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 between sphere and block
Posted 09.03.2015, 17:21 GMT-4 2 Replies
Please login with a confirmed email address before reporting spam
Hi,
Eventually i want to calculate the friction between a sphere and a magnet. To understand how contacts work in COMSOL i made two identical models, one which is 2D axisymmetric and the other one 3D. In the axisymmetric model i get the contact pairs to work but i can't reproduce these results in the 3D model although everything is the same. I get the following error:
Failed to find a solution.
Maximum number of segregated iterations reached.
In segregated group 1:
There was an error message from the linear solver.
The relative error (1e+003) is greater than the relative tolerance.
Returned solution is not converged.
Is there any difference between axisymmetric and 3D that i am forgetting?
I attached both the models.
Thanks in advance,
Janeau
Eventually i want to calculate the friction between a sphere and a magnet. To understand how contacts work in COMSOL i made two identical models, one which is 2D axisymmetric and the other one 3D. In the axisymmetric model i get the contact pairs to work but i can't reproduce these results in the 3D model although everything is the same. I get the following error:
Failed to find a solution.
Maximum number of segregated iterations reached.
In segregated group 1:
There was an error message from the linear solver.
The relative error (1e+003) is greater than the relative tolerance.
Returned solution is not converged.
Is there any difference between axisymmetric and 3D that i am forgetting?
I attached both the models.
Thanks in advance,
Janeau
Attachments:
2 Replies Last Post 29.09.2016, 11:07 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Dear Janeau,
In 2D axisymmetric you fix your model on the central axis. In 3D that would be equivalent to fixing the model in the x and y axis, on the center line.
An alternative to this is to use a sector symmetry with roller boundaries on the symmetry.
For the rest, make sure that work with similar meshes as well.
On a sidenote, generally it is more stable when you use a prescribed displacement on the top, this allows less freedom on the object. Other means of getting a more stable model is to add a spring foundation, and reducing the stiffness using an auxiliary sweep.
Best regards,
Frank
In 2D axisymmetric you fix your model on the central axis. In 3D that would be equivalent to fixing the model in the x and y axis, on the center line.
An alternative to this is to use a sector symmetry with roller boundaries on the symmetry.
For the rest, make sure that work with similar meshes as well.
On a sidenote, generally it is more stable when you use a prescribed displacement on the top, this allows less freedom on the object. Other means of getting a more stable model is to add a spring foundation, and reducing the stiffness using an auxiliary sweep.
Best regards,
Frank
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Dear Frank,
Thanks for your advice on 3D two spheres contacting. I get confused with one of your suggestion.
"An alternative to this is to use a sector symmetry with roller boundaries on the symmetry."
How to manage this operation in 3D model?
Another thing is, even though prescribed displacement node is a good strategy to converge the model, at most of time the boundary condition on top of upper sphere is loading force or stress loading, especially point loading. In this case, how to make the model convergent.
Thanks,
Zeng
Thanks for your advice on 3D two spheres contacting. I get confused with one of your suggestion.
"An alternative to this is to use a sector symmetry with roller boundaries on the symmetry."
How to manage this operation in 3D model?
Another thing is, even though prescribed displacement node is a good strategy to converge the model, at most of time the boundary condition on top of upper sphere is loading force or stress loading, especially point loading. In this case, how to make the model convergent.
Thanks,
Zeng