[edreed]

I'm just getting start with OpenFOAM and am looking to use it for external aero applications. To get started, I followed the pitzDaily tutorial in the Programmer's Guide for the rearward facing step with simpleFoam. When I try to extend this problem to 3D with a tetrahedral mesh around a simple aircraft, the solution is not coherent. I've turned off turbulence in an effort to simplify things with no luck.

Do I need to change interpolation schemes to get reasonable results using a tetrahedral mesh? What am I missing?

[pbo]

What do you mean exactly by "not coherent"?
Do you have a prismatic layer (normal extrusion of your surface mesh at the wall) between your aircraft and your tet mesh? If not, you will not be able to properly resolve the boundary layer. As regards interpolation schemes, go for higher order schemes like Gamma or linearUpwind.
It may be also worth checking your mesh with checkMesh (if the mesh is compromised, so is your numerical solution).

[edreed]

I was not setting an initial velocity for the field, just a boundary condition. Once I set an initial field, either constant or initialized from potentialFoam, I was able to get reasonable results.