deformation in a rigid link

[smhzs]

Hi,
I modeled a 2 story, 1 bay steel braced frame with tension-only braces. After confirming the modeling by checking the eigen-value results, a Nonlinear-Time History Analysis was performed. Although the braces will not undergo compression, the links experience some deformation. They are not supposed to be working this way because stiffness in the direction 1, is defined as an infinitely large value. As I tried varied strategies, it seems like a bug in SeismoStruct.

I can send you the model.
Regards,
SMHZ

[smhzs]

I would appreciate it if you could confirm the existence of problems with those Tension-only links.

[smhzs]

I am confused and do not know what to do with this model. Now that it works for eigen value analysis, it will not converge when I perform a nonlinear time history analysis.
Can you please help me? Here is the model:
https://www.dropbox.com/sh/j0po0vjosmq0jhn/yhQt9oyt-R

[Stelios_Antoniou]

The problem is the gap element. If you set d(-) to zero, the analysis runs without problems. This means that the structure cannot 'lose' the support at the link, and at the minimum loadind level that is applied initially in the time-history it fails

[smhzs]

Correct! Thanks a lot for helping.