June/July 2021

26 APEC 2021 www.apec-conf.org Issue 3 2021 Power Electronics Europe www.power-mag.com discussed, including advanced magnetics design and alternatives to the use of conventional magnetic components. One approach is to use high-Q resonators, mechanical or electromagnetic, that have better performance than resonators built from discrete inductors and capacitors, and can be used in both power conversion and wireless power transfer. The future potential and progress to date on these technologies will be reviewed. Awarded and interesting conference papers On the following pages we have selected and summarized the most interesting papers for our readers. Contact details are provided for convenience. are being treated internationally and across the entire application ecosystem. The Present and Future of Magnetics and Other Power Passives has been discussed by Charles R. Sullivan, Professor of Engineering Thayer School of Engineering at Dartmouth. High switching frequencies in power electronics are motivated by reducing the size, cost and loss of magnetic components. With advances in semiconductor switch technology, higher frequencies are becoming practical, but the typical result is not the ultra-small ultra-efficient magnetics one might hope for: rather, magnetics have become the critical factor limiting power electronics performance. Ways to overcome these limitations will be High-Density Integrated Power Electronics SiC Building Block Modular, medium-voltage (MV) converters have grown in popularity due to their ability to be scaled to higher voltage and current ratings and reduce production and maintenance costs. The power electronics building block (PEBB) concept was proposed by the Office of Naval Research (ONR) in 1997 as a universal power converter for power systems on ships. Since then, many modular MV converters have been developed using silicon IGBTs and various topologies including as a 3-level NPC and single phase 5-level H- bridge. A half-bridge SiC module prototype is fabricated and tested to provideinsight into the substrate functionality and to aid in the refinement of the iPEBB concept. Narayanan Rajagopal, Virginia Polytechnic Institute & State University (nrajagopal@vt.edu) With the advancement of WBG devices like SiC-MOSFETs, a new frontier of higher density MV power converters have emerged. Some recent demonstrations of such PEBBs include a 100 kW, 28 kHz, hybrid Si-IGBT/SiC- MOSFET three-level T-type PEBB with a power density of 27.7 kW/l; and a 100 kW, 100 kHz SiC H-bridge PEBB with a power density of 5 kW/l (PEBB 1000). It used two commercial 1.7 kV SiC half-bridge modules to form an H-bridge converter. The PEBB 1000 is comprised of many discrete components – SiC half-bridge modules, gate drivers, magnetics, and capacitors. These components are connected using additional mechanical parts, such as screws, busbars, and baseplates. The connection of multiple discrete components limits the volumetric power density, performance, and manufacturability of the PEBB. The presented design feature a 250 kW, 500 kHz integrated power electronics building block (iPEBB) for high-density applications. Simulation work was performed to design a common substrate that will serve as the electrical, thermal, and mechanical foundation of the iPEBB. SiC MOSFET bare dies and passive components that form the iPEBB topology will be directly placed on the common substrate to provide electrical interconnects and a unified platform for cooling. Substrate survey To form the common substrate, various ceramic and organic materials were explored such as organic DBC (ODBC). The ODBC in this work uses a thin polyimide film for the dielectric. The ductility of the film enables thick metal layers (>1 mm), compatibility with a variety of conductors (e.g., Cu, Al, TPG), and large footprints (>300 mm by 600 mm), enabling, in theory, the entire iPEBB to be on a single ODBC sheet. This would eliminate the need for a baseplate, especially since the ODBC can achieve thick metallization, which can provide the mechanical support for the iPEBB. Eliminating the baseplate removes a thermal interface, and can simplify manufacturing, reduce weight, and improve reliability. The ODBC is a promising technology that is compatible with a variety of conductors and comes in thick metallization and large footprints. The thick metallization offers heat spreading and mechanical support for the converter. The ODBC’s large footprint offers the ability to eliminate the need for a baseplate thereby reducing the number of thermal interfaces, reducing the weight below 16 kg, and simplifying the manufacturing process. The common substrate uses a multi-layer ODBC design to provide a common-mode (CM) screen and a low power loop inductance for fast switching application (Figure 1). But the implementation of the common substrate will require trade-offs including increased thermal resistance associated with a multi-layer design. Switching tests were performed at 1 kV, revealing a dv/dt of 27 V/ns and a voltage overshoot of approximately 96 V. The thermal resistance measurements and simulations confirmed that thicker top-side metallization will be needed for improved heat spreading. These insights will go into a multi-physics redesign of the iPEBB and will advance high density medium- voltage converters for future electric systems. Literature Design of a High-Density Integrated Power Electronics Building Block (iPEBB) Based on 1.7 kV SiC MOSFETs on a Common Substrate, APEC 2021 Procedings, pages 1-8 Figure 1: (a) Common substrate with the integrated SiC bridges, (b) top view of the integrated SiC bridges, (c) half-bridge schematic, (d) switching cell, (e) cross section of multi-layer OBDC common subtrate