maximum duration of earthquake

[yuejun]

Hi
The maximum duration of earthquake is 50s. Is there any way to exceed this limit? Say, I wanna a 100s ground motion to be analyzed.
Thanks.

[arbarbosa]

The maximum duration of the record and/or analysis is not limited to 50 seconds. It depends on your record and on the number of analysis steps that you perform.

Are you talking about the analysis procedures developed by Sylvia? If so you can write, for example

Code: Select all

 set TmaxAnalysis 60; # analysis duration 60 secs.
AnalyzeTimeHistory -TmaxAnalysis $TmaxAnalysis

[newarrived]

Hi, I used to perform dynamic analysis with various time duration, but every time I got various response. Let me explain more...
I want to perform IDA, moreover my model include P-Delta effect using geometric transformation 'PDelta'.
Earthquake time duration is 40s, In high scale ground motion say Sa(T1,5%)=2g I use time durations 50s and 60s for analyze and I get non-convergence in time about 50-55s. I mean with max. time duration equal to 50s I don't get non-convergence but whit 60s I get non-convergence.
In low scale ground motion say Sa(T1,5%)=0.5g I use time durations 50s, 80s and 110s for analyze and I get non-convergence in time about 90s. Is it logical?
Finally I want to say: when I use longer time duration, I get lower IDA curve.
Can there be a problem in OpenSees? what time duration do I have to use?

[fmk]

i would say here is more likely to be a problem with the model or your IDA stopping criteria than opensees.

what do you define failure of the building to be in your determination of the IDA?

[newarrived]

dear fmk, thanks for your attention. I'm trying to employ IDA for SAC 9-story moment frame building that was applied previously by Vamvatsikos and Cornell, 2003a. the criteria that was expressed is max. interstory drift ratio.I use exactly the same procedure namely 'hunt & fill'. I was successful to drive almost the same IDA curve with time duration equal to 50s and loading that I think it isn't exactly the same as expressed for this model. note that the pushover was fine. but since I increase time duration, the end part of my curve changes.
In any case I can't figure out why the time history analysis fail in large time duration even in low ground motion scale!

[newarrived]

I found the problem where I applied gravity load bud i didn't do gravity analysis before dynamic analysis, and my structure don't fail in large time duration now.
my new question is:
To take into account first order treatment of geometric nonlinearity we use 'Linear Transformation' and to take into account second order treatment of geometric nonlinearity we use 'P-Delta Transformation'; am I right?
when I use 'Linear Transformation' I can't see the horizontal part of IDA curve and it's almost linear!

[fmk]

yes linear and pdelta (or maybe Corotational if displacements large enough to warrant extra computation time)