AN-57

Hi.

You are absolutely right. RF3 should be taken into account, and the gain of the opto should be there as well. So where we read K_TL431, we should read K_TL431*K_Opto*(1/RF3). The rest of the equation remains the same.

We are finalizing a reviewed version of AN57.

Thanks for pointing this out 

Yes I  understand that, but are you sure the s-terms are correct?

(1 + s/(RC)) / (1 + s/ (KRC)) doesn't look right either.

While you are reviewing AN-57 pls also check "Step 8 - selection of CF1".

It states "fzero be set at 50Hz rather than 100Hz" and I believe you mean this for continuous mode (not discontinuous which is stated).

Its correct... It is a mechanism of taking the finite open loop gain of the error amplifier into account. This looks like a pole at a frequency close to 0Hz.