Marlow Model
Marlow Model
(OP)
Hello,
I have hyperelastic test data that has fit perfectly with a Marlow Model in abaqus. But for some reason when I use a Tet mesh my simulations will not converge. However, when I mesh the structure with a HEX mesh, it converges when I use "advanced hourglass control". Is there any way I can get a Tet mesh to work with the marlow model? is there a similar feature that accounts for hourglassing with hyperelasticity for a tetrahedral mesh element?
Thanks a bunch!
-Alex
I have hyperelastic test data that has fit perfectly with a Marlow Model in abaqus. But for some reason when I use a Tet mesh my simulations will not converge. However, when I mesh the structure with a HEX mesh, it converges when I use "advanced hourglass control". Is there any way I can get a Tet mesh to work with the marlow model? is there a similar feature that accounts for hourglassing with hyperelasticity for a tetrahedral mesh element?
Thanks a bunch!
-Alex





RE: Marlow Model
Are you using first-order or second-order tets? First-order tets are fully integrated and therefore don't hourglass, but they don't resolve the stress very well and can cause lack of convergence. The "regular" second-order tets (C3D10 or C3D10H) are more accurate and don't hourglass, but can cause problems with shear or volumetric locking or hard contact. The other option is the "modified" second-order tets (C3D10M and C3D10MH) which fix the contact and locking problems but can hourglass. I would try these alternatives and see if any work.
In any case make sure you're using hybrid (...H) elements for this material.
www.invariantlabs.com
RE: Marlow Model
Thanks so much. I had actually just talked with a colleague and this was exactly what he had said. My problem was that I was using linear tets instead of quadratic (second order). Now my models run beautifully. I will most likely need to account for contact in the future, so I will look into the modified second order tets for that. I really appreciate the feedback.
Thanks again.
Sincerely,
-Alex