 There
has been a great deal of conjecture as to why the Mercedes CLR did that
dramatic series of back flips at Le Mans last year. Most of
the speculation has centered on the suspension set up of the CLR, but not
nearly as much thought has been given to the aerodynamics involved.
This article describes the actual aerodynamic principles that made that
flight possible, why this is possible for other cars of this type (LM-GTP),
and why it is less likely for open cockpit LMP type cars.
For a thicker airfoil it is obvious that the path the air has to travel starting at the nose and traveling to the tail has farther to go when it travels over the top of the thicker airfoil than it does over the top of the thinner one. If we suppose that these airfoils are both in a stream of air traveling at the same speed, that air flowing over the top of the thicker airfoil is going to have farther to go in the same amount of time as the air traveling over the top of the thinner airfoil. So by Bernoulli's principal, which states that the faster the air moves the lower the pressure, we can then assume that the thicker airfoil is generating more lift than the thinner airfoil.
Now it makes sense to ask what it is that keeps these cars on the ground at 200 mph. First of all we have the approximately 2000 lb. weight of the car. Second we have ground effect aerodynamics, which are very significant on these cars. "Wait," you say, "they have flat bottoms, they can't possibly be using ground effects to keep them on the ground." Wrong. The nose part of the flat bottom is closer to the ground than the tail portion and the resultant angle is referred to as the rake angle, which is typically 2-5 degrees. That small rake angle creates a shallow ground effects tunnel which produces the vast majority of the total downforce generated by the car. As anyone who has seen these cars can attest, the angle that the flat bottom makes with the road is barely perceptible, yet that same small angle makes a huge difference in the downforce generated by the car. Typically if a team is experimenting with a large change in rake angle set-up for a particular track they might crank in a 0.100 inch change in the height of one end of the car or the other. That should indicate how critical the rake angle is.
It is obvious that once the forces lifting the car exceed those holding it to the ground the car will fly. Since most of these cars have engines in the back, and since the rules in for both the Grand Am and ALMS dictate that the drivers feet must remain behind the front axle plane, it is reasonable to expect that the center of gravity (c.g.) of these cars is well aft. Is it aft of the center of lift? It is hard to say where the center of lift for the top of the car is located without more specific information. The center of lift for the bottom of the car when at a nose up attitude can be estimated to be at less than 50% of the length of the bottom of the car. This puts the center of lift for the bottom of the car in front of the c.g. To make things worse the car has a wing on the back side generating downforce behind the rear tires and well behind the c.g. which naturally tends to lift the nose. Since the lift on top of the car increases with nose up pitch, we do not expect our car to glide like a well balanced balsa airplane. Instead, we would expect that once the nose starts up it will continue to go up as the top side aerodynamic forces increase, which is exactly what we saw in both the flip of the Mercedes at Le Mans and of the Porsche at Petite Le Mans. |
 Could
cars other than the Mercedes become airborne? Here are two photographs
taken last year at the top of the hill under the bridge at Road Atlanta.
Notice that both the Panoz and the Raffinelli Riley and Scott Mk. III were
caught with front wheels off the ground. Fortunately the speeds there were
not sufficient for flight like that of the Mercedes or even like that of
the Porsche GT98 that occurred on the back stretch of Road Atlanta at the
'98 Petite Le Mans. Certainly in the future, contenders will consider the
fate of these cars when determining spring rates and shock settings for
the rear of their cars. On the other hand consider this, a softer spring
in the rear and softer shock settings help keep the rear wheels on the
road more, so even at relatively high downforce tracks, it is not unusual
for a car to run softer springs on the back side than the front.
Beyond the setup issues, there are other kinds of incidents that can effect the attitude of the cars. In '99, Jim Downing's Kudzu DLY was put out of the Mosport race by a half shaft that fell off of another car. Fortunately Jim did not hit that half shaft with a front tire, or it could have conceivably lifted his nose off of the ground a significant amount. |
 Would
this have caused the car to flip? It is hard to say. The DLY is an open
top car and thus less likely to blow over than an closed top car as I have
discussed above. Could that same half shaft have caused a closed top car
to flip? In light of what happened to the Mercedes and Porsche cars, the
answer seems to be yes.
When we look at the height and distance that the Mercedes CLR flew at Le Mans, we all as race fans feel very fortunate that there were no spectators killed, as well as being thankful that the driver was ok. Certainly such a tragedy should be avoided and we should all embrace rules that make such events less likely. |
Let's begin with a couple of illustrations. The first is a side view
of the CLR. At first we see a vehicle which has very complicated
aerodynamics with air going in here and out there, kinks and wings and
various aerodynamic devices. Of course, in keeping with the most
fundamental of engineering principles, we attempt to simplify the aerodynamics
to make more clear what the potential problems are. Thus you see
that I have sketched an airfoil which envelopes the outline of the CLR.
In the second illustration of the Audi R8 open-top roadster, I have also
sketched an enveloping airfoil. Immediately we notice that the airfoil
enveloping the CLR is much thicker than that enveloping the Audi.
So what does that mean?
So with the statement that these cars, as illustrated, are roughly shaped
like lifting wings, we can now look for evidence that supports our hypothesis.
Perhaps the most graphic evidence could be found in the problems experienced
by the '99 Audi R8C coupe in testing. Many will remember the pictures
posted on the Internet of the Audi with its door sucked off. This
was reported to be a prevalent problem during testing of that car.
There was also a problem with the rear bodywork being sucked off the car.
Apparently the lift forces experienced by the bodywork of that car were
more than the designer had anticipated. Anyone watching the '98 Petite
Le Mans from the front straight might also have noticed the rear window
of one of the RX-7's that was sucked far into the air, not once, but twice.
Both of these events provide solid evidence to support our theory that
the top sides of these cars develop lift.
Consider this, if the nose comes up to somewhere around one degree of nose
down rake angle, the aerodynamic force generated by the bottom of the car
becomes lift. This lift pressure builds rapidly as the nose comes up. Stagnation
pressure (the maximum possible pressure) can be achieved within a few degrees
of nose up rake angle. This stagnation pressure could result in a lift
force of over 4000 lbs at 200 mph. That's over twice what is required to
lift the car off the ground. The preceding illustrations were generated
by a CFD analysis of a wing (left) in free air and that same wing in ground
effect (right). The red area represents high pressure. Notice how much
more red there is under the wing in ground effect than under the wing in
free air. This illustrates that not only is it possible to develop a great
deal of downforce with ground effects, but it is also possible to generate
a great deal of lift. In addition to the lift generated on the bottom side
of the car as the nose raises up, there is also a slower build up of lift
on the top of the car. 
