"I'll be more enthusiastic about encouraging thinking outside the box when there's evidence of any thinking going on inside it." - Terry Pratchett

***

"Automotive aerodynamics is more complex than one would think. To begin with, it is more difficult to measure aerodynamic parameters of car models than those of airplanes. Airplanes and submarines are fully submerged in air and water, respectively, while automobiles are driven on solid surfaces. Wind-tunnel experiments are performed in various ways to accommodate that difference. A complicated scheme to achieve complete similarity between experiment and reality involves running a car at a fixed location on a movable belt, a sort of treadmill for vehicles. Wind speed and belt speed must be identical, duplicating the actual driving speed. If a full-scale automobile is tested, the Reynolds number is perfectly reproduced by this scheme. Most experiments use models mounted on a fixed plate, but this method falsifies the actual drag values since the model is exposed to the wall boundary layer of the wind tunnel. Empirical corrections of aerodynamic parameters have been established for this and other testing methods. Car models can also be suspended in the free stream, just like airplane models. This of course makes full-scale simulation impractical. There have been other attempts to solve the complicated problem of the relative motions of car wind, and road, such as having two models glued together symmetrically at their wheels, but we will stop here."

--- Source unknown