Strange penetration in Contact analysis

I do a small problem in Marc contact analysis, the result shows a strange penetration in some regions. I used Segment to Segment. Does anyone have the answers or know how to solve this penetration except using remesh option?

Thanks,

the pics and .dat file are attached

Attached Files (3)

Contact_elenum1920_job1.zip

0 Kiranraj Shetty over 4 years ago

Hi Denzel Li,

You can avoid penetration without remeshing, however the solution will terminate due element distortion, Exit #1009.

Regards,

Kiran

Attached Files (1)

check.mud.zip
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 denzel.li over 4 years ago

Thanks for reply.

I run the check.mud, but the result still seems strange, as you see , the left element slide away horizontal from the boundary and the right element shows a high penetration with the boundary. Is it reasonable or acceptable for a calculation?

Attached Files (1)
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Kiranraj Shetty over 4 years ago

Hi,

In the dat file I read into Mentat 2019 FP1, There are no boundary conditions.T here are 3 contact bodis, among which body "move" has velocity defined in Y direction, and there is no table used.

When I check my results from the mud file I shared earlier, I do not see any penetration as shown in attached image ( True scale plot).

Regards,

Kiran

Attached Files (1)
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 denzel.li over 4 years ago

Yes, the model has no boundary conditions and only 3 contact bodies.

I rerun the check.mud by Mentat2020, get a similar result with the pic you show. Before I used Mentat2019. Thanks and I will compared the different settings in the check.mud
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 denzel.li over 4 years ago

Hello There is another problem, I change the boundary from a curve to element, and used the Segment to Segment contact, there is a huge penetration in the contact region, but the penetration didnot occurs when use Node to Segment method.

Attach is the data file. by the way, the data version is generated by Mentat2020

Attached Files (1)

Contact_0.05_job1.dat
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 denzel.li over 4 years ago
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 Kiranraj Shetty over 4 years ago

Hi

The material is very soft with E=10, Hence the default penalty factor for S2S contact might need to be increased. Node to seg is more robust for such scenarios.

You can log a support ticket for investigating the model.

Regards,

Kiran
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel
0 denzel.li over 4 years ago

Thanks for reply. I will try it.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Verify Answer

Cancel