First, a correction. In my earlier posting, I said that positive caster made
the inside wheel in a turn gain positive camber when steered. This is correct,
but I explained it as if it were undesirable. In fact, it desirable to have
the inside wheel go positive and the outside wheel go negative, exactly as
positive caster makes it go. So it is good for both wheels. (Brain fade.)
Akkana tells us that the weight on the front end is a factor in self-centering
of the steering. She is right, but it depends on the steering axis being
inclined inward, i.e. the lateral equivalent of positive caster. Caster is
the angle by which the steering axis is tilted backwards. When the wheels are
steered, any non-zero caster makes one wheel move up and the other down. To
a first approximation (and maybe even exactly) the amount of movement up of
one wheel is exactly the amount the other moves down, so the center of mass
does not get either higher or lower. However, if the steering axis is tilted
inward at the top, closer to the car's centerline at the top than the bottom,
then steering the wheels to *either* side will result in both wheels moving
downward, thus raising the c.o.m. of the car. Naturally, this increases the
potential energy of the car, so it "wants" to fall back down, which therefore
tries to re-center the steering.
I avoided mentioning this angle in my last posting, but since it has come up,
here is a small discussion of its effects. The self-centering action from
contact patch deformation (as I discussed before) is a dynamic one, depending
on the car being at speed. However it provides no self-centering action at
zero speed. That's one reason why steering axes are inclined inward.
The other reason to incline the axis inward is to choose a steering radius,
i.e. the distance between the center of the contact patch and the steering
axis. Under most driving the left and right wheels give equal but opposite
feedback to the wheel. But if the contact patches are way far to the outside
of the axes, the wheels' torques on the steering are balanced but at higher
forces. Should one wheel lose its traction for any reason, the other would
jerk the wheel much more, making the "road feel" too high or even impossible
to control. With no lateral inclination, the steering radius is as large as
1/2 the width of the tire plus the width required for the vertical link and
ball joints to clear the tire, and is generally way too much.
Some cars even use so much inclination that the steering axis intersects the
road surface to the *outside* of the contact patch's center. This is called
negative radius steering. It has the advantage that should you lose traction
on one wheel while braking (or accelerating with FWD), the effect on the
steering will be to yank the steering in the opposite direction from that
which the car itself will be pulled, thus enhancing stability under those
circumstances.
The disadvantage of using too much inclination is that it causes both wheels
to have more positive camber when the wheels are steered. It's good if the
inside wheel goes positive, but since the outside wheel is the important one,
the result is a net loss of cornering power. Since lower or negative radius
steering is a stability-enhancing factor but detrimental for cornering, you
generally find a larger steering radius, and thus less lateral inclination of
the steering axis, on performance-oriented cars. McPherson struts often have
less inclination because the top of the shock tower is the top of the axis,
and being out of the way of the tire, it can be moved further to the outside
than would a real ball joint arrangement. Likewise, an appropriately curved
vertical link can place the upper ball joint above the tire, giving a low
steering axis without resorting to high inclination. Assuming, of course,
that you don't want the inclination for self-centering reasons.
Jim Muller
(yawn...is he done now?)
|