CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ANSYS Meshing] Edge Sizing Problems (https://www.cfd-online.com/Forums/ansys-meshing/98613-edge-sizing-problems.html)

 nickdaish March 14, 2012 15:40

Edge Sizing Problems

1 Attachment(s)
I have been trying to use multiple edge sizing controls on a single cuboidal body to constrain the mesh and generate one that is wholly structured and consisting entirely of orthogonal hexahedral cells. The edge sizing controls use bias factors. My problem is that the behavior is unpredictable: sometimes it gives a totally regular grid, while other times it gives one that is irregular in one dimension - see attachment.

I use edge sizing controls on all the edges; in some cases, because I am using one-sided bias factors, I have to use two edge controls to cover all the edges in a given coordinate direction because, for example, the 4 edges in the x-direction have senses (assigned by DesignModeler) that are not all in the positive x-direction.

I have experimented with the above single block, and have found the appearance of irregular cells seems to be principally associated with the shape of the block, rather than the number of divisions or the type of bias factor. I have found that, for this single block, problems start when the y-dimension is less than or equal to exactly 0.5 times the x-dimension, although bizarrely the z-dimension doesn't seem to matter. I don't know whether the 0.5 value is significant, but it does seem to be exact.

I realise that there are probably some background mesh parameters floating around, even though I've turned Advanced Size Function to Off, but I thought that the explicit edge sizings would override these. I also realise that when a number of adjoining bodies are meshed together then they may "help" each other by mapping from faces already meshed - I have meshed collections of these blocks (see below) and although some are very long and thin they were nevertheless meshed correctly.

The reason I have been looking at this is because I have been trying to build a structured grid out of collections of cuboidal bodies like the one above. I thought I had managed to crack the problem with this approach, but this uncertainty makes the whole enterprise potentially futile.

Any help in explaining why this is happening and what I could do to prevent it would be greatly appreciated!

Thanks

 tauqirnawaz March 15, 2012 07:57

1 Attachment(s)
Is this something that you are looking for?
http://www.cfd-online.com/Forums/att...1&d=1331812635

 nickdaish March 15, 2012 12:44

Resolved problem for multiple bodies

Hi

I have managed to resolve the problem in the context of meshing a contiguous collection of cuboidal bodies, sub-divided into parts, by applying Multizone controls to each part: this seems to constrain the grid within each part, which is what I wanted. Nevertheless I am still wondering why this should occur at the single-body level. Perhaps we shall never know...

 tauqirnawaz March 16, 2012 09:49

Quote:
 Originally Posted by nickdaish (Post 349667) Hi I have managed to resolve the problem in the context of meshing a contiguous collection of cuboidal bodies, sub-divided into parts, by applying Multizone controls to each part: this seems to constrain the grid within each part, which is what I wanted. Nevertheless I am still wondering why this should occur at the single-body level. Perhaps we shall never know...
:) Problem occurs on single body because in the edge definition behaviour setting if you have "Soft" then mesher would sometime not follow your settings. If you simply change the behaviour to "Hard", that should solve your problem. Any ways best of luck with your computations.

 nickdaish March 16, 2012 10:52

Hard sizing isn't enough

Hi there

Thanks for replying. I was aware of the soft/hard behavior parameter, and in fact all my sizing controls were hard - hence my surprise that they were not obeyed. I can only think that there is something else going on in the background - some global parameters that take precedence in some (but not all) circumstances.

 tauqirnawaz March 19, 2012 05:34

Quote:
 Originally Posted by nickdaish (Post 349858) Hi there Thanks for replying. I was aware of the soft/hard behavior parameter, and in fact all my sizing controls were hard - hence my surprise that they were not obeyed. I can only think that there is something else going on in the background - some global parameters that take precedence in some (but not all) circumstances.
Well I think there is certainly some issue with your set up; I do not see any reason to agree with your "background issues".:D

 asurpam August 20, 2012 13:19

Hi,

I am facing the same problem. Did you find the solution to it ?
Please pass me the information if you found it.

Regards,
Ayushi

 All times are GMT -4. The time now is 01:00.