RE: suspension

[Scott Fisher <sefisher@cisco.com>]

Dave Eaton suggests:

> wishbones are damn fine for relatively flat
> surfaces where wheel travel isn't a factor (high spring rates).  struts are
> considered the best solution for times where more wheel travel is
> required/desireable.

The classic answer is that double wishbones allow the suspension
designer to control the camber gain curve, while MacPherson struts have
a camber gain curve that is linear with body roll.  

In plain English (or its American sub-dialect, anyway): Camber is the
relative tilt of a wheel and tire, measured against the surface of the
roadway.  If the top of the tire tilts out *away* from the vehicle
centerline, that wheel is said to have positive camber; if it tilts
*toward* the centerline, it has negative camber.  A change in camber is
camber gain; a dynamic change in camber as other conditions (say, body
roll) occur can be plotted in two dimensions as a camber gain curve. 
Here's why this is different for the two different suspension designs.

For an exercise to see how this works, hold out your arm, palm up, so
that your fingertips touch the computer screen.  (If your boss questions
you, tell him it's a carpal-tunnel exercise. :-)  Now bend your elbow
slightly so that your hand moves upwards, without leaning forward or
back.  Notice how, after only a few degrees of bending, your fingertips
have moved away from the screen (assuming the screen is vertical; try it
on a wall if you like).  That's an important concept -- that a straight
suspension element, when it pivots, changes the distance between the two
vertical planes at the pivot (your elbow) and the opposite end (the
computer screen).

Here's the next key: for the same number of degrees of pivot, someone
with a shorter forearm will move their fingers *farther* from the
computer screen than someone with a longer forearm.  And that's the key
to double-wishbone suspension.  Here's how it works in a car.

As the car rolls to one side during a turn, the upper and lower
wishbones both pivot the same amount.  However, in most cars, the lower
wishbone is longer than the upper.  This means that as the car leans,
the outer end of the upper wishbone moves closer to the car's centerline
than the outer end of the lower wishbone.  

That in turn means that the wheel/tire is tilted in towards the car's
centerline, or has more negative camber.  That difference is the "camber
gain."

So by manipulating the length and at-rest tilt of the two wishbones, the
suspension designers can control how much camber -- and whether it's
positive or negative -- the car's suspension dials in at different
degrees of body roll.  The relative change in camber when plotted
against body roll is the "camber gain curve."  A performance car will
typically be designed so that the tire is always pretty close to
vertical compared to the road surface, because that keeps the tire's
contact patch as large, and as evenly weighted, as possible.

Now, I explained that one first because it makes the MacPherson Strut
easier to visualize.  Basically, in a strut, there's no upper A-arm --
no person with shorter forearms.  The top of the strut is fixed to the
car's unit body, so as the car leans in a corner, only the lower arm (or
lateral link) pivots; the hub to which the wheel is affixed moves
upwards in a straight line.  (There's a little lateral movement to
compensate for the fingers-moving-away-from-the-screen effect, but no
tilt of the wheel.)  So in a MacPherson strut-equipped car, the only
thing that affects the wheel's camber is body roll, and wheel camber
changes as a linear effect of body roll; more body roll provides more
positive camber.

Other considerations: MacPherson struts are lighter and more compact,
and less expensive to produce in volume; however, the replacement parts
typically are more expensive for the consumer, because the strut has to
combine the job of shock absorber and upper suspension link.  (The next
time you gripe at the cost of new struts, compare them to the cost of
new shocks and new upper A-arms on a car so equipped.)  MacPherson
struts also typically take up less width in the car, or more
specifically move it lower down because there's no need for an upper
lateral link (wishbone).  Like transverse engine/transaxle combos in FWD
cars, the MacPherson strut is mainly a packaging issue.  

Also, as a very minor point of technical arcana: when a MacPherson strut
is used at the rear of a car (regardless of the location of the driving
wheels), it is known as a Chapman Strut.  This is in honor of Anthony
Colin Bruce Chapman, the designer and founder of Lotus Cars.  The Lotus
Elan (the original one, the 26, not the M100 of a few years ago) used
struts at all four corners, because it made the cars simpler to
construct and also lighter, and for the minimal amount of body roll that
a 1500-lb sports car will be subject to (or that its owners would
tolerate), the struts (MacPherson *or* Chapman) work just fine in a
performance application.  And I've heard they don't work too badly in
the ur-quattro, either...

Hope this isn't too much or too little for the discussion, and that the
elbow exercise doesn't cause anybody too much ridicule from your
co-workers!

--Scott Fisher