Forum for OpenSees users to post questions, comments, etc. on the use of the OpenSees interpreter, OpenSees.exe
Moderators: silvia, selimgunay, Moderators
-
crazylittleapple
- Posts: 9
- Joined: Tue Apr 12, 2005 11:14 am
Post
by crazylittleapple » Tue Apr 12, 2005 11:34 am
Could anyone please give recommendations about how to set the convergence tolerance?
-
silvia
- Posts: 3909
- Joined: Tue Jan 11, 2005 7:44 am
- Location: Degenkolb Engineers
-
Contact:
Post
by silvia » Wed Apr 13, 2005 9:12 am
i'd go with 10^-8 or so....
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104
-
fmk
- Site Admin
- Posts: 5883
- Joined: Fri Jun 11, 2004 2:33 pm
- Location: UC Berkeley
-
Contact:
Post
by fmk » Wed Apr 13, 2005 3:15 pm
the tolerance is really up to you .. it depends on your model, your units, time step, what norm you are using, i.e. disp, unbalance, or a relative one, and of course what you would be happy with.
-
crazylittleapple
- Posts: 9
- Joined: Tue Apr 12, 2005 11:14 am
Post
by crazylittleapple » Wed Apr 13, 2005 6:13 pm
Thanks a lot for you guys replies. Then, how can we know what value is good for the analysis?
From the manual, the tolerance is an absolute value (i.e, it is not a ratio, for example, for the force criteria, we set tol=0.1 means tol=0.1 N (SI units) ), right?
-
silvia
- Posts: 3909
- Joined: Tue Jan 11, 2005 7:44 am
- Location: Degenkolb Engineers
-
Contact:
Post
by silvia » Wed Apr 13, 2005 11:18 pm
the 0.1 if Newtons if that is the basic unit you are using, yes.
I want a small tolerance because i don't want the error to build up/propagate.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104