[Other] Hex meshing problem DM12

patrick · July 14, 2009, 17:45

Hi,

I'm fresh in meshing and struggling to successfully mesh simple geometry - cylinder with tangential pipe inlet and circular outlet in the bottom of the cylinder.
I'm using Ansys WB12 with DesignModeler and Meshing application for CFX and Fluent. I need to obtain structured mesh which is refined in the middle and at the outlet of the cylinder where the air core is. I've posted some examples of good mesh for cyclones and the preview of my geometry.
I was able to get custom and refined mesh (unstructured) but the quality is still poor. Then I have divided the geometry into three bodies-parts which are meshed separately: cylinder, pipe joint and straight inlet pipe. I get the problematic geometry error at the end tip of the tangential inlet pipe. I was only able to sweep the cylinder alone but the mesh quality was poor (quad/prism combination). Please find my geometry attached in DM12 format *.agdb (cylinder.zip).
I have studied tutorials and static mixer is very similar to my problem, the only problem is tangential inlet (and the mesh is poor).

I'll appreciate any help/tip or piece of advice
-Patrick

PSYMN · July 15, 2009, 10:38

Did you try to insert a MultiZone Method for this model?

Did you combine the three parts together as a Multibody part (in DM)?

Simon

patrick · July 15, 2009, 12:32

Yes, I have combined all three parts into one. The multizone mesh failed on tangential inlet

. I have no problems meshing cylinder with multizone, after creating virtual topology with bottom outlet and bottom part. The problem is with the highly unstructured and skewed mesh.

patrick · July 25, 2009, 11:22

I'm so desperate that I have divided my body into 10-part multibody (attached). Still no effect. The mesh is just random and unstructured. I'll appreciate any help.

PSYMN · July 27, 2009, 16:20

I did try it with WB, but it seems to have trouble with the sharp end of the tangential connection... I have passed the model to development for further testing.

ICEM CFD is patch independent, so in 5 minutes I was able to get a reasonable Hexa mesh with that tool. For good quality, I had to ignore the sharp end of the tangential contact point... Since it is tangential, I think it is a good compromise.

Here are some screen shots. This is just one of many ways I could have structured the topology (look for another post on CFD Online where I uploaded images of a Quarter OGrid structure )… I just did it rough with all default sizing, etc. The blocking and edge distributions could be further adjusted to produce a very nice mesh.

I am sure that ANSYS Meshing will eventually be able to take care of problems like this, but in the mean time, I am happy to help you proceed with ICEM CFD Hexa blocking.

patrick · July 27, 2009, 16:48

Hi Simon,
Thank you very much for your help and time! The mesh from ICEM is simply beautiful - exactly as I've imagined it. Hex mesh is stunning!
It's a good idea to ignore sharp point at inlet. I was also thinking of dividing the resultant contact face edge (eye-shaped) into 5 pieces. I was studying manuals and found this solution for cusps - the problem is connected with parametric surfaces. I don't know if this will help but I'll try it tonight and post results tomorrow.
-Patrick

vansyle · February 28, 2013, 17:32

Hi Simon,
Could you please tell me some step to mesh this model. I am struggling with meshing tangential connection between two pipe. Could you upload the file of this model.
Thank you very much

mustafa-uslu · January 9, 2015, 08:21

I am having similar problems with ansys meshing. Isnt there any way to carry out this on ansys meshing?

July 15, 2009, 10:38	MultiZone	#2
PSYMN Senior Member Simon Pereira Join Date: Mar 2009 Location: Ann Arbor, MI Posts: 2,663 Blog Entries: 1 Rep Power: 47	Did you try to insert a MultiZone Method for this model? Did you combine the three parts together as a Multibody part (in DM)? Simon

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Boundary layer meshing problem in Gambit	Crystal	FLUENT	0	June 12, 2009 21:58
Boundary layer meshing problem in Gambit	Crystal	Main CFD Forum	0	June 12, 2009 15:32
problem with gambit unsymmetrical meshing	meenakshi	FLUENT	1	March 28, 2008 13:46
3D-gambit meshing problem	prem	FLUENT	3	February 28, 2006 03:42
Meshing problem...	Gustaf	CFX	2	March 28, 2003 11:37

July 15, 2009, 12:32		#3
patrick New Member Patryk Wójtowicz Join Date: Jun 2009 Location: Wroclaw Posts: 8 Rep Power: 17	Yes, I have combined all three parts into one. The multizone mesh failed on tangential inlet . I have no problems meshing cylinder with multizone, after creating virtual topology with bottom outlet and bottom part. The problem is with the highly unstructured and skewed mesh.

July 27, 2009, 16:48		#6
patrick New Member Patryk Wójtowicz Join Date: Jun 2009 Location: Wroclaw Posts: 8 Rep Power: 17	Hi Simon, Thank you very much for your help and time! The mesh from ICEM is simply beautiful - exactly as I've imagined it. Hex mesh is stunning! It's a good idea to ignore sharp point at inlet. I was also thinking of dividing the resultant contact face edge (eye-shaped) into 5 pieces. I was studying manuals and found this solution for cusps - the problem is connected with parametric surfaces. I don't know if this will help but I'll try it tonight and post results tomorrow. -Patrick

February 28, 2013, 17:32		#7
vansyle New Member df Join Date: Oct 2012 Posts: 4 Rep Power: 14	Hi Simon, Could you please tell me some step to mesh this model. I am struggling with meshing tangential connection between two pipe. Could you upload the file of this model. Thank you very much

January 9, 2015, 08:21		#8
mustafa-uslu New Member Musty Join Date: Oct 2013 Posts: 8 Rep Power: 13	I am having similar problems with ansys meshing. Isnt there any way to carry out this on ansys meshing?