Date of Graduation
5-2019
Document Type
Thesis
Degree Name
Master of Science in Electrical Engineering (MSEE)
Degree Level
Graduate
Department
Electrical Engineering
Advisor/Mentor
Juan Carlos Balda
Committee Member
Roy McCann
Second Committee Member
Yue Zhao
Third Committee Member
Simon Ang
Keywords
gate signal modeling, die-to-die interactions, miller clamp, parameter variance, parasitic inductance, power packaging, SiC power module
Abstract
The main objective of this effort is to determine points of weakness in the gate network of a high-performance SiC power module and to offer remedies to these issues to increase the overall performance, robustness, and reliability of the technology. In order to accomplish this goal, a highly accurate model of the gate network is developed through three methods of parameter extraction: calculation, simulation, and measurement. A SPICE model of the gate network is developed to analyze four electrical issues in a high-speed, SiC-based power module including the necessary internal gate resistance for damping under-voltage and over-voltage transients, the disparity in switching loss between paralleled devices due to propagation delay, a high-frequency oscillatory behavior on gate voltage due to die-to-die interactions, and current equalization in the kelvin-source signal path. In addition, the analysis of parameter variance between paralleled MOSFETs and the effects of mismatched threshold voltage and on-state resistance on switching loss and junction temperature are investigated. Finally, three Miller Clamp topologies are simulated and assessed for effectiveness culminating in a solution for parasitic turn-on in high dv/dt systems such as those utilizing high-performance SiC power modules.
Citation
Curbow, W. A. (2019). Model Development and Assessment of the Gate Network in a High-Performance SiC Power Module. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/3189
Included in
Electronic Devices and Semiconductor Manufacturing Commons, OS and Networks Commons, Power and Energy Commons, VLSI and Circuits, Embedded and Hardware Systems Commons