The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Question on Op-amps used in a feedback loops
https://designers-guide.org/forum/YaBB.pl?num=1186616165

Message started by hchanda on Aug 8^th, 2007, 4:36pm

Title: Question on Op-amps used in a feedback loops
Post by hchanda on Aug 8^th, 2007, 4:36pm

Hi all,

When a differential input single ended opamp is used in a feeback , consider the case when it is used in bandgap voltage reference circuits, it forms
both positive and negative feeadback loop. Negative feedback loops are stable do not have worry about it much, but postive feebacks results in oscillations if
the gain is > 1 (Barkheusen criteria). So in the above case, I mean in bandgap circuits, we can just assure the total system to be stable by making the
loop gain of negative feedback > loop gain of positive feedback, my question is this (just this one condition loop gain of negative feedback > loop gain of positive feedback)
good enough to assure the stability or do I need to make sure that the loop gain of the positve feedback loop to be < 1.

Thanks in advance,
hc

Title: Re: Question on Op-amps used in a feedback loops
Post by tosei on Aug 8^th, 2007, 7:38pm

Hi hchanda,

This is one case for multi-loop feedback systems. Generally, when there is a common break points for all the loops involved in the feedback - which could be the case for the band-gap example you are bringing - the stability is guaranteed if any loop separately is stable. In addition, under this condition the multi-loop feedback system can be studied as a single feedback loop by breaking the loop right at that common point and looking for usual stability parameters.

If thought individually I´m temped to say the positive feedback one would have to be > 1, but I would analyze it as a single loop one as suggested above if you have a common break point.

Regards
tosei

Title: Re: Question on Op-amps used in a feedback loops
Post by rajeee1000 on Aug 8^th, 2007, 8:07pm

I agree with Tosei. To rephrase, when there are two loops, making certain the negative feedback loop gain greater than positive feedback loop gain will make the overall feedback, or the equivant single feedback, negative.

Rajesh