CFD Online Logo CFD Online URL
Home > Forums > ANSYS Meshing & Geometry

[ICEM] Splitting the blocks

Register Blogs Members List Search Today's Posts Mark Forums Read

LinkBack Thread Tools Display Modes
Old   March 28, 2011, 08:50
Default Splitting the blocks
Senior Member
Join Date: Oct 2010
Posts: 303
Rep Power: 9
saisanthoshm88 is on a distinguished road
Iíve just started exploring Hex meshing with ICEM CFD and I had a question regarding the split up of blocks. In ICEM I see that a split can be confined to a particular block alone instead of running it through all the blocks. But I learnt that this can lead to problems with the meshing later on and that it is always preferred to have continuous splits.

So can someone please elucidate the general points to bear in mind while making a split , with some images

Thanks in advance
Best regards,
saisanthoshm88 is offline   Reply With Quote

Old   March 29, 2011, 09:09
Default Propagate splits...
Senior Member
PSYMN's Avatar
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,660
Blog Entries: 1
Rep Power: 36
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
Sorry, no time for pics, but I can give the jist of it.

When you make a split thru a single block, you have a number of nodes on either side of the split edge. in order to have a pure "structured" hexa mesh, these must propagate thru even if the split does not. Therefore, if you split a single block, the adjacent blocks get implied splits... You can't see these splits, but if you go to the split edge and adjust the number of nodes, you will see that this must propagate thru. The problem is that without the split, it is harder to control the distribution on the other edges. It can be done with "link edges", but it is a hassle.

More importantly, if you later split one of these "unsplit" edges, the new split also has implied splits that must go back past other real splits and implied splits. This can cause quite a control mess.

My suggestion is that it is usually better to split thru all the blocks so you can retain this control. The down side is visual clutter, but this can be countered with other means such as turning of surfaces if you don't need them, reducing what you are looking at with index control (don't forget that index control can limit splits), etc. The other issue is that structured solver users may want fewer blocks at the end; this can be handled with merge output blocks after you have sorted out all your edge distributions, etc. This "output blocks" option allows you to keep the control of the edges, but merges them on output.

Have fun with it.
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   March 29, 2011, 09:17
Senior Member
Join Date: Sep 2009
Location: France
Posts: 323
Rep Power: 13
BrolY will become famous soon enough
Here are some images to illustrate what Simon said.

SPLIT_BLOCK_geom.jpg is the geometry (as a simple example)
SPLIT_BLOCK_mesh_3.jpg is the result of the mesh. If you need to refine the mesh at the middle of the geometry, you can see the number of nodes of your middle edge is propagating to the top and bottom edges.
SPLIT_BLOCK_mesh_5.jpg is the same mesh than the previous one but with the linked bunching option activated. So the mesh is more uniform.

But imagine that with more splits and edges, and you can guess how difficult it would be to have an uniform and nice mesh.
So, as SImon said, it's better to propagate your splits everywhere, even if it would require more associations (so you would need to create more curves and points to fully associate your blocking).

Hope it helps,
Good luck !
Attached Images
File Type: jpg SPLIT_BLOCK_geom.jpg (43.4 KB, 97 views)
File Type: jpg SPLIT_BLOCK_mesh_3.jpg (75.1 KB, 95 views)
File Type: jpg SPLIT_BLOCK_mesh_5_linkedbunching.jpg (81.1 KB, 84 views)
BrolY is offline   Reply With Quote

Old   March 30, 2011, 07:40
Senior Member
Join Date: Oct 2010
Posts: 303
Rep Power: 9
saisanthoshm88 is on a distinguished road
Thanks for the replies Simon and Boles. Simon, Can you please share some ppt that covers Advanced hex meshing techniques , tips and tricks for hex meshing. I've done a search on the customer portal but couldn't find anything as such (or) if you think that I missed out something from the portal , Please suggest me upon finding it

And can you tell me the acceptable quality range for pyramids in ICEM CFD if the solver is either CFX (or) Fluent
Best regards,

Last edited by saisanthoshm88; March 30, 2011 at 11:51.
saisanthoshm88 is offline   Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

Similar Threads
Thread Thread Starter Forum Replies Last Post
dsmcInitialise - dsmcFoam archymedes OpenFOAM Pre-Processing 93 February 11, 2014 02:22
fluent3DMeshToFoam bego OpenFOAM 27 May 29, 2013 13:08
[ICEM] Splitting a block along a stretched egdge vnakul79 ANSYS Meshing & Geometry 2 March 29, 2012 08:27
Meshing multiple blocks Hanno OpenFOAM Native Meshers: snappyHexMesh and Others 3 February 17, 2011 13:14
Merging the blocks in ICEM saisanthoshm88 ANSYS Meshing & Geometry 1 December 27, 2010 11:14

All times are GMT -4. The time now is 18:14.