[harly]

Hi,

I have a 3D case running of a sphere(d=1) with Re=50 and I want to determine the drag.

I used the following in my controlDict

functions
(
forces
{
type forces;
functionObjectLibs ("libforces.so"); //Lib to load
patches (sphere_GmshSphere); // change to your patch name
rhoInf 1.22; //Reference density for fluid
CofR (0 0 0); //Origin for moment calculations
}
forceCoeffs
{
type forceCoeffs;
functionObjectLibs ("libforces.so");
patches (sphere_GmshSphere); //change to your patch name
rhoInf 1.22;
CofR (0 0 0);
liftDir (0 1 0);
dragDir (1 0 0);
pitchAxis (0 0 0);
magUInf 0.000753;
lRef 1;
Aref 0.785398;
}
);

and

nu nu [0 2 -1 0 0 0 0] 1.5e-05;

in transportProperties

the problem is now , that my drag is too high(about 7 what should be 1.5).

Can someone see the a mistake or shall I blame the Grid ?

If someone has a case which gives good results for a laminar flow I would be happy about every input.

- Daniel

[wolle1982]

hi,

when did simulation on an airfoil, i also had too high drag. the distance to toe far boundary edge was 40 times the airfoils chord length. increasing to 100 times the chord length improved it.

--> try make the grid around the sphere very huge!

and yes, the grid is a very sensitive factor to the results. try not to get too great skewness near the boundary layer. and also refine the mesh behind the sphere.