This article was first published in 2002.
The idea that race cars undergo extensive aero development is so commonplace that it hardly needs expansion. F1, Indy, Touring Cars - in fact try naming a formula where the cars haven't spent hundreds of hours in a wind tunnel having nearly every aspect of their body shape tweaked. Well, here's one - dirt modified race cars that have up to 750hp, tube frame chasses, basic suspension and run on slick clay...
The fascinating thing is that one of these cars has in fact been inside a wind tunnel - not as part of its development, but simply to see what its drag and lift characteristics actually were. It makes for a very interesting story, with some of the results pretty much as you'd guess - and others way different!
The wind tunnel that was used was the Langley Full Scale Tunnel in the US. The test car was very similar to the ones shown here (there are variations from car to car and over years) with mostly flat body panels. Wings are not permitted in the formula, although front airdams (called 'snow-plows'), an inclined roof panel and a small plastic rear spoiler are all used. A duct-like passage is present down both sides of the car, being bounded by the side panels on their outer edges and the driver compartment on their inner edges. All four wheels are exposed, although the rears project a little under the rear bodywork. Prior to the wind tunnel experiment, it was assumed that the shape of vehicle gave large rear downforce, which was balanced by front airdams and deflectors.
The air movement in the Langley wind tunnel was obtained via two 3 megawatt electric motors driving two 11-metre diameter, four-bladed fans. The vehicle sat on a 4-way balance while being tested. The maximum airspeed used in the wind tunnel test was 100 km/h [much slower than the 130 km/h (corners) and 240 km/h (straights) that the vehicles can achieve in racing] with the data for higher wind speeds then calculated from the measured coefficients. Importantly from an aerodynamic perspective, the tail of the car is slid out during cornering, so the cornering vehicle has a yaw angle to the oncoming airflow.
The car was tested in four configurations - normal ride height (127mm) and a second ride height, 50mm lower than the first. In addition to straight-on airflow, the car was also tested with a yaw angle of -8 degrees (ie the nose of the car orientated eight degrees to the left, which is typical of a sliding cornering car) at both ride heights.
One of the most fascinating numbers was always going to be the drag coefficient (Cd). What resistance would such a 'dirty' vehicle like this pose to the airflow? The answer is a Cd of a whopping 0.69! So the drag of the shape when facing straight into the oncoming wind is more than double that of a normal modern car. When the cross-sectional area is also taken into account, at 240 km/h one of these cars needs 560hp at the tyres just to overcome the aero drag!
But what about when the car is sliding around a corner, presenting not just its front but also part of its side to the airflow? Unlike almost all other cars, the tested dirt-modified car's drag actually dropped marginally at 8 degrees of yaw - from 0.69 to 0.68. (As a comparison, a Land Rover Freelander SUV's drag co-efficient goes up from 0.41 to 0.43 with 8 degrees of yaw.)
Decreasing the ride height - even by only 50mm - improved the Cd, with it dropping from 0.69 to 0.66 with zero degree of yaw, and from 0.68 to 0.65 at 8 degrees of yaw.
So, as expected, the dirt-modified race car is a very high drag shape - not surprising with the mostly open cabin, large wake and exposed front and side crash-bars and roll-cage. But what about front and rear axle lift?
With no crosswind component, at 100 km/h the rear downforce was already 95kg! So far, so good. But at the front the story wasn't nearly as encouraging - here a measured lift of 16kg was occurring. At 240 km/h the rear downforce was an astonishing 594kg - while at the same time front lift was 102kg... With a total car mass of as little as 1100kg (up to 1200kg in some classes), and with only about 40 per cent of this mass on the front wheels, the front end had the potential to get about 25 per cent lighter at 240 km/h!
The designers - who expected the shape to develop rear downforce - were right, but their guesses about the front airdams being able to counteract that rear downforce were very wrong. Part of the reason for this is the forward location of the rear axle - push down on the rear of the body and the front see-saws upwards.
With a yaw angle of 8 degrees (ie equivalent to sliding the car into a corner), at 130 km/h the rear downforce was 158kg and front-end lift was 24kg. Cars of the same shape as the one tested were known for understeering on corner entrances - it was becoming increasingly obvious why!
But lowering the car gave an interesting change - with zero yaw angle and at 240 km/h, front-end lift decreased 17 per cent from 102kg to 85kg, while rear downforce scarcely altered (594 to 588kg).
In addition to drag and lift measurements, flow visualisation was carried out with the use of tufts (stuck on pieces of yarn), smoke streams, and oil placed on the bodywork. These techniques showed that the flow over the roof and side panels was attached, but that the flow along the upper surface of the side pods was separated - probably because of the upstream interference caused by the front suspension uprights.
The airflow directly behind the driver on the rear deck was swirling, however there was cleanish air getting to the rear spoiler either side of the cabin - but the central part of the spoiler directly behind the driver was doing nothing. The flow under the car was smooth (opening up the possibility of using this to create front downforce), while the airflow over the front airdam was disturbed by the network of crash-bars positioned in front of it.
Just decreasing ride hight by 50mm reduced the amount of power this car needed to achieve 240 km/h by 37hp! In addition front-end lift was considerably reduced. However, the imbalance in the front/rear lift co-efficients would always represent an impediment to handling and stability - major body changes would be needed to give overall better results.
So as you can see, no matter what its shape, any race vehicle has important aerodynamic influences acting on it.
SAE technical paper 2000-01-3548