Natural Periods in SeismoStruct and other programs

[FoadEng]

Hi
I am using SeismoStruct for running some displacement based adaptive analyses. But I think I have got some problems with modeling and eigen value analysis in SeismoStruct. I wanted to know about the period difference between same models in SeismoStruct and other programs. For example for a 2 story (3D model) first mode period is 0.65 in SeismoStruct but 0.7 in SAP or ETABS. Personaly I think they should give us same results. So I just wanted to know if this difference is acceptable or not?! If it's not how much difference is good?
Thank you very much
Foad Mohajeri Nav

[seismosoft]

Hi,

Please do make sure the models are all truly identical. If they are, it is then up to you, as a user, to decide if a 10% difference in eigen value results is acceptable or not (it all depends really on the objective of your analyses).

We can also add that we are planning to replace, sometime in the future, the current eigensolver (based on Hughes' Lanczos algorithm) with a Ritz solver. We hope that this will help eliminating differences such as the one you have observed.

We re-emphasise, however, that often, such differences are caused by slight differences in the models employed in the different programs. Indeed, we are currently putting together a Verification Manual, comparing SeismoStruct's results with those obtained with other programs and/or theoretical solutions, and have in general not come across differences as large as you report.

Seismosoft Support

[fakharifar.mostafa]

Dear SeismoStruct Developers

I'm doing my MS thesis regarding to RC buildings under earthquake exitations utilizing SeismoStruct software. In order to justify my advisor professor about the utilization of SeismoStruct in my thesis, I have accomplished a set of verification and calibration works with SeismoStruct and comparing the results with similar programs like IDARC. Results from SeismoStruct agree reasonably well with those softwares.

[maysamr]

1-I am modeling a 8 story rc frame in seismostruct and etabs. etabs calculates 1st period as 1.76 but sesimostruct gives only 0.65.
2-I am using diaphragm constraints for each story but for a 2d 8 storis model why should it found 10!!!! modes?????
where can i upload my file to get help. it is urgent. plz help. my email is meysamr@yahoo.com.

[Romain]

Hi,
A fundamental period of 0.65s might be reasonable for an 8-storey building. Do you have any indications that it should be different (in this case T=1.76)?

Regarding the number of fundamental modes:
The use of rigid diaphragm constraints seam more reasonable to use in 3D models. Did you try instead rigid links or elastic stiff elements in your 2D frame?
If you are using distributed masses and/or your flours are not effectively stiffed, the additional modes that you obtain are probably local ones, with very low mass participation.

Cheers,
Romain

[seismosoft]

Dear maysamr,

We may add also that 10 modes are the default number of modes output by SeismoStruct in an eigenvalue analysis. If you wish to have a larger number, go to the Settings and increase such value.

Seismosoft Support