Hi,
I am doing a model which tests a composite beam-column sub assemblage. The model requires the use of "hinges" connected sometimes up to 4 elements. I have modeled the "hinges" as link elements which have a very high stiffness in all DOF except the M2 DOF which is very small. When I try to run the program it gives me a "sol_tol" message and is unable to apply the load. do you have any suggestions as to fix the sol_tol problem?
Thanks
sol tol
-
seismosoft
- Posts: 1271
- Joined: 06 Jul 2007, 04:55
Re: sol tol
Hi,
Very difficult question to answer, especially without seeing the model itself.
As usual in these cases, we suggest you to try to solve the problem in an as much as possible simplified subset of your model, and then go from there into the full model.
Then, make sure the model is statically balanced (i.e. that no unstable mechanism is formed when you introduce all the hinges you seem to be introducing); this is a typical scenario for problems such as the one you describe.
Finally, you can always "play" with the values of stiffness of the link elements or with the 'Iterative Strategy' settings; however, the error you are getting leads us to think more of an unstable structure (e.g. double hinges on all columns of a given storey, hinges on all elements converging into a given node, etc) rather than of a model with convergence difficulties.
Hope the above helps somehow.
Seismosoft Support
Very difficult question to answer, especially without seeing the model itself.
As usual in these cases, we suggest you to try to solve the problem in an as much as possible simplified subset of your model, and then go from there into the full model.
Then, make sure the model is statically balanced (i.e. that no unstable mechanism is formed when you introduce all the hinges you seem to be introducing); this is a typical scenario for problems such as the one you describe.
Finally, you can always "play" with the values of stiffness of the link elements or with the 'Iterative Strategy' settings; however, the error you are getting leads us to think more of an unstable structure (e.g. double hinges on all columns of a given storey, hinges on all elements converging into a given node, etc) rather than of a model with convergence difficulties.
Hope the above helps somehow.
Seismosoft Support