Axisymmetric Penetration

[skilroy]

Hi all,

I have been modelling, with Abaqus Explicit, an analytic rigid body contacting a solid deformable body that is partitioned into two seperate materials. The interaction used for this contact is a surface-to-node interaction, with the master surface being the rigid body and the nodal surface being the entire deformable body. This worked fine in 2D using plane stress elements (see below), but the problem requires an axisymmetric analysis.

I created axisymmetric parts and created all properties, interaction loads, mesh etc. exactly the same as in the 2D model. However, when running the job for this model, the rigid body penetrates through the surface of the deformable body without any contact occurring (see below)

Initially, I thought that this had to do with the interaction. I change from penalty to hard kinematic with no luck. I also switched to a surface-to-surface contact but this did not change anything.
I also tried refining the mesh but to no avail either. Using a deformable body instead of the analytical rigid and refining both meshes did not change anything either.

Can any of you smarter Abaqus peoples think of anything else that could be causing this problem? Thanks!

 

[FEA way]

Try using general contact, this type of contact is currently the recommended one and works great in most cases.

 

[skilroy]

Hi FEA way, thanks for the reply.

The problem is I receive this error when trying to use general contact, so I have been trying to avoid it:

***ERROR: *CONTACT cannot be defined on 2D element facets. If the model
contains 2D or axisymmetric elements then *CONTACT INCLUSIONS, ALL
EXTERIOR cannot be specified, and the first surface name on each
data line under *CONTACT INCLUSIONS or *CONTACT EXCLUSIONS must not
be blank.

 

[FEA way]

Which version of Abaqus do you use ? General contact for axisymmetric elements in Explicit analyses was introduced in Abaqus 2019.

 

[skilroy]

I use the 2020 version, but I am running it on a supercomputer which uses a 2018 edition I believe. I guess that explains the error. I know that this problem has been modelled before using a surface-to-node interaction, so I am hoping I can figure it out using this interaction. It's strange that everything works for a 2D planer stress analysis but not axisymmetric, do you have any hunch as to why?

 

[FEA way]

Is point mass/inertia assigned to the reference point of this rigid body ? This helps Abaqus calculate proper contact stiffness and often solves penetrations issues like this one.

 

[skilroy]

No, I will try that out. Thanks for the help.

 

[skilroy]

No luck with assigning a point mass to the rigid body either unfortunately. Thanks for the help anyways.