A Discussion on Ultrahigh Efficiency and Ultrahigh Power Density DC–DC Converter Technologies

Abstract

This paper discusses the challenges in achieving ultra-high efficiency, defined here as higher than 99% efficiency, as well as high power density of more than 2kW/in <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> for DC-DC converters. For simplicity, our discussion focuses on 48V to 12V converters aimed primarily at data center applications, but the concepts discussed are broadly applicable to DC-DC conversion at a wide range of voltage levels. This paper will present a review of the fundamental sources of losses in DC-DC converters and how to minimize them, as well as a more in-depth look at some of the most efficient and dense topologies presented in literature thus far. Based on this analysis and review the key concepts that enable DC-DC converters to achieve higher than 99% efficiency at a power density of more than 2kW/in <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> will be summarized which include: easily paralleled “modular” designs to reduce conduction loss, multi-level structures that reduce individual component voltage stresses, utilizing lower switching frequencies to reduce switching and quiescent losses, operating with full duty ratio to ensure maximum utilization of the power components, and utilizing novel circuit topologies that nearly eliminate bulky, lossy magnetic components when compared with conventional topologies.