[turbine123]

Hi everyone,

I am trying to simulate a NACA 23015 airfoil in OpenFoam 8. The source of the results is a technical report written by NASA "AERODYNAMIC CHARACTERISTICS OF 15 NACA AIRFOIL SECTIONS AT SEVEN REYNOLDS NUMBERS FROM 0.7 x 10^6 TO 9.0 x 10^6"
I started with a opensouce motorbike case and tried to change it for the airfoil simulation. (I am unfortunately an OpenFoam newbie)
The model for the simulation was created in Catia. I am trying to study an "endless airfoil" which means that the model is larger in y-direction than blockmesh.
The airfoil length is 24 inch (0.6096m).

My blockmesh:
vertices
(
(-3 -0.05 -5)
( 8 -0.05 -5)
( 8 0.05 -5)
(-3 0.05 -5)
(-3 -0.05 5)
( 8 -0.05 5)
( 8 0.05 5)
(-3 0.05 5)
);

blocks
(
hex (0 1 2 3 4 5 6 7) (110 1 100) simpleGrading (1 1 1)
);
The origin of the coordinate system is also the origin of the airfoil

After a few tests I noticed that I have to change the turbulence model in a low reynolds model, because my y+ is lower than 5.
I have chosen the kklOmega model. Now my case is running, but I get wrong results. My result for cd should be at 0.007 for a Re=2*10^6 but I get 0.0038. The cl should be 0.1-0.12 and I get 0.141

I am going to add the systems folder so that you can take a look but in the following lines i will summary the most important things:

snappyhexmesh:
refinementBox2
{
type searchableBox;
min (0.35 -0.05 -1);
max ( 4.0 0.05 1);
}

nCellsBetweenLevels 6;

edge refinement: level 9
surface-wise refinement: 8 9
region-wise refinement: levels ((1E15 5))

nSurfaceLayers: 10
min. Thickness: 0.1
finalLayerThickness: 0.3

fv solution

for u,kl,kt and omega:
{
solver smoothSolver;
smoother GaussSeidel;
tolerance 1e-8;
relTol 0.1;
nSweeps 1;
}

residualcontrol
{
p 1e-4; // 1e-2
U 1e-5; // 1e-3
"(kt|kl|epsilon|omega)" 1e-5; // 1e-3
}

fvschemes
divSchemes
{
default none;
div(phi,U) bounded Gauss linearUpwindV grad(U);
div(phi,kt) bounded Gauss upwind;
div(phi,kl) bounded Gauss upwind;
div(phi,omega) bounded Gauss upwind;
div((nuEff*dev2(T(grad(U))))) Gauss linear;
}

Please don´t be confused that the airfoil is named motorBike(stl)


I hope you can help me. If you need more information just let me know.

Thanks for your help!

[klausb]

It's probably insufficient boundary layer mesh refinement, try y+ < 1, use up to 10000 timesteps and skip residual control.

[HPE]

In addition to Klaus' remarks: I don't want to blame the turbulence model at this point, but the model is transitional model. In your case, do you expect any transition?