User login

Navigation

You are here

Why getting trouble with Abaqus UMAT subroutine when using interaction features like contact or tie?

Rudraprasad Bhattacharyya's picture

I am implementing a constitutive relation in UMAT subroutine using the commercial software Abaqus. First, I tested the UMAT subroutine with 'one element model' first. Then I tested with comparatively bigger geometry. It worked fine. Now when I am trying to do a models involving contact, i.e. use of the 'Interaction module'. I am getting error of over-constraint warning for 'Split Tension test of concrete cylinder'. The same input file (except the material definition) is running on Abaqus perfectly with Mohr-Coulomb criteria. I can see a reasonable stress and strain contour. But when I am using UMAT, the simulation runs with no warning or error message. The ODB file shows zero stress in the specimen. Interestingly, the Abaqus message file gives following information
1. " ***WARNING: Solver problem. Zero pivot when processing D.O.F. 3 of 2 nodes. The nodes have been identified in node set WarnNodeSolvProbZeroPiv_3_2_1_1_1".
2. "110 SEVERE DISCONTINUITIES OCCURRED DURING THIS ITERATION. 10 POINTS CHANGED FROM CLOSED TO OPEN. 100 POINTS CHANGED FROM SLIPPING TO STICKING"

My questions are:
1. Why the same input file with Abaqus' in-built material definition did not lead to above mentioned warning?
2. Is there a way to numerically assign a unsymmetrical solver to my UMAT material?

Subscribe to Comments for "Why getting trouble with Abaqus UMAT subroutine when using interaction features like contact or tie?"

More comments

Syndicate

Subscribe to Syndicate