Earthquake Response Spectrum Analysis - Problem

Posted by: davlasc13

Earthquake Response Spectrum Analysis - Problem - 02/25/20 11:35 AM

Hello,

I realized a calculation model under CAESAR II 2019 SP1.
Several static cases and static combinations have been successfully created and executed.
I wish to carry out the analysis of the earthquake of this model by a spectral dynamic calculation.
So I configured dynamic analysis with the load data for my project.

When the modal analysis is executed, it never stops.
Forcing the stop of the modal analysis leads to the generation of the results but the Sturm is not good. The analysis is incomplete.
I have to determine the modal analysis until a frequency of 33hz.
I have set a maximum limit of 1000 modes.

I don't understand where the problem comes from.
We have the same problem on other computers.

Thanks for your help.
I can share the model if necessary.
Posted by: davlasc13

Re: Earthquake Response Spectrum Analysis - Problem - 03/03/20 03:13 AM

Hi,

Has anyone ever encountered a similar problem on a modal analysis?
The model was also sent to Support France, no solution to date ... It was sent in level 2 ...

Dear Dave Diehl and Richard Ay, do you have any help with your level of expertise?

Thanks in advance.
Posted by: pooria1978

Re: Earthquake Response Spectrum Analysis - Problem - 03/10/20 04:11 AM


Hi davlasc,

as someone who has dealt with so many dynamic calculations I reckon your problem must be either for the fact that you have a lot of rigid elements in your model or the "estimated no. of significant figures in eigen solvers" would have to be increased.
I do not have CII-2019 so I can not check your calculation.

good luck!
Posted by: Richard Ay

Re: Earthquake Response Spectrum Analysis - Problem - 03/10/20 10:26 PM

Actually we've been looking at your model today. Increasing the number of significant figures has helped the solution (see Advanced Control Parameters). We do not have a definitive answer for you yet, but Tech Support is looking into your problem.

Your model is very (very) flexible - almost 50 frequencies below 4 hz. Additionally the similarity between the various lines means each has about the same frequencies. These will all have to be extracted. Also note that there is no guarantee that the frequencies are extracted in increasing numerical magnitude.