Good morning,
I am working on my dissertation and my topic is vulnerability assessment and retrofitting.I added to the original structure conventional braces considering an initial camber (imperfection) at the middle of the brace length, because i want the braces to buckle. I try to run an IDA analysis, but it allways stops at 0.4g or earlier, because off prd_ite, Diverge or fbd_tol message. I have been performed an Eigenvalue analysis and it is ok, i changed the convergency tolerance, elem. loop conv., max. it. numb., div. numb., but it doesn't help. I have to run the analysis until collapse, how i know if the analysis stops because of collapse or convergency problem, I know that the hysteretic curve can tell me if the collapse happened.
Could you help me with some advice.
Best regards,
Timea Jozsa
Error messages
-
seismosoft
- Posts: 1192
- Joined: 06 Jul 2007, 04:55
Re: Error messages
Please refer to the Convergence and Divergence flags page of the help system and the manual Have you tried to increase the number of steps?
Seismosoft Support
Seismosoft Support
-
jozsatimea
- Posts: 2
- Joined: 27 Jul 2016, 06:20
Re: Error messages
Thank you for the reply!
I use time step 0.005, displ/rot tolerance is equal to 0.001 and the element loop conv. tol. is 0.0001.
the maximum number of it. is 60 and the divergence it. is 40, but the analysis stops due to fbd_tol at PGA=0.3g. After checking the Histertic curves i can say that Fbd_tol has not been appeard because of collapse.
Should i increase more the convergence tolerances?
What is the highest convergence tolerance value, which ensure accurate results?
Is it possible that the buckling of braces causes the convergence problem?
I use time step 0.005, displ/rot tolerance is equal to 0.001 and the element loop conv. tol. is 0.0001.
the maximum number of it. is 60 and the divergence it. is 40, but the analysis stops due to fbd_tol at PGA=0.3g. After checking the Histertic curves i can say that Fbd_tol has not been appeard because of collapse.
Should i increase more the convergence tolerances?
What is the highest convergence tolerance value, which ensure accurate results?
Is it possible that the buckling of braces causes the convergence problem?
-
seismosoft
- Posts: 1192
- Joined: 06 Jul 2007, 04:55
Re: Error messages
What type of units are you using? If you are using m 1e-3 is too large. If you are using mm 1e-3 is too low. In general try to use between 1e-4 and 1e-5 for both displacement-based criteria (values in m for the displacement criterion).
Further, note that if you have collapse at a particular step, then you will not get convergence at that step, since the structure will not be able to withstand the gravity load and achieve convergence.
Finally, in the case of Fbd_tol, you can uncheck the relevant checkbox.
Seismosoft Support
Further, note that if you have collapse at a particular step, then you will not get convergence at that step, since the structure will not be able to withstand the gravity load and achieve convergence.
Finally, in the case of Fbd_tol, you can uncheck the relevant checkbox.
Seismosoft Support