Thank you for explaining in detail. Actually I had a look at text books. I'm not very illiterate and I actually know what causes oscillation and the idea of phase lead and compensation. Actually I have read what you correctly said about reason of oscillation in many text books and I understand them completely but as I said in first post providing alternative feedback path as suggested by you and Mr. Steve M. Taylor has some drawbacks and the most important is reduction of sensitivity of error amp especially at higher frequencies, so error amp can't control output voltage precisely. I'm looking for other solutions. I'm not suggesting that error amp should not have alternative feedback path at all but amount of this feedback should be very limited and other parts of this loop should be optimized so we can have both stability and precision. The question is how, I have read in brochure of some laboratory instruments that they have introduced loop stability factors that I guess are used to deal with this problem. As much as I could find information about them, they are used to limit bandwidth of power amp, I don't know how, and exactly what they are. Do you know anything about them?