The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> convergence problem in sim
https://designers-guide.org/forum/YaBB.pl?num=1206628802

Message started by xuedashun on Mar 27^th, 2008, 7:40am

Title: convergence problem in sim
Post by xuedashun on Mar 27^th, 2008, 7:40am

I designed a regulator which has several loops insides and am trying to check the stability for some loops by using dc&ac sim.

I use a loop breaker (large L&C, 1V ac source inside) to cut one loop and keep other loops unchanged. and use nodeset to give
desired voltage values at some important nodes. The ac sim works well at typical case (say 26 degree C, tpy MOS RES CAP...).
But it stops at some corners. I check the sim log and found that even the dc can not work well at the corners because of convergence
problem, ac also did not run because of convergence problem.

Then I tried run a tran sim where the stable operational point is reached and got good result. I check the output date file and
the voltage values look right and run ac again by setting prevoppoint YES in ac option and using same loop breaker. but ac sim
still does not work.

Is anyone can help me and give me some suggestions on this sim problem? Thank you very much!

Title: Re: convergence problem in sim
Post by mohta on Apr 16^th, 2008, 11:51am

Try to use STB/PSTB analysis - makes life simpler. Look at the Spectre documentation or search this forum. This topic has been discussed many times before. If you can identify one path that breaks as many loops as possible, then breaking the loop over there will give you necessary and sufficient conditions for stability of at least the broken loops.

Title: Re: convergence problem in sim
Post by thechopper on Apr 16^th, 2008, 6:52pm

One of the last discussions concerning multiloop circuits is here
http://www.designers-guide.org/Forum/YaBB.pl?num=1198131494

Hope it helps
Tosei