aimed at AI data centers moving to higher-voltage power distribution. The PowerCore DC/DC architecture spans the rack, server, and GPU board, and is designed to take an 800 VDC rack supply down to the voltage rails used by GPUs.
PowerCore is built as a modular, three-stage conversion chain:
The RackCore module is a 100 kW rack power shelf that converts 800 VDC to a 50/48 VDC bus. Enteligent says this stage eliminates the need for supercapacitor and battery backup shelves, and the company ties that to more available rack space, lower cost, and higher reliability. Enteligent also claims the approach can enable up to 50% higher compute density.
The MaxCore module is server-mounted and converts 50/48 VDC to a regulated 7 VDC rail. Enteligent describes the design as a multi-phase, multi-path architecture intended to improve efficiency and simplify implementation.
The ChipCore module sits near the GPU as a compact, 16-phase stage converting the intermediate rail to 1.2/1.7 VDC for final point-of-load delivery.
For data center power engineers, the interesting part isn’t that Enteligent has another VRM story—it’s the explicit attempt to treat 800 VDC distribution, intermediate buses, and point-of-load conversion as one coordinated electrical design problem. At high rack power, conversion efficiency, transient response, thermal dissipation near accelerators, and EMI control all stack up fast, and a “just optimize each stage” approach can create ugly tradeoffs in conductor sizing, filtering, and board real estate.
Enteligent says the modules can be evaluated as a complete chain or integrated individually with compatible downstream power designs. “As AI systems move toward much higher rack power, every conversion stage has to work as part of one system,” said Sean Burke, CEO of Enteligent. “The PowerCore stack draws on our DC-to-DC experience in high-density power electronics and computing to give AI infrastructure and server OEMs a modular path from 800 volts at the rack to the GPU.”
Source: Enteligent
