NACA 0012 k epsilon LRC - convergence problem

Bernoulli666 · May 15, 2017, 07:14

I am trying to run the NACA0012 validation case with k-epsilon model. The data that I am using are in the folders attached. I am not reaching convergence as you can see in the picture. I am using the mesh given by the NASA website, with size smaller than 449-129 I reach convergence but results are not accurate, with mesh size higher than 449-129 I have the residuals as in the picture. Which parameters do I have to change? I tried increasing and decreasing relaxation factors but it does not reach convergence anyway.

Flowkersma · May 15, 2017, 09:01

k-epsilon turbulence model uses wall functions and therefore the first cell wall distance should be y+ > 30. The meshes from NASA have y+ < 1 and therefore should not be used with standard wall functions.

Bernoulli666 · May 15, 2017, 09:33

So which kind of boundary conditions do I have to use for k epsilon and nut in the airfoil?
I read in the forum that, if y+<1, zeroGradient is ok. Could you please explain me why?
Thank you

Flowkersma · May 15, 2017, 10:01

Basically, what I am saying is that you cannot use that mesh with k-epsilon model. However, this is not exactly true because there are also enhanced wall functions which can be used with y+<30. I think I have read somewhere that enhanced wall functions were also recently implemented to OpenFOAM but I would still suggest to use some other turbulence model such as k-omega SST or Spalart Almaras.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] issue occur after extrude 2D airfoil mesh and convergence problem in CFX	shiyun	ANSYS Meshing & Geometry	4	May 9, 2012 20:55
Submerged fin, Convergence problem	supermouniette	FLUENT	10	July 6, 2009 11:47
Epsilon Convergence Trouble	Carlos	FLUENT	4	August 27, 2007 12:22
CONVERGENCE PROBLEM - oil boiler	MM	FLUENT	1	February 15, 2007 06:24
convergence problem with SIMPLER	NURAY KAYAKOL	Main CFD Forum	1	February 24, 1999 14:43

May 15, 2017, 09:01		#2
Flowkersma Senior Member Mikko Join Date: Jul 2014 Location: The Hague, The Netherlands Posts: 243 Rep Power: 13	k-epsilon turbulence model uses wall functions and therefore the first cell wall distance should be y+ > 30. The meshes from NASA have y+ < 1 and therefore should not be used with standard wall functions.

May 15, 2017, 09:33		#3
Bernoulli666 New Member Ajeje Brazov Join Date: Apr 2017 Posts: 14 Rep Power: 9	So which kind of boundary conditions do I have to use for k epsilon and nut in the airfoil? I read in the forum that, if y+<1, zeroGradient is ok. Could you please explain me why? Thank you

May 15, 2017, 10:01		#4
Flowkersma Senior Member Mikko Join Date: Jul 2014 Location: The Hague, The Netherlands Posts: 243 Rep Power: 13	Basically, what I am saying is that you cannot use that mesh with k-epsilon model. However, this is not exactly true because there are also enhanced wall functions which can be used with y+<30. I think I have read somewhere that enhanced wall functions were also recently implemented to OpenFOAM but I would still suggest to use some other turbulence model such as k-omega SST or Spalart Almaras.