Unphysical oscillations in tangential forces for contact problem #24632
Assignees
Labels
Comments
recuero
added a commit
to recuero/moose
that referenced
this issue
Jun 9, 2023
…b#24632) This commit changes the way the normal contact force is used in the node-face penalty friction algorithm. The test that failed with this change only tests a point contact for a couple of time steps. That failure was related to a too-large penalty factor. With the previous approach, a too-large penalty factor may not initially have a negative influence because the algorithm was using a cached normal force value. Slightly decreasing that value, however, makes a more extended approach to penalty contact work for the single point test modified in this commit. Another example of a frictional cylinder is added to avoid regressions. That test checks for tangential slips in opposite directions. The previous formulation is unable to run this example.
recuero
added a commit
to recuero/moose
that referenced
this issue
Jun 9, 2023
…b#24632) This commit changes the way the normal contact force is used in the node-face penalty friction algorithm. The test that failed with this change only tests a point contact for a couple of time steps. That failure was related to a too-large penalty factor. With the previous approach, a too-large penalty factor may not initially have a negative influence because the algorithm was using a cached normal force value. Slightly decreasing that value, however, makes a more extended approach to penalty contact work for the single point test modified in this commit. Another example of a frictional cylinder is added to avoid regressions. That test checks for tangential slips in opposite directions. The previous formulation is unable to run this example.
recuero
added a commit
to recuero/moose
that referenced
this issue
Jun 13, 2023
…b#24632) This commit changes the way the normal contact force is used in the node-face penalty friction algorithm. The test that failed with this change only tests a point contact for a couple of time steps. That failure was related to a too-large penalty factor. With the previous approach, a too-large penalty factor may not initially have a negative influence because the algorithm was using a cached normal force value. Slightly decreasing that value, however, makes a more extended approach to penalty contact work for the single point test modified in this commit. Another example of a frictional cylinder is added to avoid regressions. That test checks for tangential slips in opposite directions. The previous formulation is unable to run this example.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Look into formulation for penalty node-face contact, particularly the definition of tangential traction from the normal pressure and its Jacobian. Look at one benchmark problem.
See discussion below.
Discussed in #24562
The text was updated successfully, but these errors were encountered: