Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Contact problem. Static vs Dynamic

Please login with a confirmed email address before reporting spam

I have an impact model; steel presses down soft tissue.

I can solve this model in static analysis with prescribed displacement.

But when I solved the model in dynamic analysis with prescribed velocity, penetration occured.

Besides prescribed movement and solver, everythings are the same.

I do not understand why the model cannot be solved in dynamic analysis.

Actually error does not occured but there shouldn't be penetration.

I attached two files,

0-hyper.mph - static analysis
00-hyper.mph - dynamic analysis

Any help will be appreciated.

Thanks,


Jay


9 Replies Last Post Jun 17, 2013, 9:46 p.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 30, 2013, 4:25 p.m. EDT
I saw in document that slave boundary cannot penetrate master boundary, but it does.

hmmmmm....
I saw in document that slave boundary cannot penetrate master boundary, but it does. hmmmmm....

Jim Freels mechanical side of nuclear engineering, multiphysics analysis, COMSOL specialist

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 30, 2013, 11:03 p.m. EDT
I have not looked at your model. Have you set up contact pairs ? Of course, to do that, each part in contact must be part of an assembly of parts rather than a union (default) in the final step of the geometry creation. Then, the contacts will not cross boundaries. It also becomes much more difficult, and tricky, to solve; but, it does work quite well.
I have not looked at your model. Have you set up contact pairs ? Of course, to do that, each part in contact must be part of an assembly of parts rather than a union (default) in the final step of the geometry creation. Then, the contacts will not cross boundaries. It also becomes much more difficult, and tricky, to solve; but, it does work quite well.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 31, 2013, 12:12 a.m. EDT
I did. There is contact pair and they are part of assembly.
I looked at forum and example files a lot, and I think I did everything I am supposed to do.
I did. There is contact pair and they are part of assembly. I looked at forum and example files a lot, and I think I did everything I am supposed to do.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 12, 2013, 12:54 p.m. EDT
I think the contact boundaries should be only the 2 top surfaces, not the lateral faces of block number 1
I think the contact boundaries should be only the 2 top surfaces, not the lateral faces of block number 1

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 12, 2013, 1:05 p.m. EDT
When I don't include lateral faces, penetration occurs.

Do you have any ideas besides contact boundaries?
When I don't include lateral faces, penetration occurs. Do you have any ideas besides contact boundaries?

Nagi Elabbasi Facebook Reality Labs

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 14, 2013, 8:57 a.m. EDT
Hi Jay,

I noticed that the 7 m/s velocity you applied is too high. It is much higher than the wave speed through the lower block (around 3 m/s). Probably Young’s modulus for that block (9 kPa) is incorrect.

Nagi Elabbasi
Veryst Engineering
Hi Jay, I noticed that the 7 m/s velocity you applied is too high. It is much higher than the wave speed through the lower block (around 3 m/s). Probably Young’s modulus for that block (9 kPa) is incorrect. Nagi Elabbasi Veryst Engineering

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 17, 2013, 4:49 a.m. EDT
There are some general problems with your model.
1) Lateral faces are not contact surfaces, so you don't have to include them.

2) Mesh in contact models is always a problems. I'd use a swept mesh + boundary layer to have finer mesh where it is needed (at least in destination boundaries only). A coarser mesh can cause numerical problems at contact boundaries

3) Too high prescribed velocity, try to reduce it. I can't try to solve your model right now, so i can't say if this is enough or if there are other problems

A question, why simmetry in 2 later faces?

Try also to use identity pair under "Geometry form assembly" and contact pair under "definitions" (I never really understood the difference between this solution and contact pair under "Geometry, form assembly" + contact pair under "definitions", so perhaps it's a stupid suggestion)

Let me know if it works

Stefano
There are some general problems with your model. 1) Lateral faces are not contact surfaces, so you don't have to include them. 2) Mesh in contact models is always a problems. I'd use a swept mesh + boundary layer to have finer mesh where it is needed (at least in destination boundaries only). A coarser mesh can cause numerical problems at contact boundaries 3) Too high prescribed velocity, try to reduce it. I can't try to solve your model right now, so i can't say if this is enough or if there are other problems A question, why simmetry in 2 later faces? Try also to use identity pair under "Geometry form assembly" and contact pair under "definitions" (I never really understood the difference between this solution and contact pair under "Geometry, form assembly" + contact pair under "definitions", so perhaps it's a stupid suggestion) Let me know if it works Stefano

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 17, 2013, 5:24 a.m. EDT


Try also to use identity pair under "Geometry form assembly" and contact pair under "definitions" (I never really understood the difference between this solution and contact pair under "Geometry, form assembly" + contact pair under "definitions", so perhaps it's a stupid suggestion)



The pair type that you select under "Geometry form assembly" is the type of pairs that will be automatically generated based on geometrical proximity. They will then appear as such under 'Definitions' where the type can be changed if necessary.

The idea is that you select a type under "Geometry form assembly" which is suitable for the majority of your pairs, and then switch the type of the others under Definitions. It is only the type a pair has under 'Definitions' which determines whether it can be used in 'Contact' in Solid Mechanics or not.

Regards,
Henrik
[QUOTE] Try also to use identity pair under "Geometry form assembly" and contact pair under "definitions" (I never really understood the difference between this solution and contact pair under "Geometry, form assembly" + contact pair under "definitions", so perhaps it's a stupid suggestion) [/QUOTE] The pair type that you select under "Geometry form assembly" is the type of pairs that will be automatically generated based on geometrical proximity. They will then appear as such under 'Definitions' where the type can be changed if necessary. The idea is that you select a type under "Geometry form assembly" which is suitable for the majority of your pairs, and then switch the type of the others under Definitions. It is only the type a pair has under 'Definitions' which determines whether it can be used in 'Contact' in Solid Mechanics or not. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 17, 2013, 9:46 p.m. EDT
My results look like attached file.

Even if I decrease velocity, contact part is torn off.

Any one has a idea?
My results look like attached file. Even if I decrease velocity, contact part is torn off. Any one has a idea?

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.