Delta Debuts DC Power System for AI Data Centers
Delta Electronics unveiled a prefabricated solid state transformer based DC power system in a press release at Energy Taiwan 2026. The system combines solid state transformers, DC/DC converters and control equipment for AI data centers and high density computing facilities.
Delta says its solid state transformers reach efficiency of up to 98.5%. The company says the system can improve efficiency from the grid to chips by 4% to 5% and reduce on site deployment time by up to 50% through its modular design.
The system can include a high C rate battery that provides millisecond level support during rapid changes in AI computing loads. Delta also presented microgrid components including energy storage, intelligent power dispatch and solid oxide fuel cells, with trial production of its fuel cell solution planned by the end of 2026.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.
Also, consider following us on social media:
More from Data Centers
Oct 14 QuickLogic secures eFPGA contract for automotive and industrial SoC Oct 14 Solidion Says PEAK UPS Can Cut Battery Footprint by 30% Oct 14 LITEON Shows 800 VDC Power and Cooling Systems for NVIDIA Vera Rubin Oct 14 SmartSens Shows Automotive Vision Sensors at AutoSens Europe Oct 14 IDC Raises 2029 OCP Infrastructure Spending Forecast to $800 BillionSilicon Brief
Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Delta Develops Power and Cooling Infrastructure for NVIDIA DSX AI Factories
Schneider Electric Unveils Software Defined Switchgear for AI Data Centers
DIGITIMES Examines 800 VDC Power for AI Data Centers
Huawei Shows AI Energy Systems at IDEE 2026
Huawei Unveils Grid Interactive AI Data Center Solution
Daily AI Brief: the AI news that matters, in your inbox.