Improve Mesh quality - airfoil simulation

Lukas84 · June 21, 2010, 12:19

Hello everyone,

I'm working on a NACA 0012 airfoil analysis in 2D at low Reynolds-Number (Re=1.6e^5, chord length= 0,25m, v=10 m/s) at different angle of attack. I'm using k,omega-SST turbulence model with segregated flow, implicit unsteady with a time step around 1e^-4 s, velocity inlet and pressure outlet, ideal gas (incompressible).
My problem is when i start the simulation it doesn't converge at all. The residuals increase rapidly, so do velocity (http://yfrog.com/muvelj), temperature, pressure etc. until a floating point exception error occurs.
I assume it has something to do with the mesh quality at the (blunt) trailing edge. Star ccm+ retracts the prism layers at the trailing edge(s. http://yfrog.com/jnhinterkante1j ) and the mesh quality close to the trailing edge is very bad to ( http://yfrog.com/j9cellqualityj --> threshold of mesh quality between 0 and 0.1 of the volume mesh before converting to 2D).
Can anyone help me with improving the mesh quality? I tried a lot like mesh refinement etc. but it didn't work. Any suggestions? If the problem has nothing to do with the mesh quality, what else could it be? btw, i checked for negative volume cells: There aren't anyone.
I hope someone can help

JBeilke · June 21, 2010, 19:53

http://www.beilke-cfd.de/naca_0012_5_degree.sim.gz

Lukas84 · June 22, 2010, 05:19

Hallo Joern,

thank you for your quick reply. Can you explain me how to create a structured mesh like the one in your .sim-file in star ccm+? Or do you need to use another program for creating it? Do you have any idea how to improve the mesh quality in my unstructured grid, using e.g. polyhedral model with prism layers?

JBeilke · June 22, 2010, 16:41

The mesh was created with GridPro. I never tried it with ccm+. Before using GridPro I did it with prostar (many years ago), but it was not very comfortable to do :-)

abdul099 · July 6, 2010, 11:07

You can try to play around with the prism layer options. Try to increase the layer reduction percentage (ccm+ will decrease the number of prism layers on the trailing edge where the thickness is reduced) and increase the minimum thickness percentage (if the thickness would fall below this value, prism layer thickness wil be set to zero in this area).

Also try to refne the mesh near the trailing edge.

Best regards

June 21, 2010, 12:19	Improve Mesh quality - airfoil simulation	#1
Lukas84 New Member Join Date: May 2010 Posts: 3 Rep Power: 16	Hello everyone, I'm working on a NACA 0012 airfoil analysis in 2D at low Reynolds-Number (Re=1.6e^5, chord length= 0,25m, v=10 m/s) at different angle of attack. I'm using k,omega-SST turbulence model with segregated flow, implicit unsteady with a time step around 1e^-4 s, velocity inlet and pressure outlet, ideal gas (incompressible). My problem is when i start the simulation it doesn't converge at all. The residuals increase rapidly, so do velocity (http://yfrog.com/muvelj), temperature, pressure etc. until a floating point exception error occurs. I assume it has something to do with the mesh quality at the (blunt) trailing edge. Star ccm+ retracts the prism layers at the trailing edge(s. http://yfrog.com/jnhinterkante1j ) and the mesh quality close to the trailing edge is very bad to ( http://yfrog.com/j9cellqualityj --> threshold of mesh quality between 0 and 0.1 of the volume mesh before converting to 2D). Can anyone help me with improving the mesh quality? I tried a lot like mesh refinement etc. but it didn't work. Any suggestions? If the problem has nothing to do with the mesh quality, what else could it be? btw, i checked for negative volume cells: There aren't anyone. I hope someone can help

June 21, 2010, 19:53		#2
JBeilke Senior Member Joern Beilke Join Date: Mar 2009 Location: Dresden Posts: 528 Rep Power: 20	http://www.beilke-cfd.de/naca_0012_5_degree.sim.gz GAP likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 07:20
Convergence moving mesh	lr103476	OpenFOAM Running, Solving & CFD	30	November 19, 2007 15:09
mesh quality and independency..!??	michelle	CFX	0	October 15, 2007 06:50
Compressible transonic airfoil RAE2822 simulation	Stefano	Siemens	9	June 21, 2006 11:47
Mesh Quality at the Interface	Can Atli	FLUENT	0	November 23, 2005 11:32

June 22, 2010, 05:19		#3
Lukas84 New Member Join Date: May 2010 Posts: 3 Rep Power: 16	Hallo Joern, thank you for your quick reply. Can you explain me how to create a structured mesh like the one in your .sim-file in star ccm+? Or do you need to use another program for creating it? Do you have any idea how to improve the mesh quality in my unstructured grid, using e.g. polyhedral model with prism layers?

June 22, 2010, 16:41		#4
JBeilke Senior Member Joern Beilke Join Date: Mar 2009 Location: Dresden Posts: 528 Rep Power: 20	The mesh was created with GridPro. I never tried it with ccm+. Before using GridPro I did it with prostar (many years ago), but it was not very comfortable to do :-)

July 6, 2010, 11:07		#5
abdul099 Senior Member Join Date: Oct 2009 Location: Germany Posts: 636 Rep Power: 22	You can try to play around with the prism layer options. Try to increase the layer reduction percentage (ccm+ will decrease the number of prism layers on the trailing edge where the thickness is reduced) and increase the minimum thickness percentage (if the thickness would fall below this value, prism layer thickness wil be set to zero in this area). Also try to refne the mesh near the trailing edge. Best regards