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.
Different between modeling basic contact problem 2D and 3D
Posted Aug 27, 2017, 5:21 a.m. EDT Version 5.2a 2 Replies
Please login with a confirmed email address before reporting spam
Hi All,
I did model a basic contact problem (cylinder &plate) in 2D, and it worked.
However, when I try the same problem but in 3D model, it doesn't work even if I used the same procedure like boundaries conditions, mesh, contact,....
The error just shows the "The relative residual (0.018) is greater than the relative tolerance" which just appears many times when I have tried many other modification to boundaries conditions.
I am quite a new Comsol user, and I have spent days to find the solution but I haven't figure out the problem.
Attachmetn is my 2D because I cant upload 3D (size is large here) , but I did the same with 3D.
I am looking for any helps.
Thank you very much
Yen
I did model a basic contact problem (cylinder &plate) in 2D, and it worked.
However, when I try the same problem but in 3D model, it doesn't work even if I used the same procedure like boundaries conditions, mesh, contact,....
The error just shows the "The relative residual (0.018) is greater than the relative tolerance" which just appears many times when I have tried many other modification to boundaries conditions.
I am quite a new Comsol user, and I have spent days to find the solution but I haven't figure out the problem.
Attachmetn is my 2D because I cant upload 3D (size is large here) , but I did the same with 3D.
I am looking for any helps.
Thank you very much
Yen
2 Replies Last Post Aug 29, 2017, 4:13 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
7 years ago
Hi,
The most common mistake leading to non-convergence is that you have too few constraints to suppress all possible rigid body motions.
Note that if you remove mesh and solution from the model, you can reduce the file size significantly, and thus possibly be able to upload it.
Regards,
Henrik
The most common mistake leading to non-convergence is that you have too few constraints to suppress all possible rigid body motions.
Note that if you remove mesh and solution from the model, you can reduce the file size significantly, and thus possibly be able to upload it.
Regards,
Henrik
Please login with a confirmed email address before reporting spam
Posted:
7 years ago
Hi Henrik,
Thank you for your reply. Since the problem is symmetric (cylinder and plate), so I modeled half of them in 3D and set up properly boundary conditions. It runs successfully. There is no point load at the middle of the cylinder, so I just use edge load instead.
However, the stress and contact pressure are just so low as in pictures.
I don't really understand why the results are wrong, even though it runs successfully, and the mesh seems fine enough.
Thank you
Yen
Thank you for your reply. Since the problem is symmetric (cylinder and plate), so I modeled half of them in 3D and set up properly boundary conditions. It runs successfully. There is no point load at the middle of the cylinder, so I just use edge load instead.
However, the stress and contact pressure are just so low as in pictures.
I don't really understand why the results are wrong, even though it runs successfully, and the mesh seems fine enough.
Thank you
Yen
Attachments: