Problem Modeling JKR contact using Abaqus
I have been working on solving contact problems relating to fibrillar adhesives over the past 3 yrs of my PhD work. In the past, I used Comsol FEM to model contact interactions by implementing the Dugdale-Barenblatt traction separation law at the contact interface.
In my recent work, I will like to carry out situations that involve materials that are not originaly in contact, but adheres upon contact. To do this I have switched to Abaqus because of its good reputation in adhesive contact modeling. With the help of online user manuals and examples, I have been able to successfully simulate various contact problems with the exception of adhesive contacts.
As a barebone model, I have made several attemps at modeling the JKR contact - the case of a rigid sphere indenting a soft surface in the presence of adhesive interractions, but failed. I am using a 2D axisymmetric hemisphere as indenter. I specified contact surfaces with the sphere bottom as the master surface and the top nodes of a rectangular substrate as the slave surface. The contact interaction is a node to surface contact (abaqus/ standard) with the default traction-separation behaviour specified along with damage/ damage-evolution properties at the interface. I used axisymmetric stress CAX4R elements for the mesh of both parts.
The model fails with the error 'too many attempts made for this increment'. I am using a max # of increment of 1e6, with increment size of initial=5e-5, minimum=1e-11, max=0.01.
I have attached images of 2 separate cases at the last increment before failure. In the first case, the crack does not initiate even though the stress is high enough to initiate damage. In the second (low damage parameter), the last node in the center of the sphere does not break from contact.
I will greatly appreciate any suggestions of what may be the problems with the model.
I have also shared links to the work files: