CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

Octree Grid Refinement

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

Reply
 
LinkBack Thread Tools Display Modes
Old   September 24, 2008, 14:38
Default Octree Grid Refinement
  #1
Tristan
Guest
 
Posts: n/a
I've been playing around with ICEM to make a tet mesh. I have a global maximum element size of 3.5 mm and then I specify a smaller maximum element size on some parts where I need local grid refinement. Here's a sequence of (part maximum element size-total number of elements) pairs for my domain (global maximum element size is always 3.5 mm):

(1.5 - 3,298,490); (1.0 - 11,166,651); (0.875 - 2,2465,04); (0.7 - 4,293,775); (0.6 - 6,766,108); (0.4375 - 2,811,941)

This doesn't make a whole lot of sense. Clearly, powers of 2 in the ratio between the global max element size and the part max element size are golden. If I don't use a power of two then I see grid refinement throughout the whole domain instead of just in the region near the parts with the smaller maximum element size. Is there a general rule here for Octree meshing that you should always make your local grid refinements in powers of 2?

Thanks,

Tristan
  Reply With Quote

Old   September 25, 2008, 09:39
Default Re: Octree Grid Refinement
  #2
rohit
Guest
 
Posts: n/a
I have not exactly understand your problem, but i can make out that you need local refinement in some areas. Try using "mesh density" for local refinements..it is available under the mesh tab.

  Reply With Quote

Old   September 25, 2008, 13:13
Default Re: Octree Grid Refinement
  #3
Tristan
Guest
 
Posts: n/a
Rohit,

The problem is that as I decrease the maximum size on the parts where I need grid refinement, the total number of elements in my domain does not increase monotonically. When I set the maximum size on some of my parts (surfaces where I need a finer grid) to 1 mm (global maximum size is 3.5 mm), I end up with over 11 million total elements. However, if I set the maximum part size slightly smaller to 0.875 mm (which is exactly one fourth of the global maximum size) then I get just over 2 million elements.

I'll take a look at using a mesh density.

Tristan
  Reply With Quote

Old   September 26, 2008, 05:33
Default Re: Octree Grid Refinement
  #4
rohit
Guest
 
Posts: n/a
Tristan,

It's a strange problem, but can i know whether you are defining a surface size for the rest of the surface apart form the ones where you need refinement.

I would suggest you do define a surface size of 3.5mm(i.e, your global mesh size) to all the remaining surface, and lets us see how it behaves with both the refinement sizes.

Rohit.
  Reply With Quote

Reply

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
snappyHexMesh aborting Tobi OpenFOAM Native Meshers: snappyHexMesh and Others 0 November 10, 2010 03:23
Grid Refinement Yamuna FLUENT 0 September 9, 2008 13:24
Grid refinement. Sas Main CFD Forum 8 September 2, 2005 08:36
Combustion Convergence problems Art Stretton Phoenics 5 April 2, 2002 04:59
Grid refinement with wall functions Mark Render Main CFD Forum 8 May 2, 2000 07:09


All times are GMT -4. The time now is 04:43.