The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Mixed-Signal Design >> Stability of SC CMFB
https://designers-guide.org/forum/YaBB.pl?num=1180121097

Message started by analog2000 on May 25^th, 2007, 12:24pm

Title: Stability of SC CMFB
Post by analog2000 on May 25^th, 2007, 12:24pm

Hi,
I am designing a sc cmfb for a differential sc integrator. Would anyone tell me on how to simulate the stability of a sc cmfb circuit ? I have simulated the stability of continous time cmfb circuit with stb analysis
Any help would be greatly appreciated
thanks

Title: Re: Stability of SC CMFB
Post by Berti on Jun 4^th, 2007, 12:25am

Hi,

similar to a continuous-time CMFB: break the loop and check the open-loop behavior.

I can recommend you an excellent paper about SC-CMFB:

Ojas Choksi and L.Richard Carley, Analysis of Switched-Capacitor Common-Mode
Feedback Circuit, IEEE Transactions on Circuits and Systems II, 2003.

Title: Re: Stability of SC CMFB
Post by Frank Wiedmann on Jun 4^th, 2007, 1:26am

If you have a recent version of SpectreRF, you can use pstb analysis. If you have an older version of SpectreRF which does not have pstb analysis, take a look at http://www.designers-guide.org/Forum/YaBB.pl?num=1171399171.

Title: Re: Stability of SC CMFB
Post by analog2000 on Jun 4^th, 2007, 3:37pm

Hi,
Thanks for the responses.

1) I don't have pstb available in the version of spectreRF that I am using

2) When I break the loop at a certain point in time, using "prevoppoint" option within the stb analysis. The opamp is not in the correct biased state. The transient sims run just fine and I get predictable response. Any ideas ?
thanks a lot

Title: Re: Stability of SC CMFB
Post by Frank Wiedmann on Jun 4^th, 2007, 11:27pm

In most cases, it makes no sense to analyze a switched-capacitor circuit in one of the two states only, because the switching action is necessary for the proper operation of the circuit and both states must be taken into account. You must either use some kind of pss analysis (pstb or pac), transient simulation (apply some disturbing signal to the loop and watch the response) or a model that combines both states (e.g. a state-space averaged model).

Title: Re: Stability of SC CMFB
Post by Berti on Jun 5^th, 2007, 12:25am

As far as I understand SC-CMFB circuits, a larger capacitance usually increases the return ratio and consequently decreases
the phase margin of the CMFB.

You are therefore mainly interested in the phase where both capacitors are connected to the amplifier (but I agree with Frank
that it never harms to check both phases in a SC circuit).

To get the right operating point first run a transient analysis and use the writefinal option so that it can be recalled to act as a nodeset
for a later DC analysis. Then cut the loop, provide an input signal and run an AC analysis with the prevoppoint option set to yes.

Cheers, Thomas

Title: Re: Stability of SC CMFB
Post by analog2000 on Jun 13^th, 2007, 2:56pm

Hi,
1) The prevoppoint is not working for me ? I have filed a request with cadence so that they can take a look at it. The transient end point does not match with the ac analysis operating point, when I had prevoppoint set to yes ? Any ideas

2) I got upgraded, so that I now have pstb available. When I tried to run it, it gave bogus results. My gain margin was >0dB but my phase margin was around 0 degrees even though my transient response i.e. the step response is just fine. Any special techniques that I need to setup for pstb analysis ?
thanks
analog20000