[ICEM] Meshing Ground Effect and a Leading Edge Rotating Cylinder Correctly

HarryCP · March 9, 2017, 14:50

Hi my thesis involves examining a NACA0015 with a Leading Edge Rotating Cylinder (LERC) in Ground Effect.

So far I have successfully meshed and achieved good results for: NACA0015 in a freestream, NACA0015 in Ground effect, NACA0015 LERC in freestream.

Now when I mesh the NACA0015 LERC in ground effect and import into fluent to solve I am unable to get the simulation to start as divergence is detected and some strange flow behaviour appears.

Attached are some photos of the meshes I have toyed around with.

Any suggestions or help would be greatly appreciated!

bluebase · March 10, 2017, 05:47

Hello Harry,

you have some highly skewed cells in the throat area behind the cylinder.
I marked them in red in the following picture.
Fortunately, this can be fixed with a few extra splits, see the picture.
bettergrid.jpg

Always check your mesh quality. For example check Determinate and (Eriksson) Skewness. I believe there are mathematical proofs about numerical stability, but here are my rule of thumbs: I wouldn't start a simulation until the skewness is above 0.3. I recall some options of fluent to allow high-skewed-mesh simulations, but the solution quality is reduced.
The determinante in my experience also shouldn't be below 0.5. If possible above 0.7.

Another problem i see in your fifth picture is the extreme change in cell size. There are proofs that jumps in cell sizes shouldn't be above 1.2, relatively.
As you can see in your picture, in the throat area you have a tiny boundary-layer-cell right next to a (relatively) huge cell. This is a very unstable configuration.

With regards,
Sebastian

HarryCP · March 11, 2017, 07:56

Hi Sebastian,

Thank you very much for your swift response and I appreciate your time,

I have begun re-meshing, introducing the new splits as you recommended. Unfortunately I am still having no luck, but I reckon it is because of mesh quality and the drastic changes in cell size which you pointed out. I am currently working on smoothing the mesh in order to improve skewness/determinant and hopefully I'll crack it soon.

I find the aspect of geometric symmetry an issue, I understand ICEM/FLUENT prefers a nice symmetric mesh and in my model I need to decrease the distance to ground to analyse the impact it has upon the aerofoil. Therefore the mesh above the aerofoil will be different to the mesh below, any tips on how tackle this problem?

Kind regards,
Harry

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] meshing airfoil in ground effect	cfd_peter	ANSYS Meshing & Geometry	12	March 14, 2017 18:30
Flow Simulation : air around an rotating cylinder using Solidworks Flow Simulation	Wyrold	Main CFD Forum	0	October 22, 2015 09:48
[ICEM] Meshing a moving car on ground	lihuang	ANSYS Meshing & Geometry	0	March 15, 2011 11:50
ground effect fluent modelling	meenakshi	FLUENT	0	June 11, 2008 02:09
Rotating Cylinder IGE	Jason Mc Beth	FLUENT	0	January 23, 2008 08:02

March 10, 2017, 05:47		#2
bluebase Senior Member Sebastian Engel Join Date: Jun 2011 Location: Germany Posts: 567 Rep Power: 21	Hello Harry, you have some highly skewed cells in the throat area behind the cylinder. I marked them in red in the following picture. Fortunately, this can be fixed with a few extra splits, see the picture. bettergrid.jpg Always check your mesh quality. For example check Determinate and (Eriksson) Skewness. I believe there are mathematical proofs about numerical stability, but here are my rule of thumbs: I wouldn't start a simulation until the skewness is above 0.3. I recall some options of fluent to allow high-skewed-mesh simulations, but the solution quality is reduced. The determinante in my experience also shouldn't be below 0.5. If possible above 0.7. Another problem i see in your fifth picture is the extreme change in cell size. There are proofs that jumps in cell sizes shouldn't be above 1.2, relatively. As you can see in your picture, in the throat area you have a tiny boundary-layer-cell right next to a (relatively) huge cell. This is a very unstable configuration. With regards, Sebastian

March 11, 2017, 07:56		#3
HarryCP New Member Harry Clare-Paule Join Date: Nov 2016 Posts: 2 Rep Power: 0	Hi Sebastian, Thank you very much for your swift response and I appreciate your time, I have begun re-meshing, introducing the new splits as you recommended. Unfortunately I am still having no luck, but I reckon it is because of mesh quality and the drastic changes in cell size which you pointed out. I am currently working on smoothing the mesh in order to improve skewness/determinant and hopefully I'll crack it soon. I find the aspect of geometric symmetry an issue, I understand ICEM/FLUENT prefers a nice symmetric mesh and in my model I need to decrease the distance to ground to analyse the impact it has upon the aerofoil. Therefore the mesh above the aerofoil will be different to the mesh below, any tips on how tackle this problem? Kind regards, Harry