Camber
Theory
Camber is the angle of the wheel relative to vertical.
If the top of a wheel leans in towards the chassis, it is said to have negative camber;
if the top of a wheel leans away from the car, it has positive camber.
The cornering force that a tire can develop is highly dependent on its angle
relative to the road surface, and so wheel camber has a major effect on the road holding of a car.
Since most independent suspensions are designed so that the camber varies as the wheel moves up and
down relative to the chassis, the camber angle that we set when we align the car is not typically what
is seen when the car is in a corner. Nevertheless, it's really the only reference we have to make camber
adjustments. For competition, it's necessary to set the camber under the static condition, test the car,
then alter the static setting in the direction that is indicated by the test results (usually by
studying the temperature change across the width of the tire).
When at rest, a tire (with no static camber) will sit squarely on the ground. The area of contact
between the tire and the road is as big as possible. When cornering, this happy state gets disrupted in
two ways.
Firstly, forces acting through the rubber carcass of the tire cause the inside edge (of the loaded,
outside tire) to lift off the ground. To counter this, some static -ve camber is usually added to the
front wheels, so that the (outside, loaded) tire remains flat on the road while cornering. This is most
noticeable on modern Formula 1 cars, which have huge amounts of -ve static camber on their front
wheels.
Secondly, a well-designed suspension will keep a tire flat on the road while the car rolls around a
corner. However, excessive body roll will also cause the tops of the outside, loaded tires to lean
outwards from the chassis (dynamic +ve camber) and the inside, unloaded tire to lean inwards toward the
chassis (dynamic -ve camber).
Also, when accelerating, the rear suspension will go into compression (squat),
causing the tops of the two rear
tires to tilt in towards the chassis (dynamic -ve camber).
When braking, the front suspension will go into compression (dive),
causing the tops of the two front
tires to tilt in towards the chassis (dynamic -ve camber).
(Suspensions can be designed so that there is no change in camber during squat or dive.
However, this is highly undesireable since it is the generation of camber during
suspension deflection that compensates for camber change caused by body roll.)
Strangely, a tire develops its maximum cornering force at a small negative camber angle,
typically around -0.5°. This is due to the contribution of camber thrust,
which is an additional lateral force generated by elastic deformation as the tread rubber
pulls through the tire/road interface (the contact patch).
Application to GPL
When developing a setup, you should set the camber to zero degrees while you
develop the ride height, wheel rate and roll bar rates.
These things affect the setup more 'deeply' than camber.
Only when you have achieved a
balance with these parameters should you start to add camber to the wheels.
These cars have soft suspension and not much aerodynamic downforce, so it's
generally ok to ignore the unloaded/inside wheels when setting camber.
I don't use asymmetrical camber; the hotlappers might be able to
tolerate the weird handling that it creates, but I can't.
|
That done, the two golden rules of camber are:
- The more camber you add, the worse the car will handle;
- Camber should be added at both ends of the car to keep it in balance.
For example, suppose you have a balanced car with zero camber on each wheel.
You add -1.00° of camber to the front wheels and try a lap of, say, Monaco.
You'll probably notice two things:
- The car will now oversteer in the longer corners when you touch the throttle.
There is now more grip at the front while cornering, and so relatively less at the rear.
- The car will be twitchier, especially over bumps and under braking.
You can restore the balance of the car by adding some -ve camber onto the rear wheels.
This may (or may not, depending on your physics model) give you more overall grip in the
corners, but it won't get rid of the twitchiness.
How much camber you add to each wheel can be guided by observing the
spread of temperatures across the outside tires after going through a long corner.
If the inside edge of the outside tire is cooler than the outside edge of the outside tire
(by more than 1 or 2 degrees)
then the inside of the tire must have been pressing down on the road less than the
outside, and so you can add a click of negative camber to try and flatten the tire on the road
while cornering. Conversely, if the inside edge of the tire is hotter than the outside,
make the camber more positive.
You'll probably find that you end up with more -ve camber on the fronts
than the rears, by about half to a quarter of a degree; for example you might end
up with -1.00 on the fronts and -0.50 on the rears.
You generally set camber in concert with the ride height and roll bars;
the more you allow the car to roll, the more static camber you're likely to apply to each wheel.
(I'm not convinced that GPL models this accurately - a higher, softer car certainly
rolls more, but it doesn't seem to affect the tires in the way it does in the real world.)
A long, flat, fast corner (e.g. Monza/Curva Grande) is a good place to start testing;
with a neutral-handling car, start out with no camber and note the slight understeering
tendancy it produces. Slowly add -ve camber, one click a time, and feel the understeer
disappear. When the car starts oversteering, back off one click. Do this separately
for both ends of the car.
All the while, keep an eye on the tire temperatures at the exit of
the corner.
Finally, find a big braking area (e.g. Zandvoort/Tarzanbocht) and test your braking ability
against a known opponent (e.g. the A.I.); too much -ve front camber can compromise
the braking of the car.
Camber thrust: hold an eraser-tipped pencil vertically, rubber end down,
push it down gently on a flat surface, and move it sideways.
You will feel a certain amount of resistance.
Now tilt the top of the pencil a little away from the direction of travel.
You get more resistance.
|
Don't bother to keep the inside temperature higher than the outside
- to take advantage of the legendary camber thrust - because it doesn't appear to be
modelled in GPL. In any case, Carroll Smith reckons that you need to keep the inside of the tires
about 15°F hotter than the outside to take full advantage of it - which would require
unacceptable amounts of static camber in GPL.
|
Dr Bunsen Honeydew experiments...
Theory says that under acceleration,
the rear of the car will squat and generate its own -ve camber.
So, to get the best contact patch under acceleration, you need less -ve camber at the
rear - you may even need some +ve camber.
On the main straight at Monza, I got my assistant to accelerate
up and down between the two Firestone barriers
that prevent access onto the banking. I found that you get an even contact
patch with 0.00° camber on the rear tires, which suggests that dynamic camber
change due to squat is not modelled in GPL.
Doug Arnao: “[David Kaemmer's] only "flaw" in the model is he doesn't do
dynamic camber and roll center change with roll. That's just too CPU intensive
for right now.”
|
