Pushover Analysis Error

[sh.hosseinzadeh]

Hi there,
I have a building modeled with SeismoStruct 2016, while I am trying to perform pushover analysis it will end up getting me weird responses and/or not converging at all. I have checked the file several times to make sure there is no material error. but I can not fix it.
Can anyone diagnose this issue? I also enclose the dropbox link such that you can download the file.

https://www.dropbox.com/s/06r1j51orkowg ... 7.spf?dl=0

Regards,

[seismosoft]

But you have not connected the control node (and all the centres of the slabs in the different floors) with your structure. The rigid diaphragms is not enough. Better to use a rigid link.

Seismosoft Support

[sh.hosseinzadeh]

Thanks for your reply,
Can you elaborate a little bit? what I did is using a rigid diaphragm besides introducing a node at the center of mass of the structure and I connected all of them together using a rigid diaphragm constraint.
how can I use rigid link instead since it doesn't yeild the same results when I am doing eigenvalue analysis?

Regards,

[seismosoft]

The Z+RX+RY DOFs of the control nodes are not connected in any way with the rest of the structure. This is what is mentioned in the warning message that you get, when you start the analysis (always pay attention to the warnings at the beginning!)

Regards,
Seismosoft Support

[obaidullah khan]

hello,
i did static pushover analysis and IDA of my structure got the results, but this time when i added 2 shear walls in building and run the PO analysis, the analysis is not running and giving the error "warning: could not re-orthogonalise all Lanczos vectors".why is it so, what could be the reason as i have connected the shear wall nodes to rigid diaphragms and control node of story through rigid link.

[z.gronti]

Dear obaidullah khan »,

If you carry out a search in this Forum on "could not re-orthogonalise all Lanczos vectors" you will find posts with hints on this issue.

[obaidullah khan]

Dear Zoi Gronti,
Thanks for your response I carry out the research and I got some connectivity issue in my model, I solve that now my analysis run completely and all values got converge but the issue is software is not considering the shear walls which I have added. Program is giving the same base shear value as my other model was without retrofitted. I have connected the shear wall to rigid diaphragms and also connected to control node of each story through rigid link constrains

[z.gronti]

Dear obaidullah khan,

Please check the connectivity of the added walls with your model and the walls' support conditions.

[obaidullah khan]

Hi,
what could be the reason in static pushover analysis when a curve studdenly break each and every time after attaning its peak value and does not complete the analysis/curve,gives error elm_inv every time. I have increase the values of element iterative strategy, global iterative strategy an convergance criteria to there maximum limit in the setting tab, also make alteration in the loading phase (steps, max displacement) but still analysis does not completely run.kindly tell what could be the reason?or where else i can improve?i have checked all the things throughly there is no mistake in the modelling on sections definitions etc as the T.P of my seismostruct matches with the T.P of my Etab structure.Just the prob is POA doest not run completely.
Regards

[huffte]

I have encountered situations similar to that you describe. The issue was a real failure of the material based on fracture strain being achieved. I wonder if you might check the parameters in your material definition and in your performance criteria. Then compare limiting values you specified to the values attained just prior to the "breaking" you describe. It could be a legitimate failure point or it could be an error in performance criteria and material specifications. These are only two of many possibilities.