convergence problem with contact analysis

[albert1981]

Dear all,

I have some convergence problem with a contact analysis. I run the model of a quarter of the plate applying a cyclic load for each node. After one cycle I release the node from its symmetry and I apply the cycle at the following node. I put an analytical surface in the symmetry line in order to avoid the overclosure due to a plastic wake which has been formed after applying several cycles. This model run fine but when I input initial residual stress by using the SIGINI subroutine I have some convergence problems. The contact model that I used is a hard contact type between rigid elements made by an analytical surface and a seris of nodes.
In attached there is the msg file.

Do you have any idea on whch could the problem be?

thanks in advance

Albert

 

[albert1981]

Dear all,

I tried to use an hard modified contact interaction property, but the error persists and is the "Severe discontinuity iteration". According to the contact algorithm in 12.4.5 of the ABAQUS docomentation if there are changes in contact then it does not even check the equilibrium.
Is there anybody who can give me some suggestion?

Thanks a lot,
Albert