Every control loop you build should be properly design and measured before you ever ship a power system. This page provides tips on how to do this quickly and reliably.

Loop Gain Measurement Setup with AP 200 USB Frequency Response Analyzer (Isolated Feedback Power Supply)

Use the setup above to measure transfer function of an isolated-feedback power supply. By moving the probes for Channel A and Channel B, you can measure several different transfer functions without changing the injection point:

• Channel A at point 1 and Channel B at point 2 to measure loop gain
• Channel A at point 3 and Channel B at point 4 to measure power stage and modulator gain
• Channel A at point 4 and Channel B at point 2 to measure compensation and isolation gain
• Channel A at point 1 and Channel B at point 3 to measure control chip buffer gain
  

Loop Gain Measurement Setup with AP 200 Parallel Frequency Response Analyzer (Non-Isolated Feedback Power Supply)

Use the setup above to measure transfer functions of a direct feedback power supply. By moving probes for Channel A and Channel B, you can measure the following transfer functions without changing the injection:

• Channel A at point 1 and Channel B at point 2 to measure loop gain
• Channel A at point 1 and Channel B at point 3 to measure compensation gain
• Channel A at point 3 and Channel B at point 2 to measure power stage and modulator gain
  

Hints for Successful Loop Gain Measurement

• Always ground the control circuit being measured whenever possible for both safety and noise immunity. If the circuit is ac-line referenced, use an isolation transformer to provide the power supply input, and ground the circuits appropriately. This will also make circuit debug and monitoring easier.
• If it is impossible to ground the control circuit, the signals to the
  frequency response analyzer must be isolated. Do not rely on capacitors to achieve this isolation since they can cause catastrophic failures, especially during transients. One useful technique for signal isolation is to use an oscilloscope which has an analog output which provides a very safe and effective buffer.
• When starting a sweep at low frequencies (10 Hz), use the maximum available signal from the network analyzer oscillator (1.77 V). As the sweep frequency approaches the loop crossover frequency, reduce the amplitude of the oscillator signal. Monitor critical parameters such as amplifier outputs to make sure that the control circuit is not being limited due to over-modulation.
• If you are not sure if you are using the right signal level, reduce the signal by 6 dB. If the loop gain remains the same (ignoring the increased noise), the previous level was OK, and you can increase the signal again. If the loop gain changes significantly, stay with the lower signal level.
• When measuring switching power supplies, use the lowest bandwidth for the best noise rejection. For a faster sweep speed, use a higher bandwidth receiver setting.
• Loop measurements above 30 kHz can be very difficult due to instrument grounding and high-frequency crosstalk between cables on injected and return channels. This is true of any network analyzer, and great care must be taken if you want to extend measurements beyond this frequency.(See paper on Multi-Loop Control for Very High Frequency Quasi-Resonant Converters).
• Inject the signal into a low-ripple part of the circuit if at all possible.

 

Power Supply Loop Gains Measured with AP 200 USB Frequency Response Analyzer

© copyright Ridley Engineering, Inc. 2010