Power Design Tip 8: Reduce EMI performance by changing the power frequency (see PDF for full text!)

MOS power IC full range

KKST patch plug-in active passive crystal oscillator hot sale 24.576000MHZ

description:

When measuring EMI performance, did you find that no matter what method you use, there are still problems that exceed the specification by a few dB? There is a way to help you meet EMI performance requirements or simplify your filter design. This approach involves modulation of the power switching frequency to introduce sideband energy and to change the narrowband noise to wideband emission characteristics to effectively attenuate harmonic peaks.

It should be noted that the overall EMI performance has not been reduced, but has been redistributed. With sinusoidal modulation, the two variables of the controllable variable are the modulation frequency (fm) and the magnitude of the power switching frequency (Δf) that you change. The modulation index (Β) is the ratio of these two variables: Β = Δf / fm The choice of modulation frequency and frequency shift are two important aspects. First, the modulation frequency should be higher than the EMI receiver bandwidth so that the receiver does not measure both sidebands at the same time. However, if the frequency you choose is too high, the power control loop may not be able to fully control this change, resulting in a change in output voltage at the same rate. In addition, this modulation can cause audible noise in the Power Supply. Therefore, the modulation frequency we choose cannot generally be much higher than the receiver bandwidth, but greater than the audible noise range.

Figure 3 herein compares EMI performance measurements with frequency modulation and no frequency modulation. At this point the modulation index is 4, and as we expected, the EMI performance at the fundamental frequency is reduced by approximately 8 dB. Other aspects are also important. The harmonics are smear into the frequency band corresponding to their number, ie the third harmonic extends to three times the fundamental frequency. This situation is repeated at some higher frequencies, resulting in a noise floor that is much higher than the fixed frequency. Therefore, this method may not be suitable for low noise systems. However, many systems have benefited from this approach by increasing design margins and minimizing EMI filter costs.