Help me to clarify one detail with error amplifier for dc-dc converter

[AMSA84]

Hi guys,

Basically:

I had a type II compensator for a buck converter. With the design I have, specially with the drivers I have and according to all the known theory on how to design the control, I got a certain values for the components (that needed a bit of tunning). After I got the result I wanted and now the detail, the output of the error amplifier (symmetrical OTA used here) the output voltage was around 1.4 and with step load ranged from 1.2 to 1.8 or so. Everything ok.

I re-design the drivers (I concluded that here the driver delay plays an important role) and consequently the compensator network. After that I got another suitable response on the output (without yet trying the step load). That said, I checked the error amplifier output and now that voltage is around 2V, which is bad, because my pwm comparator can compare only voltage up to 2.3V and before hand I know that when I give the step load I will have a problem.

Question:

Why that error amplifier voltage has increased?

How can I decrease that voltage?

Hope made myself clear.

Regards and thanks in advance.

 

[dick_freebird]

Not much to go on, here. Does your error amp set the
duty cycle directly (voltage mode) or does it set the
current? If the former, did you do something that has
made the duty cycle change (like, say, very asymmetric
"driver" (power train) delays)? Or have you made some
change to current sensing without scaling other things
appropriately, such that you have to throw more current
(e.g. a low level loop instability that causes current to
slosh in and out of the output filter, wastefully and
needing to be made back up by more overall current
throw?

Seems to me like there is some debug activity lacking.