High-Performance Hybrid Switched-Capacitor Converters and Coupled Magnetics for Direct 48 V to Point-of-Load Vertical Power Delivery

Date: 17/03/2026

Time: 11:00 am

Presenter: Yicheng Zhu

Abstract: High-performance processors are the driving force behind modern data center computing platforms. As processor current demands surpass 1000 A, the large power distribution network (PDN) inherent to the conventional two-stage lateral power delivery (LPD) architecture leads to severe voltage drops and unacceptable conduction losses, making this “last inch” the bottleneck of data center power delivery systems. In pursuit of a more efficient and compact alternative to the existing solution, this talk will introduce a family of high-performance 48 V to point-of-load hybrid switched-capacitor (SC) voltage regulators, named the switching bus converter (SBC). Based on a single-stage vertical power delivery (VPD) architecture, the SBC significantly reduces the PDN size and power conversion losses. Moreover, benefiting from the hybrid SC approach, the SBC effectively leverages the greatly superior energy density of capacitors compared to inductors, as well as the better figure-of-merit of low-voltage switching devices over their high-voltage counterparts. Finally, this talk will present design-oriented modeling and multi-objective optimization techniques for coupled inductors, which are widely adopted in point-of-load applications to achieve reduced inductor volume and improved dynamic response.

Yicheng Zhu is an Assistant Professor and a Friends of Alec ECE Fellow in the Chandra Family Department of Electrical and Computer Engineering at The University of Texas at Austin. He received his B.Eng. and M.S. degrees in electrical engineering from Tsinghua University, Beijing, China, in 2017 and 2020, respectively, and his Ph.D. degree in electrical engineering and computer sciences from the University of California, Berkeley (UC Berkeley), CA, USA, in 2024. He was a Postdoctoral Scholar at UC Berkeley from 2024 to 2025. His research interests include circuit topologies, control techniques, analytical modeling, and high-performance implementations of power electronics for computing and energy systems. Dr. Zhu received the IEEE Power and Energy Society (PES) Outstanding Student Scholarship, the Outstanding Tsinghua Master’s Thesis Award, and the Berkeley Fellowship in 2020. He was the 2023 recipient of a Best Paper Award at the IEEE 24th Workshop on Control and Modeling for Power Electronics (COMPEL), a Best Paper Award at the Open Compute Project (OCP) Future Technologies Symposium, and the NVIDIA Graduate Fellowship, awarded annually to five Ph.D. students worldwide involved in research spanning all areas of computing innovation. In 2024, he received the Outstanding Graduate Student Instructor Award and the Teaching Effectiveness Award from UC Berkeley, as well as the Ross N. Tucker Memorial Award from the Department of Electrical Engineering and Computer Sciences at UC Berkeley for outstanding research. In 2025, he received a Best Presentation Award at the IEEE 40th Applied Power Electronics Conference (APEC) and the IEEE Power Electronics Society (PELS) Ph.D. Thesis Talk (P3 Talk) Award.