Reply to thread

what you can do is use a sig gen  ( isolated if necessary  )  to inject a sine wave perturbation into the control loop  -  i.e.  into just after the volt ref or demand spot  -  insert resistor  ( say 100 ohm  ) if necessary, to isolate the Vref from the sig gen  -  and then look at the output of the converter on  AC coupled.

look at the now modified Vref input  with the other scope channel,

at low freqs the gain can be easily seen  ( keep the sig gen amplitude down until you get the feel for it  )  -  and the phase lag will be minimal,

you can write down   (  or track and store on a fancy scope )   the gain and phase delay at  10, 50, 100,  200,  500,  700, 1kHz  and so on  to 10 - 50kHz or so

and then plot.   This gives you the plant   ( i.e.  power converter only,   including modulator  )   gain and phase.

Fitting an 100 ohm  ( say )   injection resistor in the feedback loop,  usually after the last error amp,  or at the end of the Volt divider as it goes into the Vea   -  you can look at both sides with your scope and get the full loop gain/phase,

beware injecting noise with your probes  -  solder a 22k resistor to each point, to go to your probe(s)   to stop this  -   p.s.  you need a decent scope with good averaging capabilities.

The sig gen goes across the 100 ohm,   you will likely need an isolating injection transformer with near flat characteristics from 10Hz to  100kHz.  ( or a couple of large resistors either side back to the sig gen  -  say  150k  each  -  the sig gen then needs higher o/p capability -  10Vrms gives 3.33mV across the 100 ohm ).

You will need less signal at lower freqs,  and a little bit more at higher freqs  . .  .

happy plotting