|
In previous pages, I have discussed the importance of equal rear tire loading when drag
racing a non-IRS RWD car. It appears that I have omitted, however, an adequate description of how the inequality happens to
exist.
You might think it intuitively obvious that the driveshaft torque would tend to unload the right rear tire. But, it's not
as simple as that. If you place your feet on the bristle end of a push broom, straddling its horizontal handle, you can twist
the handle all you like and you'll never pick up one end of the broom. The forces and torques involved are "internal" and
are not "seen" by the floor below. If, however, you could grab the edge of a nearby table as you twisted, the results would
be different.
Now, suppose the twist is applied, by the driveshaft, to the rear axle assembly. The reaction...to that twist...is taken by
the engine and transmission mounts. But, the presence of the front wheels provides a situation more like that when "the edge
of a nearby table" is grabbed. In other words, if those front wheels were not there, the forces and torques involved would,
again, be "internal" and there would be no tendency for the driveshaft torque to unload the right rear tire.
This does not mean, however, that, when both front tires are lifted from the track surface during a hard launch, the rear
tires are instantly loaded equally. If the tires were unequally loaded when the fronts lifted, there would be a time interval
during which the car would "bounce" from left to right. Only the internal damping of the tires brings this bouncing action
to a stop and anyone who's dribbled a tire and wheel assembly like a big basketball knows there's not a whole lot of damping
available!
In order for the torque applied to the rear axle assembly to be cancelled, it is necessary that the rear suspension springs
deflect. In other words, the chassis must "roll" with regard to the rear axle assembly. (This assumes that the suspension
linkage and spring positions and rates are symmetrical. See other pages at this site for asymmetrical considerations.)
The amount of chassis rotation is related to the "roll stiffness." The front and rear suspensions usually have different values
of roll stiffness, usually expressed in "inch-pounds per radian" or "foot-pounds per degree" or something similar. Obviously,
if the value of the front suspension roll stiffness was zero or, at least, very small compared to that at the rear, the front
suspension would have no ability to absorb any part of the driveshaft torque and, consequently, there would be no unloading
of the right rear tire.
This, then, is the reasoning behind the common "trick" of installing a large anti-sway bar at the rear and removing...or,
at least, disconnecting...the bar at the front.
(Though there is no "suspension" involved, it should be pointed out that the long wheelbase of the dragster acts to minimize
the "loss" of reaction torque to the front wheels and thereby also minimize a loading disparity of the rear tires.)
In addition, speed equipment suppliers now offer an "anti-roll bar." This is somewhat confusing, since the OEM anti-sway bar
is sometimes called an anti-roll bar. But, the anti-roll bar is essentially an anti-sway bar with two apparent differences:
First, the torsional rate is much larger than that of an OEM anti-sway bar, and, secondly, there is provision for preloading.
The advantage of the higher rear roll stiffness, brought about by the anti-roll bar's higher spring rate, has already been
discussed. Unfortunately, the greatly increased rear roll stiffness can result in a dynamic instability problem called "oversteer"
when the car is driven under "normal" road conditions. For this reason, the suppliers strongly recommend that the bar be connected
only at the dragstrip.
With the preloading provision, there is included the opportunity to alter the static wheel loading. (See another page at this
site for details.)
I must emphasize that the use of a large anti-sway bar at the rear or the use of the "anti-roll bar" is an attempt to statically
cancel the driveshaft torque and, as such, is inferior to those techniques which dynamically cancel. Again, see other pages
at this site for examples of dynamic cancellation.
|
|
