Dear Cervenka Consulting,

I have a question about modeling the reversed-cyclic loading on a RC slab-column connection. First of all, I have tried to model the reversed cyclic load as a static loading as displacement-controlled test (Newton-Raphson method) with very little increment for each steps in order to get higher possibility of convergence. However, there will always be some steps where the analysis stops and cannot continue further due to the divergence criteria is trespassed. I have read a lot of discussion about this problem in any other forums (a lot of another commercial FE software have this convergence problem for reversed-cyclic loading) and there is one idea to anticipate this convergence problem by using dynamic analysis instead of static.

My question is, do you think it can be appropriate to model the reversed-cyclic loading in Dynamic Problem with a very slow time increment for each steps and introduce a little value of damping to stabilize the analysis? If it is not, do you have any suggestion to solve this convergence problem. I have read all of the troubleshooting from ATENA and I also have tried to use:1) elastic stiffness updated each step;2)reduce the incremental static displacement for each steps;3)repeat no converged times (5 times with 0.25 ratio). I also have use the latest GiD script and ATENA version 5.3.2 where this cyclic issue should have been adressed.

I am looking forward to hear your response and suggestion.

Best regards,

Andri Setiawan