Simulation of an indentation test in Abaqus.

[ShiningEngineer]

Hello, I have already been able to simulate a compression test over the entire surface with my model in Abaqus and the test data fits.

Now I would like to simulate an indentation (local) compression test, but my simulation terminates prematurely.

I am using a self-created and validated HyperFoam material model and the iterative solver as I have non-linear deformation properties and large deformations. I also tried to do the contact definition in the initial step via self-contact (default), but this is not possible due to my discrete rigid stamp, so I currently used the Abaqus function "Contact Pairs".

I use C3D8R as the element of my rectangular body. My bottom side is fixed in U3=0, and I transfer the forced deformation via my reference point (RP) with U3=12.

I would be grateful for any suggestions, regards.

https://files.engineering.com/getfi...689-47fb-906a-51b740be737b&file=question1.png

 

[rb1957]

please, terminology is important. I don't think you're talking of a "compression" test, but rather an Impact test.

Ok, you may have a model and applied a distributed compression load to it. But when you talk of an "indentation" test, this is (I think) an impact test.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[ShiningEngineer]

Hello,
I am talking about an indentation test according to ISO 2439, in which a cylindrical punch is inserted into the part of a total surface up to a deformation of 40% of the initial height.

I think my model represents this case, or am I wrong? Thanks for the input!

 

[ShiningEngineer]

The compression test according to iso 2439 is an extended compression test. My choice of words may have caused confusion.

 

[FEA way]

At what point does it diverge ? What are the stresses and strains then ? It would be better to fix all the DOFs for the sample and all apart from U3 for the indenter. Also, dynamic implicit quasi-static (or maybe even explicit if you can’t get rid of the convergence issues) step would be recommended. Finally, general contact should be used in most cases.

 

[rb1957]

NP ... my ignorance of your specific testing.

The indentation hardness of flexible cellular materials is a measure of their load-bearing properties. ISO 2439:2008 specifies four methods (A to D) for the determination of indentation hardness and one method (E) for determination of compressive deflection coefficient and hysteresis loss rate of flexible cellular materials. Annex A provides a summary of test parameters and typical force-indentation graphs obtained with these methods.

These five methods are applicable only to latex foam, urethane foam and PVC foam of the open-cell type.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[ShiningEngineer]

Hello,

thank you for your input. I have fixed all DOFs except for U3 for the forced deformation.

I am currently still using Static, General. I am not sure if I have defined my contacts correctly.

The cylinder is a discrete rigid body and the rectangle is my model to be compressed by the forced deformation through the RP (of the cylinder). My rectangle is composed of four materials that I have already been able to validate. Due to material grading, the stiffness increases from the surface to the lower surface. The data comes from experimental test data and is stored in the material models.

If I perform a compression over the entire surface of the rectangle with a corresponding stamp, the same test can be reproduced in reality. But as soon as I use a cylindrical body stamp, there are significant deviations as the results show. I think the error lies in my contact conditions. I have also tried tetrahedron elements (linear/square), with the same result.

Unfortunately, it is not possible to set contact conditions via General Contact, as I always get the comment from Abaqus with double-sided surfaces.

The contact pairs option causes my simulation to stop at about 20% deformation. I have varied the coefficient of friction between 0.01 and 0.6 without success.

If the coefficient of friction is too low, the simulation terminates prematurely. Perhaps someone has a tip for me on how to visualise the contacts differently?

https://files.engineering.com/getfi...01f-817f-2c73a3dda1a8&file=question_forum.png