Lift on symmetric body at 0deg AoA

Jason

Ok... I have a mortar round (like they have on the old war movies... drop it into the tube, it makes a "thump" noise, and now you have a good old-fashioned projectile, but it goes "boom" at the end). We're analyzing the system for some possible retro fits...

The physical geometry is 60deg periodic (6 tail fins), but because of an angle on the fins a periodic CFD model can't be used at angles of attack. Symmetry is gone too because of this. So I had to model the entire thing. I used an unstructured mesh (complicated geometry... someone out there can probably do a structured mesh on it, but I am not that someone) and ran it using the segregated solver (low speed Mach ~0.2, first order converge then second order converge). On the converged solution I have a lift on the fuze assembly (which is far enough forward of the fins that there is no effects caused by their angle)... but this part of the model is fully axisymmetric, and therefore at 0deg AoA, there should be no lift! It's a small force (<10% of the axial load which is small because it's a pretty aerodynamic shape to begin with).

This leads me to believe that even though this part of the geometry is symmetric, since I'm using an unstructured mesh, the mesh has an asymmetry, which is causing an asymmetry in the flow, which is causing an asymmetry in the loads, and therefore I get lift when I shouldn't. Is there a way to prove this? If I simply refine the mesh, I might still get this asymmetry (or a different one) and get loads in directions I shouldn't. Is there a way to find out where in the mesh the problem is so that I at least have an idea of where to refine? Is there something else that could be causing this that I'm missing?

I really appreciate any help.

Thanks, Jason