Time History Analysis on 3D RC Frame: Convergence problem at only one column

[elifyildirim]

Hi everyone,

I modeled a 3D RC frame+shear wall structure and performed time history analysis. All columns, beams, and walls have been modeled with fiber section forcebeamcolumn elements. Shear walls were modeled with rigid elements using rigid elasticbeamcolumns.
Periods, reactions from gravity analysis, pushover curves in X and Y seem reasonable.

However, when I perform time history analysis, only one column raises a problem and I could not find the reason. For example, while using a 20-sec ground motion, the analysis has stopped at 15 sec due to convergence problem at that column. My model has a code that changes algortihm to avoid convergence problems.

In the StressStain results of that column, I see no stress at section 1 but abnormal stress on section 5 (like twice). [#integration points=5]

I checked nodes, element connections, loads, diaphragms, and fiber sections. I could not find any problem regarding modeling.

Can anyone make any suggestions?

Thank you very much in advance.

[mhscott]

Make things elastic in order to isolate the error, then start adding back nonlinearity.

https://portwooddigital.com/2020/08/21/ ... -a-sprint/

[selimgunay]

You can try reducing the number of fibers, especially in the cover.

[elifyildirim]

thank you very much for your suggestions!