Unstructured mesh (on Gridgen) doesn't run in FLUENT

Santiago · November 15, 2012, 10:35

Hi Guys,

I'm not an expert using unstructured meshes on FLUENT. The thing is that I had to make a hybrid mesh (prisms + tets) amid my (3D structured) domain in an attempt to reduce the number of nodes. So what a I did was to extrude the unstructured mesh on the walls and to initialize an unstructured block on the core of the block.

When I run a 1st order SIMPLEC with a k-e closure, everything is fine. But when I try to shift gears to a 2nd order scheme the case automatically diverges. It doesn't even bother to make one time step.

Thanks in advance,

S

LuckyTran · November 15, 2012, 12:06

I would check grid quality and try to improve it as much as you can. For meshes with skewed cells can easily lead to those symptoms. Look for the cells between the prism layer and core mesh. This is usually where the skewed cells show up.

eishinsnsayshin · November 16, 2012, 12:46

I've heard SimpleC can have big issues with any skewness since it sets under-relaxation factors to 1 for convergence speed up....did you try using Simple?

LuckyTran · November 16, 2012, 13:02

Quote:

Originally Posted by eishinsnsayshin

I've heard SimpleC can have big issues with any skewness since it sets under-relaxation factors to 1 for convergence speed up....did you try using Simple?

I don't think this is the case, since a 1st order scheme converged without difficulty, but one can always try and find out if there is time.

Santiago · November 16, 2012, 14:48

What would be a good quality criteria for unstructured meshes? The same as for structured (Jacobian/skewness centroid/aspect ratio)?

November 15, 2012, 10:35	Unstructured mesh (on Gridgen) doesn't run in FLUENT	#1
Santiago Senior Member Santiago Lopez Castano Join Date: Nov 2012 Posts: 354 Rep Power: 15	Hi Guys, I'm not an expert using unstructured meshes on FLUENT. The thing is that I had to make a hybrid mesh (prisms + tets) amid my (3D structured) domain in an attempt to reduce the number of nodes. So what a I did was to extrude the unstructured mesh on the walls and to initialize an unstructured block on the core of the block. When I run a 1st order SIMPLEC with a k-e closure, everything is fine. But when I try to shift gears to a 2nd order scheme the case automatically diverges. It doesn't even bother to make one time step. Thanks in advance, S

November 16, 2012, 14:48	Quality	#5
Santiago Senior Member Santiago Lopez Castano Join Date: Nov 2012 Posts: 354 Rep Power: 15	What would be a good quality criteria for unstructured meshes? The same as for structured (Jacobian/skewness centroid/aspect ratio)?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] export hexa mesh to fluent	Wieland	ANSYS Meshing & Geometry	37	January 23, 2013 03:27
[ICEM] Tips for 3D tetra mesh quality for Fluent transient run. Rotary Engine	Catthan	ANSYS Meshing & Geometry	0	June 18, 2012 12:46
Transport mesh from ICEM CFD, to Fluent, to Sysnoise	Wieland	FLUENT	2	April 15, 2012 06:28
Exporting structured mesh from ICEMCFD to Fluent?	jeevan kumar	FLUENT	1	January 23, 2012 11:21
unstructured mesh grid independence for Fluent	Shane Schouten	FLUENT	0	October 11, 2006 16:50

November 15, 2012, 12:06		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,668 Rep Power: 65	I would check grid quality and try to improve it as much as you can. For meshes with skewed cells can easily lead to those symptoms. Look for the cells between the prism layer and core mesh. This is usually where the skewed cells show up.

November 16, 2012, 12:46		#3
eishinsnsayshin Member Join Date: Jun 2011 Posts: 48 Rep Power: 14	I've heard SimpleC can have big issues with any skewness since it sets under-relaxation factors to 1 for convergence speed up....did you try using Simple?