Quantification of chaotic spectrum and EMI suppression in converters

EMI of switching converter is mainly decided by characteristics of spectrum of PWM drive waveforms. So it is of great meaning to quantify and discuss that. This paper takes PWM drive waveform as a time series process or event. Based on statistical theory, probability density function and characteristic function are analyzed for periodic frequency-spreading scheme and chaotic frequency-spreading scheme. Based on invariable distribution, characteristics of power spectral are calculated for chaotic frequency-spreading further. Finally, chaotic frequency-spreading scheme is fulfilled successfully on a flyback prototype with PWM control chip SG6846, system operating stable, the peak of ripple keeping constant. And spectrum of the experiment system is expanded, so EMI is suppressed effectively.