|
I have only skimmed over the discussion, but I do have a few comments that hopefully are not redundant. Back in the late '80's, CFD codes had a huge accuracy problem in regions of flow separations due to difficulties with the turbulence modeling. I have been told that the situation is better now. It is actually much easier to model an aircraft with CFD than a car because the flow generally stays attached for almost the entire length of a well streamlined airplane (unless it is at a high angle of attack). The flowfield around a car is generally characterized by large flow separations particularly on the trailing end. As for optimizing the rear deck angle, you should find a copy of Aerodynamics of Road Vehicles by Wolf-Heinrich Hucho. There is usually a drag spike at an angle of about 30 degrees down from the horizontal. Also bear in mind that a car is often not aligned with the flow direction due to cross winds (Beta angle). The drag coefficients that Ford reports are "wind-averaged" values from testing performed over a range of yaw angles. It also appears that you have not included a moving ground plane in your simulation - this has a significant effect on the aerodynamic forces. One way to simulate this effect is to attach a mirror image of the vehicle about the horizontal plane (where the tires contact the pavement). Moving ground plane effects are another reason it can be more difficult to simulate a car flowfield as opposed to an airplane flowfield.
|