spider bites - oh, no, not again!

[Mike Arman <armanmik@n-jcenter.com>]

I've been keeping a safe distance from the ongoing spider bite flame war,
but watching it nevertheless. I guess it is our list's answer to TV's Pro
Wrestling. "In this corner, from the United Kingdom . . . and accompanied
by . . .. " etc.

What I *think* I'm hearing is that certain of our cars, when equipped with
certain drive train components, and when pushed hard around corners
SOMETIMES exhibit a phenomenon as follows: First, it understeers, then it
oversteers, then it understeers again.

If I understand that correctly, when the car enters the corner, the driver
turns the wheel and the car doesn't turn as much as he expects (continues
to go straight), so he turns the steering wheel some more. Next thing that
happens is that the car begins to suddenly oversteer, and dives toward the
inside of the turn. Finally, the car returns to the understeer condition.
The resultant path through the turn is a lurching zig-zag, and the driver
emerges somewhat puckered. This non-linear response to steering inputs is I
think what is being called spider bite.

It has been blamed on the center differential's varying the power sent to
each wheel in response to differing load conditions through the turn - and
it appears that the war is being fought on two fronts: First, does spider
bite exist at all, and second, exactly how does it work? There has recently
been some anecdotal evidence (from a pretty reliable source) that wheel
alignment has major effects on the onset of spider bites.

Unpredicatable oversteer and understeer can be caused and affected by a
number of things. Speed, vehicle weight, tire selection, pavement
conditions, suspension deflection, suspension geometry, worn parts in the
suspension, spring and shock rates, weight distribution of the vehicle,
driver skill, and I am sure the phase of the moon all have effects.

Note the understeer-oversteer phenomenon exhibited by 911 series P-cars and
old Corvairs - both rear engine cars with much of the weight at or near the
rear axle. Both cars seem to handle marvelously, steering lightly and
predictably (which builds driver confidence) until the limit of adhesion at
the rear is lost, and you go spinning off the road (how's THAT for
understeer?). It sounds to me that what is happening here is that the car
is initially understeering, so the driver turns the wheel further, and then
the rear breaks loose suddenly and completely. This is the FIRST HALF of
the spider bite! And with 2WD, the ONLY half of the spider bite you will
get, and not recognized as such, because in a 2wd car, there never IS a
scond half.

On Audi, with all wheel drive, the SECOND half of the bite, the return to
understeer, just might be the RECOVERY from the spider bite! What is being
interpreted as oversteer, the dive toward the apex of the turn, may in fact
be the start of a classic spin-out. (The attitude of the car is the same-
the rear end is too far out both at the start of a spin, and during
oversteer. Except in a spin, the rear end contines to depart,
uncontrollably!) The AWD system (torsen) senses the loss of traction and
compensates for it (and puh-leese don't ask me the technical details of
this) and manages to restore the car to the normal "understeering" condition.

It occurs to me that the speed and degree of response of the torsen's
reaction to the loss of traction affects the driver's perception of the
severity of the spider bite. Does the torsen respond BANG- NOW with all or
nothing? or is the response a gradual feeding in or reduction of power to
the skidding wheeel? If the response is abrupt, the car's change in
attitude will be, too. If the response is very smooth, we may not even
notice it. Are we in fact driving at the limits of the TORSEN instead of at
the limits of the car?

Result: While the driver emerges from the turn somewhat puckered, the car
has been more or less under control all the way around, and has negotiated
the turn successfully - the message the "spider bite" send is sending us is
"SLOW DOWN; you are at the LIMIT for this car!" P-cars and Corvairs don't
send this message - they spin and crash.

If what I am saying here bears even the slightest relation to reality, it
seems to me that most of us, who do NOT drive our cars at or even near
their limit of adhesion, will never experience this spider bite. (Although
it certainly can't hurt us to know what it is, and what it feels like just
in case we DO someday.) The listers who are most vociferous about it are
the listers who normally drive fast and hard anyway, much closer to the
limits of their vehicles. The comment about wheel alignment affecting
spider bites is useful, in that if the handling of your quattro gets
squirrely, checking wheel alignment might be a good place to start looking
for a cure, especially on type 44s, which are known to be sensitive to
wheel alignment.

We now return to our regularly scheduled flame throwing!

Nomex on, and ducking for cover, Best Regards,

Mike Arman