Hyperelastic material properties

[Anonymous]

Hi Team

I was trying to input the material properties from the compression test and it said that for compression test the stress values should be negative. However, when I enter the negative values of stress the error comes up saying Singularity detected.

Any help would be appreciated.

Thanks.

Regards,

Tarun 

[Anonymous]

Hi

I have attached the model setup, the table for material data entry and the output file for the same in this post.

Also when I increase the number of bisections limit to 100 the computation keeps going and won't progress after the 6 iteration. It keeps coming back from the 6th iteration to 1 and the loop goes on forever.

Thanks.

Regards,

Tarun

[John_Holtz]

Hi @Anonymous 

I think stress and strain need to be negative for compression 

Otherwise, there is nothing useful in the static images or log file you provided. You have the model: what is going on with the results?

I suggest that you output the results from all time steps. ("Nonlinear Setup  1 > Intermediate Results > On".) Has the strain exceeded your stress-strain curve (and therefore is it "extrapolating" to what?). Is something else going wrong that would explain why it cannot continue beyond 80%? Is there a hollow space in your model that you image doesn't show, and that is causing problems? 

[Anonymous]

Hi @John_Holtz 

Thanks for coming back to me. 

The geometry is 1/8 of the model and I have applied symmetry conditions to all sides.

Using the negative data it was showing me singularity error. 

Well, I don't completely understand what you are saying. I am attaching the model and the data for further investigation.

Regards,

Tarun

[John_Holtz · Accepted solution]

Hi @Anonymous 

I think you should talk with a hyperelastic material expert, someone who has practical experience. But I do not know who that would be. The questions that come to my mind about your analysis are things like this:

1. The calculated stress (at 81.5% load) is 41.788 MPa. The stress-strain data goes up to 17 MPa. What data is it using to get to 41 MPa?
2. I think Mooney Rivlin creates a parabolic stress-strain curve from the material data. The Neo-Hookean model that you are using may be parabolic, or maybe just a straight light. I cannot image that either one is a good fit to your data! I think you need to use a higher order sub type (Yeoh? Polynomial?).
3. You only have simple tension/compression data. That may be sufficient for the simple tension test you are doing now, but I assume that you want to do something that is more complex. Once the analysis starts to have shear and other effects, I think you are going to need to have additional material data.

[Anonymous]

Hi @John_Holtz 

Thank you for the feedback. I'll investigate the issue further.

Regards,

Tarun