Menu

Discussion Forum

New Discussion
Filtern

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.

LARGE DEFORMATION ALE (moving mesh elements)

Posted 10.04.2011, 05:57 GMT-4 6 Replies

Giulio Ferrazzi
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I everyone,

I'm solving a 2D contact mechanic problem in the large deformation range. Specifically I have two bodies where the first of the "enters" in the second with a prescribed displacement. The study is parametric so I solve the problem for an increasing value of the initial prescribed dislacement.

Since the second body is 1000 times softer than the first one, I have problems of convergence when the mesh deforms too much. I'm am trying to use ALE but it seems that I'm not able to set the boudary conditions. I have imposed the same prescribed dispacement for the mesh that belongs to the soft body, but I don't get suitable results.

Here I attached the file of the model. Is everything ok? Thanks

Giulio

Attachments:

6 Replies Last Post 09.05.2012, 09:01 GMT-4
Amer Malik
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 10.04.2011, 13:22 GMT-4
I haven't looked at your file yet, but did you define the Contact pair between the right boundaries. By the looks of it the contact isn't defined properly and the bodies are penetrating into each other, hence causing the mesh to deform too much
I haven't looked at your file yet, but did you define the Contact pair between the right boundaries. By the looks of it the contact isn't defined properly and the bodies are penetrating into each other, hence causing the mesh to deform too much
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 10.04.2011, 17:56 GMT-4
Hi

Normally you apply a displacement or a force on one of the solid bodies, and prescribe the displacements of the ALE to follow the u,v of the deformed solids.

You are imposing a displacement on part of the ALE, on top of the contact. And do not forget that the symmetry condition in the solids must also have an equivalent in the ALE (there is no symmetry, you must block the x motion in the symmetry plane in your case

Check out a few of the examples of large deformations and contacts in the model library. Then I'm sure you will get the BC better defined

By the way, why not drop the ALE to start with and look at the deformation with a scale of "1". I do not really see why you need it, at least until you get your contact case running. Another hint use small steps such as a range(0,0.1,10)


--
Good luck
Ivar
Hi Normally you apply a displacement or a force on one of the solid bodies, and prescribe the displacements of the ALE to follow the u,v of the deformed solids. You are imposing a displacement on part of the ALE, on top of the contact. And do not forget that the symmetry condition in the solids must also have an equivalent in the ALE (there is no symmetry, you must block the x motion in the symmetry plane in your case Check out a few of the examples of large deformations and contacts in the model library. Then I'm sure you will get the BC better defined By the way, why not drop the ALE to start with and look at the deformation with a scale of "1". I do not really see why you need it, at least until you get your contact case running. Another hint use small steps such as a range(0,0.1,10) -- Good luck Ivar
Giulio Ferrazzi
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.04.2011, 05:49 GMT-4
Thank You Ivar,

do you know if, in order to find it´s better to use a small range step?

Did COMSOL use, to find out the solution at steo i, the solution at the step i-1 ?
Thank You Ivar, do you know if, in order to find it´s better to use a small range step? Did COMSOL use, to find out the solution at steo i, the solution at the step i-1 ?
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.04.2011, 09:08 GMT-4
Hi

its always better to use "small" steps to allow the solver to follow its optimum, but not smaller than required, that means that sometimes 1 step is enough ;)

--
Good luck
Ivar
Hi its always better to use "small" steps to allow the solver to follow its optimum, but not smaller than required, that means that sometimes 1 step is enough ;) -- Good luck Ivar
Sanket Kelkar
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12.10.2011, 13:58 GMT-4
Can anybody send me a very basic implementation of ALE like shrinking of a circle or square with time. I am trying to implement ALE without any physics (for simplification). But I am not seeing the desired result. I just get back the initial geometry. I don't know what mistake I am making.

Thanks
Sanket
Can anybody send me a very basic implementation of ALE like shrinking of a circle or square with time. I am trying to implement ALE without any physics (for simplification). But I am not seeing the desired result. I just get back the initial geometry. I don't know what mistake I am making. Thanks Sanket
Djuno Mbm
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 09.05.2012, 09:01 GMT-4
Hello Giulio,

I tried to run your .mph file but i don't see any convergence problems ! ! ! Are you worried because the total displacement computed of 0.01 [µm] is much bigger than the prescribed one of 4.10^-7 x 1[nm]???

According to what Ivar told you, what kind of changes did you make to make it work better? could you upload the final file please?

Best regards,
Hello Giulio, I tried to run your .mph file but i don't see any convergence problems ! ! ! Are you worried because the total displacement computed of 0.01 [µm] is much bigger than the prescribed one of 4.10^-7 x 1[nm]??? According to what Ivar told you, what kind of changes did you make to make it work better? could you upload the final file please? Best regards,

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.

Suggested Content