I created an airfoil in Blender, exported it as an FBX file and when I imported it Unity recognizes it as a mesh. I can't seem to use this mesh in the geometry game object and have the aerodynamics work. It seems that the aerodynamics only work with Unity primitive meshes.
I have a work around where I use a cube game mesh for the geometry to create a wing as described in one of the first tutorials. Then I added a game object into the geometry with my blender mesh and scaled it so that it's larger than the cube mesh. There are two drawbacks to this approach. First is the fact that it's not using the airfoil surface to compute the lift and drag forces (so I'm lying!). Second is that I can't add a control surface.
Is it possible to use different meshes than the few that I've seen including sphere, cube, cylinder, capsule?
Hi there Michele
Thanks for your question. One of the core principles we use within AO is that we approximate the aerodynamics of bodies by assuming they are ellipsoids (or combinations of), and we dont use a mesh as such to define geometry. In the tutorials and examples we use primitive meshes for graphics to emphasise the point that the modelling doesnt take into account detailed surface geometry. That said the aerodynamic models are pretty good at capturing the right behaviour even with simple approximations. For an aerofoil, the main thing AO picks up from its shape is its maximum thickness to chord ratio which determines drag and maximum lift. Another important parameter is the camber, which you have to specifically enter yourself. For the overall 3d aerodyamics the other important part is the aspect ratio which gives the lift curve slope, which again comes from the ellipsoid body.
So to answer your question, it is the shape of the aerodynamic object ellipsoid that controls the aerodynamic model, not the attached mesh. You can obviously add what mesh you like on to the object, but it will only use its ellipsoid shape for the aerodynamics.
If you have a control surface on your mesh you can link the transform rotation of this to the control surface component on the aero object if have one. The control surface aerodynamics are pretty good in the AO model.
Please try not to be too disappointed that you cant use 'real' aerofoil section geometry! There is some myth that their detailed shape is the secret to flight. This is not the case - if you have the thickness to chord ratio and camber of a wing section plus the wing aspect ratio you will be at least 80% right all the time. If you want to be more accurate than that you will need a higher order computational method for the aerodynamics which is beyond the scope of AO.
Hope this helps somewhat
Bill