Plant measurement through Loop Gain Meter

Question

We are using PLECS standalone 4.5.5 (64 Bit). we have a problem with the AC sweep analysis results related to the boost converter model.

We want to perform AC analysis to measure the plant once with perturbation/response block in open loop (control to output) as Figure ''Plant in OL" and once with loop gain meter block in feedback as Figure "Plant in CL".

If the loop gain meter can measure the open loop transfer function in the closed loop, the results of both tests must be the same, but they are not like the figure "Results".

We want to know what the reason is, because we will use the PLECS to simulate the test bench.

On the test bench, we also use the AC sweep method, but the plant must be measured in closed loop, so it is very important for us to trust the simulation result. If we want to simulate what we do on the test bench, we should use the loop gain meter block in our simulations.

Comments

Farzad, can you post your model so I can look at your analysis settings?

Answer 1

These results will not be the same as the converter is at two very different operating points.  The average duty for the first test case is 0.5 (assuming the carrier peak is 100) and the average duty in the second case is ~0.12. 

Correcting for the difference by adding an offset to the initial reference improves the results (Duty Correction).

Then the other issue is that small signal analysis requires the inputs to be “small signal”.  With the high level of current ripple in the converter being fed into the modulator input, I don’t believe this condition is being met.  If we add a low-pass filter to the current output, then the results align better (Duty Correction + Filter).

The results might be a bit different from yours as I had to guess at the switching frequency and peak value for the carrier ramp.