Online frequency response analysis: A powerful plug-in tool for compensation design & health assessment of digitally controlled power converters

Modeling the effects of digital control and digital PWM can be complex for power converters with multiple switches and unique PWM switching schemes; hence an approach of measuring the plant and open loop response is typically followed to verify the modeling or use frequency response data to identify the plant. This paper presents a cycle efficient software frequency response analyzer (SFRA), that is added to the software of the digital controller being used to control the power converter. SFRA needs no extra hardware connections and hence can be run periodically. The gathered frequency response data can be used to correlate with the health and changes in stability margins of the power converter; which can change due to aging and degradation of components caused by electrical and thermal stresses. The paper compares fixed and floating point implementation of SFRA and discusses a dual core approach to run SFRA for high switching frequency power converters. A bi-directional buck boost converter is used as a case study to demonstrate the SFRA, where first SFRA is used to extract the plant TF, which is then used to design the compensation and the designed compensation is then verified using SFRA.