CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > ANSYS Meshing & Geometry

[GAMBIT] 3D meshing of a tortuous blade

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

Reply
 
LinkBack Thread Tools Display Modes
Old   January 14, 2014, 06:54
Default 3D meshing of a tortuous blade
  #1
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
Hello,

I intend to mesh a turbine blade with hex cooper. But problems with skewed an even inverted elements arise on the leading edge

I meshed the tip seperately, linked it to the hub face... due to the different geometry there it gets skewed already... when meshing the volume it takes quite a while to generate the mesh and then produces skewed and inverted cells

Is there a way to for example spilt the faces in a way so that hex cooper is possible? I have not dealt with inverted elements and am not sure how to approach here
Attached Images
File Type: jpg inverted Elements.jpg (92.9 KB, 15 views)
File Type: jpg 2dmesh leading edge.jpg (92.5 KB, 13 views)
Persil is offline   Reply With Quote

Old   January 14, 2014, 07:11
Default
  #2
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
Where do inverted (skewed) elements come from?
from your BL or from your core-mesh?
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 14, 2014, 07:26
Default
  #3
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
I did a BL around the airfoil at hub and tip, there the elements look fine. So it must be the core mesh. Which is the face meshed with quad pave on the attached picture....

...but I am not entirely sure about the inverted elements. It kind of looks like those are in the BL and the skewed elements are caused by the meshed face at the leading edge
Persil is offline   Reply With Quote

Old   January 14, 2014, 07:30
Default
  #4
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
before you try some splits, you may try to increase the BL height
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 14, 2014, 07:56
Default
  #5
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
ok thank you, I tried that which still causes the linked face to end up skewed. See the picture: the face in front is ok, but the other one is skewed and will produce even more skewed elements when meshing the volume...

is it possible to do hex cooper without face linking the source faces? or do you have an idea to improve the quality of the linked face
Attached Images
File Type: jpg BL_increased.jpg (95.5 KB, 9 views)
Persil is offline   Reply With Quote

Old   January 14, 2014, 08:03
Default
  #6
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
you can also mesh your face with tri. The volume will still be meshed with cooper (wedge)
When I see your picture, it seems you mesh the first surface (in front of you ) first. And Gambit try to project your source-surface on the other (back surface),
Try to mesh the back surface first.
If it won't fix your problem, then you will have to split both source surfaces and you will try to mesh one source with quad (no pave)
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 14, 2014, 09:08
Default
  #7
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
what do you mean with your last sentence? I cannot imagine meshing one face in quad (why no pave?) and the other differently.... those faces need to be linked to my understanding
Persil is offline   Reply With Quote

Old   January 14, 2014, 10:28
Default
  #8
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
pave is unstructured. Map is structured, and you have more control on your mesh.
You have to understand of course that both surfaces are map (or pave). But you only have to mesh one before you go and mesh the volume. In your case both surfaces don't have to be linked.
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 14, 2014, 13:06
Default
  #9
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
when I only mesh 1 of the faces and then mesh the volume, the remaining face will be automatically meshed by gambit, so I achieve nothing... gambit crashed :/

why do you think, that there is no face linking needed in my case?
Persil is offline   Reply With Quote

Old   January 14, 2014, 13:22
Default
  #10
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
Link faces only if you need periodicity.
In your case, Gambit meshes the other face because of the link.
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 15, 2014, 05:00
Default
  #11
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
ah well, then I have not given you enough information: I definitly need the periodicity because this mesh will be extended circumferentially
Persil is offline   Reply With Quote

Old   January 15, 2014, 05:35
Default
  #12
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
But according to your pictures I cannot imagine your source faces are periodic
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   January 15, 2014, 05:47
Default
  #13
Member
 
Join Date: Nov 2013
Posts: 43
Rep Power: 3
Persil is on a distinguished road
yes, those faces itself are not periodic.... but the faces connecting those 2 are.

maybe the picture will help explain this, you see 1 blade, where the tip and hub are meshed via linking so that the outer faces (red and opposite) are periodic
Attached Images
File Type: jpg periodic.jpg (96.5 KB, 8 views)
Persil is offline   Reply With Quote

Old   January 15, 2014, 05:58
Default
  #14
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 2,965
Rep Power: 30
-mAx- will become famous soon enough
yes, now it is clear.
What is the quality if you mesh this volume with tetra? (just to be sure that your BL is clean)
If you want you can also provide me the dbs file or at least only the volume which causes skewed elements
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   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
[ICEM] High Stagger Rotor Blade Meshing eromon84 ANSYS Meshing & Geometry 12 August 14, 2013 06:40
[ICEM] Incomplete blade Hexa meshing m5edr ANSYS Meshing & Geometry 3 June 22, 2013 16:36
[ICEM] Hexa meshing of Wind Turbine blade. Help for Newbie please federvo.mala ANSYS Meshing & Geometry 4 June 8, 2013 03:19
Meshing a 3D, tapered and twisted blade with Gambit-TGrid teskio ANSYS Meshing & Geometry 0 July 3, 2009 10:36
Question on blade meshing Jeffrey CFX 0 March 7, 2008 18:13


All times are GMT -4. The time now is 21:01.