Kristofer Eberle
Director of Business Development,
Plexim, Inc., USA
Biography: Mr. Eberle is responsible for business development and strategy for the Americas at Plexim, which develops the industry-leading PLECS simulation software platform for power electronic systems. Having joined the company as an Applications Engineer and the founding member of the U.S team in 2010, he has a deep understanding of Plexim’s portfolio, as well as the needs and challenges of its customers. Mr. Eberle also leads the company’s collaboration with semiconductor manufacturers to support device models for PLECS and custom web-based simulation environments, and advises the technical roadmap and vision for future development efforts in these areas. Mr. Eberle received his BSEE from Northeastern University in Boston, MA, and his MSEE from the University of Wisconsin-Madison, with a focus on power conversion and control systems. He is an IEEE PELS member.
Title
Hot Topic: Intuitive Semiconductor Thermal Loss Estimation for System-Level Design
Abstract: Thermal management is an important aspect of power electronic design and becomes more critical with increasing demands for compact packaging and higher power density. As wide bandgap technologies rapidly penetrate the application landscape, engineers are faced with learning how to deploy higher switching frequencies and unfamiliar control strategies. The primary challenges to investigate the thermal side of a new design are the limitations of device models and lack of consistent and complete public manufacturer datasets.
While traditional SPICE-based device modeling approaches determine semiconductor switching losses from current and voltage transients, such models are commonly configured for a very specific and limited use case, or have no temperature dependence at all, and resultant simulation speeds are often unbearably slow.
An alternative method deploying ideal semiconductor switch models yields fast and robust simulations, while leveraging detailed lookup-table (LUT) data for accurate switching and conduction loss estimation over a wide range of operating conditions. As major device manufacturers increasingly support such LUT models for direct application in commercial software tools, designers can not only predict performance and efficiency, but also investigate tradeoffs of topologies, components and their drive circuits and control schemes, confidently size cooling systems, and predict device lifetime reliability.
