I am trying to run a simulation for a 3d symmetry case. The questions are:

1. Can I simulate only half the model (symmetry) in Fluent?

2. For the coefficient, such as Cd and Cl, do I need to time by 2 the simulated Cd and Cl for the real value?

3. Do I need to use the Size Function to control my mesh size?

Thank you.

1) Yes, just use the symmetry b.c. on your symmetry plane. Remember, if you've got separated regions along the symmetry plane, you'll get more of a time averaged solution due to the symmetry plane cutting the separated region in half.

2) Well, a common way of dealing with this is to only use half the reference area (since you only have half of your model) that way the coefficients are correct for the full model. If you're looking at forces directly (in Newtons or Lbs, or whatever units you're in) then you will be only seeing half of the force (once again, because you only have half the model). Fluent doesn't assume anything special when you apply a Symmetry B.C. It's only a way of dealing with one of the surfaces on your volume mesh.

3) Depends on your meshing. If you're using structured mesh (or if you're using the cooper meshing scheme for volumes), it's easy to control the volume mesh by meshing edges, then faces, then volumes (and uses less memory). If you're using a tet mesh, then the easiest way to mesh a volume is using sizing functions. The other option is to decompose the C.V. into smaller volumes. These volumes would have to be small enough so that the mesh size is about constant through the volume. In a 3D tet mesh, Gambit tries to apply whatever element size you have specified. If there isn't a lot of variation in mesh size on on that volume then this isn't a big problem, but if you have a refined area, then the growth rate off of this refined area tends to be too large.

Hope this helps, and good luck, Jason

Well Jason, thank so much for your help. It really help me a lot.