hyperelasticity -large deformations problem

[lscheerl]

I have the following problem:

I have a silicone material and I have the uni-axial tension test data results and his poisson ratio.
I filled these data in,in a hyperelastic material with Marlow.

I simulate a tension test on a dogbone to see wether my material act in Abaqus how it should react. The dimensions of the total dogbone are 150mm x 20mm x 1mm.I use Z-symmetry so my model has a thickness of 0,5mm.

My mesh: I use C3D8H elements

My step: I use NLGeom=on and my beginning increment=0.005

When I have strains up to 0,05, the stresses are OK but with higer strains , the stress is too high. The curve of my material model is a convex curve and the strain-stress curve I become with Abaqus is a concave curve.

I already tried the following things:
- I evaluated my hyperelastic material, it is stable for all strains and follows the test points
-I used other potential strain energy models like polynomial, N=2.
- I refined my mesh and have convergence for my results. So mesh is fine enough, I think.
- I used C3D20H elements en C3D8HR elements, the same results.
-I have no warnings that my elements are too heavily distorted.

Can you help me? I don't know what I can change to solve the problem.

 

[Dave442]

Hi,

I'm not sure how helpful this may be but are you inputting your experimental data and post-processing your numerical data correctly?

For hyperelastic materials ABAQUS requires nominal stress and nominal strain but outputs true stress and true strain. Maybe this is causing a discrepency?

Dave

 

[lscheerl]

Thanks Dave,

I forgot that Abaqus gave the true stress and strain.
Now it looks a little bit better

Lies