Turning a shaft through conact pressure in ABAQUS/standard?

Hello all,

This is my first post here. This seems be an informative and helpful forum.

I am trying to model the rotation of a "hard" shaft as a result of motion of 2 strips of soft materials (please see attached figure).

The strips are deformed due to external forces and as a result, contact pressures are applied on the shaft. The forces are then varied on the strips leading to net rotational forces on the shaft.

I am interested in the torque applied on the surface of the shaft and the angle of rotation (from some abituary changes in forces). I am not interested in the kinematics of the problem.

I have set up the model as follow in ABAQUS/Standard:

• The material of the strips are described by a custom user routine and have been tested to be working correctly.
• There are BC's constraining the end of the strips.
• The shaft is fixed in all degrees of freedom except with rotation with axis out of plane (UR3).
• The actual model in Abaqus is a 3D model. The strips are represented with membrane elements, and the shaft by 3D deformable elements.

I have tested the model with frictionless contact. The contact surfaces are the edge of strips and the surface of the shaft. The model worked as expected, the strips conform to the surface of the shaft and slides across the surface of the shaft (due to absent of friction).

However, when I add in friction, the solution fails to converge. This was most likely due to opening and closing of the contact surface when the shaft is turning (i.e. the strips exert some turning force, turning the shaft. But near ends of contact, they come off contact and possibly lead to "chattering" as described in ABAQUS manual). Would anyone be able to give me pointers into preventing this from happening or some other set up to get a converged solution?