The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Jul 18th, 2024, 5:34am
Pages: 1
Send Topic Print
transient simulation become very slow at a point (Read 2594 times)
sear
New Member
*
Offline



Posts: 5

transient simulation become very slow at a point
Nov 11th, 2010, 5:42am
 
I'm running transient simulation with AMS->APS->spectre. From 0 to 100us, it take 1 hour then suddenly at 100us, it become very very slow. The step become about 100as. Any one has seen this kind of situation? I check the input, output signals, there is nothing special at this point.

Thanks.
Back to top
 
 
View Profile   IP Logged
sheldon
Community Fellow
*****
Offline



Posts: 751

Re: transient simulation become very slow at a point
Reply #1 - Nov 15th, 2010, 10:16am
 
Sear,

  More information would be helpful. Usually the time step gets small
when there is a convergence issue. There are many phenomena that
may cause this type of issue, it might be best to contact your local
Cadence support.

                                                              Best Regards,

                                                                 Sheldon
Back to top
 
 
View Profile   IP Logged
wave
Senior Member
****
Offline



Posts: 117
Silicon Valley
Re: transient simulation become very slow at a point
Reply #2 - Nov 15th, 2010, 12:47pm
 
It's sometimes helpful to twiddle with transient settings.  
Increase Cmin, change timestep, and perhaps the convergence algorithm.

Good luck!
Back to top
 
 
View Profile   IP Logged
sheldon
Community Fellow
*****
Offline



Posts: 751

Re: transient simulation become very slow at a point
Reply #3 - Nov 15th, 2010, 6:06pm
 
Wave,

   Yes, you are correct that tweaking can help, in particular,your
example of setting cmin is good. Setting cmin is usually very effective
without greatly effecting the results. On the other hand, almost all of
the convergence issues that I have seen recently have ended up
being related to incorrectly formed models: models with large
negative resistance, non-monotonic functions(higher order derivatives
blow-up), etc.  These issues can be worked are by tweaking the
simulator parameters but it is better to identify the issue and fix it.
The bonus is that fixing the problem often significantly improves
simulation time. The down side is that typically developer's will need
access to the testcase in order to understand the issue.

                                                        Best Regards,

                                                           Sheldon
Back to top
 
 
View Profile   IP Logged
Pages: 1
Send Topic Print
Copyright 2002-2024 Designer’s Guide Consulting, Inc. Designer’s Guide® is a registered trademark of Designer’s Guide Consulting, Inc. All rights reserved. Send comments or questions to editor@designers-guide.org. Consider submitting a paper or model.