Hello Hrv,

thanks for your reply, I initialize Uz with 0. My U for initializing and inlet is (29.21 0 0). If I run potentialFoam first I get a velocity in z-direction in the range of -10^-17 to 10^-17, but I think it is just the normal noise. 10^-17 might be 0 for OpenFOAM.
The results for Ux and Uy look great from potentialFoam.

So you think, because of this z-velocity, all my results are pointless? That would be bad, because I need this data for my diploma work.

RW

Hello again,

I made a mesh with blockMesh. After running the case, I still got the same velocity in z-direction. As I didnt know how to solve the problem anyway, because all possible options I could think of had a tryout, I started to calculate the pitzdaily-case from the simpleFoam-tutorial with second order.
After computing this case, I got a velocity in z-direction too, so might this be a bug?

My schemes-setup (pitzDaily):

div(phi,U) Gauss GammaV 0.2;
div(phi,k) Gauss Gamma 0.2;
div(phi,epsilon) Gauss Gamma 0.2;
div(phi,R) Gauss Gamma 0.2;
div(R) Gauss linear;
div(phi,nuTilda) Gauss Gamma 0.2;
div((nuEff*dev(grad(U).T()))) Gauss linear;

After 500 iterations I had a velocity about 0.29 m/s, not much, but there shouldnt be a Uz-velocity, because the boundary condition of front- and backplane is "empty". The look of this z-velocity areas in paraFoam look exactly like in my airfoil-cases.
Any ideas?

RW

[sradl]

Had a similar problem (mesh import from Cubit): it turned out that my cell faces in the "empty" patch were not perfectly perpendicular to the z-direction.

So one has to be careful when meshing stuff!

Is there any effort in OF to circumvent this problem and allow more "dirty" meshes to be processes in 2D?

br
Stefan Radl