MS Thesis Defense: Muhammad Hassan Kiani

Monday, July 16, 2018, 10:30am

Rm. 100 (Spanos Auditorium), Cummings Hall

“Modelling, Optimization, and Comparison of Hybrid Resonant Switched Capacitor DC-DC Converters”

Abstract

Recent decades have seen rapid growth in portable-embedded computing and communications, which in turn has accelerated the need for high performance power management and delivery. This translates to a demand for high power density, low cost, small size, and high efficiency DC-DC converters. Traditionally magnetics based designs (like buck or boost) have dominated the DC-DC conversion realm. However, these architectures have a tradeoff between size of passives and efficiency. Hybrid-resonant switched capacitor (ReSC) DC-DC converters have shown great potential for applications that require both high density and efficiency. These approaches are known to benefit from the favorable active-device-utilization properties of switched capacitor (SC) converters and the relatively high energy density of capacitors compared to inductors, while mitigating the charge sharing losses inherent in SC designs. Compared to buck or boost DC-DC converters the hybrid approach promises to reduce the size of magnetic components by reducing the volt-second or flux linkage requirements.

While hybrid-resonant SC converters have shown a lot of promise, it has thus far been difficult to assess which topologies are most favorable given constraints on size (area or volume), energy storage, or power loss. This work presents a comprehensive study to model the losses in hybrid-resonant switched capacitor (ReSC) DC-DC converters. Both the losses in the active semiconductor devices and magnetic components are considered. The formulation presented in this work allows comparison between different converter architectures across conversion ratios given some common constraints. Optimization techniques are presented to minimize the losses in the individual components and the overall converter. Finally, given the drive for miniaturization of passive components, a technique is developed that leads to optimal allocation of size between flying capacitors and inductor(s) with a constraint on total size of passives.

Thesis Committee

For more information, contact Daryl Laware at daryl.a.laware@dartmouth.edu.