Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Problem with the meshing of a structure

Please login with a confirmed email address before reporting spam

I am trying to mesh a structure. I have a block and objects inside that block. The blocks gets meshed insanly coarse, while the objects inside get meshed rather fine. The block has elements that are in the size of it's diameter, this just cannot be. Where do I make a mistake.


10 Replies Last Post Feb 7, 2012, 4:10 a.m. EST

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 2, 2012, 3:43 p.m. EST
I downloaded your file and added a mesh 2. I just used a physics controlled mesh with normal element size. It meshed fine. Says it has 2428 elements. Are you trying to do something where you have extreme gradients? Sometimes you can use the default mesh and it works just fine.
I downloaded your file and added a mesh 2. I just used a physics controlled mesh with normal element size. It meshed fine. Says it has 2428 elements. Are you trying to do something where you have extreme gradients? Sometimes you can use the default mesh and it works just fine.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 2, 2012, 5:06 p.m. EST
I want a mesh with exatly symetric nodes on the boundaries, therefore all the mirroring. It worked with other structures, but here the mesh of the block is so coarse (elements spanning the whole diameter) that this mesh has to be broken. Also an external program that sums up the element volumes of all elements comes to a sum that is lower than the volume of the block.
I want a mesh with exatly symetric nodes on the boundaries, therefore all the mirroring. It worked with other structures, but here the mesh of the block is so coarse (elements spanning the whole diameter) that this mesh has to be broken. Also an external program that sums up the element volumes of all elements comes to a sum that is lower than the volume of the block.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 3, 2012, 2:17 a.m. EST
use scale variable option ... and which ever direction you need to make the mesh finer, please given an higher value than 1 .. just solves the issue !
use scale variable option ... and which ever direction you need to make the mesh finer, please given an higher value than 1 .. just solves the issue !

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 3, 2012, 5:24 a.m. EST
I used "Scale" if this is was you ment by "Scale variable". I managed to make the mesh finer by using small values (values larger than 1 only made the mesh coarser). Still my external program says me that the mesh does not fill the volume. Also the mesh starts to look very ugly (Extremly fine in places it does not need to be fine).
Did I misunderstand what you meant by "Scale variable"?
If not, how else can I prevent this mesh from getting broken, how can Comsol realise it has to mesh coarser in the middle of the bar.
I used "Scale" if this is was you ment by "Scale variable". I managed to make the mesh finer by using small values (values larger than 1 only made the mesh coarser). Still my external program says me that the mesh does not fill the volume. Also the mesh starts to look very ugly (Extremly fine in places it does not need to be fine). Did I misunderstand what you meant by "Scale variable"? If not, how else can I prevent this mesh from getting broken, how can Comsol realise it has to mesh coarser in the middle of the bar.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 4, 2012, 2:19 a.m. EST
Hi i guess this is what you wanted .. hopefully i am right !!

the objects inside cube are symmetric and the elements are coarser on the main cube !
Hi i guess this is what you wanted .. hopefully i am right !! the objects inside cube are symmetric and the elements are coarser on the main cube !


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 6, 2012, 4:11 a.m. EST
Thanks for your attempts to help me. The mesh of the structure you sent me seems to be correct. The problem is that I need a symetric mesh. I will subject this mesh to periodic boundary conditions. Every node on the surface has to be symetric with a node on the opposite side. By opposite side I mean a node in an XY plane has another node in the XY plane on the other side of the object, with identical XY coordinates, just differing in the z value. A node in the YZ plane has another node with identical YZ coordinates, just a different X coordinates and so on.
However thank you for what you sent me. I have a mesh that is symetric but broken, and you sent me a mesh that is not completly symetric but not broken, I will try to combine both together to get a symetric not broken mesh.
Thanks for your attempts to help me. The mesh of the structure you sent me seems to be correct. The problem is that I need a symetric mesh. I will subject this mesh to periodic boundary conditions. Every node on the surface has to be symetric with a node on the opposite side. By opposite side I mean a node in an XY plane has another node in the XY plane on the other side of the object, with identical XY coordinates, just differing in the z value. A node in the YZ plane has another node with identical YZ coordinates, just a different X coordinates and so on. However thank you for what you sent me. I have a mesh that is symetric but broken, and you sent me a mesh that is not completly symetric but not broken, I will try to combine both together to get a symetric not broken mesh.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 6, 2012, 5:34 a.m. EST

Thanks for your attempts to help me. The mesh of the structure you sent me seems to be correct. The problem is that I need a symetric mesh. I will subject this mesh to periodic boundary conditions. Every node on the surface has to be symetric with a node on the opposite side. By opposite side I mean a node in an XY plane has another node in the XY plane on the other side of the object, with identical XY coordinates, just differing in the z value. A node in the YZ plane has another node with identical YZ coordinates, just a different X coordinates and so on.
However thank you for what you sent me. I have a mesh that is symetric but broken, and you sent me a mesh that is not completly symetric but not broken, I will try to combine both together to get a symetric not broken mesh.


Its simple ... ' I have got an idea .. if u need symmetric mesh just divide the object into 4 by cutting it using two planes .. then mesh one volume and copy the same thing to the other four ... ! ..

first do what i did to the one fourth geometry and then copy domains ... !
[QUOTE] Thanks for your attempts to help me. The mesh of the structure you sent me seems to be correct. The problem is that I need a symetric mesh. I will subject this mesh to periodic boundary conditions. Every node on the surface has to be symetric with a node on the opposite side. By opposite side I mean a node in an XY plane has another node in the XY plane on the other side of the object, with identical XY coordinates, just differing in the z value. A node in the YZ plane has another node with identical YZ coordinates, just a different X coordinates and so on. However thank you for what you sent me. I have a mesh that is symetric but broken, and you sent me a mesh that is not completly symetric but not broken, I will try to combine both together to get a symetric not broken mesh. [/QUOTE] Its simple ... ' I have got an idea .. if u need symmetric mesh just divide the object into 4 by cutting it using two planes .. then mesh one volume and copy the same thing to the other four ... ! .. first do what i did to the one fourth geometry and then copy domains ... !

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 6, 2012, 7:50 a.m. EST
While I was trying it out I realized that restricting the maximum element size also solves the problem, thanks for your efforts to help me.
While I was trying it out I realized that restricting the maximum element size also solves the problem, thanks for your efforts to help me.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 6, 2012, 12:23 p.m. EST
@ all,
i have a problem regarding meshing a comsol model file.. before that i have another problem that is i want to upload .mph file for sake of my actual problem, but i cant upload the file showing an error that 'file extension error'..
please help..
thank you all..
@ all, i have a problem regarding meshing a comsol model file.. before that i have another problem that is i want to upload .mph file for sake of my actual problem, but i cant upload the file showing an error that 'file extension error'.. please help.. thank you all..

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 7, 2012, 4:10 a.m. EST
I never had this file extension error. Maybe you would be more succesful to get answers if you asked your question in a new thread.
I never had this file extension error. Maybe you would be more succesful to get answers if you asked your question in a new thread.

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.