The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> stability for ckts having one feedback and one forward loop
https://designers-guide.org/forum/YaBB.pl?num=1599231943

Message started by tulip on Sep 4^th, 2020, 8:05am

Title: stability for ckts having one feedback and one forward loop
Post by tulip on Sep 4^th, 2020, 8:05am

I designed a circuit having one feedback and one feed forward loop, the feed forward loop is used to provide a zero to improve the phase margin of the feedback loop.
When I break the feedback loop, the stb analysis shows the loop is stable, but when I break the the feedforward loop, the stb analyais shows the loop is unstable.
The transient reponse of the circuit is stable.

In cadence help file "spectreref.pdf", it says: The loop-based algorithm can only be performed on individual feedback loops to ensure that they are stable. Although the stability of all feedback loops is only a necessary condition for the whole circuit to be stable, the multi-loop circuit tends to be stable if all individual loops are associated with reasonable stability margins.

Does cadence mean that if one of the loop of a multi-loop circuit is unstable, the whole circuit is unstable?
How can I check the stability of the circuit I designed?

Thank you very much.

Title: Re: stability for ckts having one feedback and one forward loop
Post by Rakesh on Oct 7^th, 2020, 8:28pm

Can you post some schematic diagram

Title: Re: stability for ckts having one feedback and one forward loop
Post by tulip on Oct 9^th, 2020, 6:30am

This circuit is a buffer(common mode feedback circuit not shown)

Title: Re: stability for ckts having one feedback and one forward loop
Post by tulip on Oct 9^th, 2020, 6:43am

I added diffstbprobe at A, and plot the magnitude and phase response. The circuit seems stable.

Title: Re: stability for ckts having one feedback and one forward loop
Post by tulip on Oct 9^th, 2020, 6:50am

I added diffstbprobe at location B, and plot the magnitude and phase response. at 269mdB gain, its phase margin is only 22 degree. From this figure, it seems the circuit has the risk of oscillate.

Title: Re: stability for ckts having one feedback and one forward loop
Post by Frank Wiedmann on Oct 12^th, 2020, 5:06am

If the path through B is a feed-forward path, then there should be no significant feedback loop in that place. Anyway, examining a local loop with the global loop being active often gives rather strange results because the signal also goes around the global loop.

Title: Re: stability for ckts having one feedback and one forward loop
Post by tulip on Oct 15^th, 2020, 6:03pm

Thank you very much. After these days of thinking, I deemed that I should break the loop at point A to caculate the phase margin of the opamp.

The reason is : the capacitors are used to provide frequency compensation, after conpensation, the opamp can be considered as an amplifier with a stable transfer fuction T(s), so, the nyquist plot of T(jw) can be used to determine the feedback amplifer's stability, the position A is the appropriate position to break the feedback loop.