Learning Forum

[manuel.uruena]

I am modelling in APDL an axial loaded shell tube subjected to local buckling, with a force load and arc-length method. I have introduced geometric imperfections.

My solution always stops due to uncovergence at the maximum load before the negative slope of the load-displacement curve, and local buckling does not take place yet. The error message is that the arc-length method reached a minimum. If I use a small value for minimum mutiplier in arc-length method, unconvergence still occurs, but no longer with the reached minimum error message. All coloured lines of solution suddenly drop down.

I have tried playing with a large number of substeps, a very small MINARC value, lowering the MAXARC value to 1, changing from CERIG to RBE3 constraints, increasing number of maximum iterations to 50, lowering the equivalent plastic strain in cutback controls to 0.05,... etc.

How can I achieve convergence?

[Ashish Khemka]

Please see if the following discussion helps you.

Arclength method does not gives converged solution in buckling analysis ÔÇö Ansys Learning Forum

Regards Ashish Khemka

[Bill Bulat]

I confess I've had trouble using ARCLEN successfully in the past myself. I was surprised to see so many ARLEN examples in our Help (image below). Maybe one of these bears similarity to your model and gives you insight into how to resolve the issue?
Kind regards Bill