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.

Re: Moving Boundaries Problem (ALE) / Inverted Mesh Warnings

Please login with a confirmed email address before reporting spam

Hi,

I am working on a 1-D moving boundaries problem using ALE application mode. In this model, two boundaries are moving on one side with different velocities.

Let me explain clearly with number. Consider 3 boundaries, where the first boundary is fixed. Second and third boundaries are moving towards the first. Here, the third boundary is moving with a higher velocity than the second one. So, I set up a if condition such that it can use it's own velocity until it hits the second boundary and then the third boundary must use the velocity of the second one. So that both boundaries move together.

When i tried to model the above mentioned problem, I got inverted mesh warnings. I have tried several things...for example, allowing a definite (small) distance between the second boundary and the third one. so that i can get rid of inverted mesh warnings. but unsucessful.

Can i build a model in COMSOL where two domains collapse into one after a certain time?


Any suggestions/comments are greatly appreciated.

Thank you,

Manohar

3 Replies Last Post Feb 5, 2011, 6:47 a.m. EST
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 5, 2011, 4:41 a.m. EST
Hi

One advice: be sure that the "third" point (boundary) NEWVER comes all the way up to the second, because then you remove a domain (=line in 1D) and that is a topological change that COMSOL cannot cope with as the domain id's are renumbered etc. Normally the ALE should work OK so long you do not change the topology, so leave this last line lengt ther but with a "small" value.

Still you might need to tweak a little the ALE settings to help the system to have a regular contraction of all nodes along the line.

For me I do expect your model to work well, if you respct the point above

--
Good luck
Ivar
Hi One advice: be sure that the "third" point (boundary) NEWVER comes all the way up to the second, because then you remove a domain (=line in 1D) and that is a topological change that COMSOL cannot cope with as the domain id's are renumbered etc. Normally the ALE should work OK so long you do not change the topology, so leave this last line lengt ther but with a "small" value. Still you might need to tweak a little the ALE settings to help the system to have a regular contraction of all nodes along the line. For me I do expect your model to work well, if you respct the point above -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 5, 2011, 4:51 a.m. EST
Hi Ivar,

Thanks for your reply. I actually tried the same thing by leaving a small distance between the second and the third one, but still i am getting a lot of inverted mesh warnings.

Sorry i forgot to mention that initially there is a 4 orders of magnitude difference between the two domains. I just want to see how a reactant particle converts into product. So initially i used a small 2nd domain (i.e. products).

I meshed differently with interactive meshing to see if it works. But even that didn't helped.

Any ideas on how can i model this problem with out having any inverted mesh warnings.

Thanks
Hi Ivar, Thanks for your reply. I actually tried the same thing by leaving a small distance between the second and the third one, but still i am getting a lot of inverted mesh warnings. Sorry i forgot to mention that initially there is a 4 orders of magnitude difference between the two domains. I just want to see how a reactant particle converts into product. So initially i used a small 2nd domain (i.e. products). I meshed differently with interactive meshing to see if it works. But even that didn't helped. Any ideas on how can i model this problem with out having any inverted mesh warnings. Thanks

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 5, 2011, 6:47 a.m. EST
Hi

yuo have the "remeshing" option as one thing, i.e. check the mesh size and remesh or remeah systematically

but also the ALE mesh laws: "You can choose between Laplace smoothing, Winslow smoothing, and hyperelastic smoothing"

Not sure which one to propose though

--
Good luck
Ivar
Hi yuo have the "remeshing" option as one thing, i.e. check the mesh size and remesh or remeah systematically but also the ALE mesh laws: "You can choose between Laplace smoothing, Winslow smoothing, and hyperelastic smoothing" Not sure which one to propose though -- Good luck Ivar

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.