Non-Convergence of time history cases - Instability and ill conditioned structure

[BOCH]

Hi guys,

I am currently facing a problem with SAP 18.0.1 while running a model that is seismically isolated using triple friction pendulums.
9 non-linear time history cases are applied in the triplet directions (XYZ in same case).
Some of these cases are converging and some are not .. at the end of the non-linear analysis an instability and ill conditioned warning is shown as well.
I have checked the model many times, but I could not figure out what could be the reason of this non-convergence and the instability ..
Does anyone have an idea ?

Thanks,

 

[UcfSE]

Did you run analyses with the standard solver and fix any errors that came up?

 

[BOCH]

Yes I tried this but I haven't seen the any errors, it looks to me like the advanced solver ..
Maybe I have done something wrong, how could you see in the standard solvers such errors ?

 

[LQQD]

Hi,i think you may already know this but instability warnings should come out on the analysis log. I have been working with V18 and it was soooo slow, 4 times slower actually, i had to go back to V17. Did you notice any trouble? Did you try running the model on V17?

 

[Shz713]

There are some reported issues with analysis running with this version.

What are the degree of tolerance? Did you check carefully lower order modes mybe there are some out of range/local failure which causing these errors. Investigate modal shapes by Ritz vectors, then crosscheck he connectivity of any used links. Plus, ensure adequate time-step, total duration, damping, and material properties.

Shoot for the Moon, even if U miss, U still land among Stars!

 

[BOCH]

Hi LQQD, is it possible to re-run the model with version 17 since it is already saved and modified in version 18 ?

Hi Shz713, thanks I will have a look at these issues.

 

[LQQD]

BOCH, yes it is possible. You just have to follow these steps

1) open de .s2k or .$2k file extension file. And then search the word "version" and change the version number with the one you will be opening it

3) Open a new instance of sap2000, blank model and go to file--> import ans select to impor the .s2k file

 

[BOCH]

Thanks LQQD,
I have tried to run the model in version 17.3.0, the instability issue have disappeared however the convergence issue for triplet time history cases with triple pendulum is still there ..