Hello again,

I was looking over your worksheets on this site and came across this one. My question is about the first question, and follow-up question 2 states as follows:

Follow-up question #2: what would have to be done to reduce the ripple voltage at the integrator's output?

I'm wondering what the suggested answer to this is, if any. Thanks

 

I don't like that they describe the low pass filter as a passive integrator. While it is true that a low pass filter does behave as an integrator at frequencies well beyond the cutoff frequency (when suitably scaled), the terminology is a little misleading in this case. The circuit, as shown, will give the correct DC average value, while a true integrator would be unstable and would keep integrating the average value until the system is saturated at the voltage limit.

Anyway, once we realize that it's a low pass filter and not a true integrator, we can say that ripple can be reduced either by shifting the cutoff frequency to lower values, or by increasing the order of the filter so that the PWM frequency is attenuated more.