RE: Arachnophobia (long)

[Sargent Schutt <sargent@novagate.com>]

Scott, Jeff, Dave(s), et al:

I can't say I've ever been 'bitten' by the Torsen spider, as that
analogy infers a sort of 'surprise' at the event. But perhaps I fall on
the "duh" side of this rather than the "huhh?"(as QSHIPQ put it).

Anyone who thinks the Torsen center diff is 'all that' and more has
never done the following, bitten or not:

In moderate rain on pavement, enter a sharp 90 degree corner, trail
brake hard and late enough to swing the rear end around a little (rear
wheels are sliding sideways a little), as you near the apex, transition
to full throttle. The Torsen will almost invariably think the fronts
have less traction than the rear (rears are sliding sideways and thus
turning slower) and send the rears significant power, furthering
oversteer. :-) As the turbo spools and the rears break loose, the Torsen
redistributes the power back to the front. Since the suspension is
loaded (you're still cornering at 10/10's w/oversteer, full throttle
now) and the fronts are getting even more power than the rears were a
second ago, the inside front will break loose. :-( The car will make a
sloppy jerk back into line as the Torsen has shut down the party in the
rear and overdosed the inside front wheel (no more oversteer, car
flattens out, inside front settles down). I am really perturbed by this
sequence of events. It is no fun when you want to play 'Dukes of
Hazard'.

I've done it in my 90q 20V and my 91 200q. Same results, though the
dynamics of the respective cars are different (not 'nuf HP in the 90,
and a shorter wheel base).

A non-Torsen car (like my 86 5ktq) with the center diff locked is much
more predictable, b/c even if you break the inside front loose the rears
are still getting half the power, and so you can continue to control
your slide.

In the snow, you can still slide the type 44 car easily as you can
overpower both the fronts and rear; in the rain it is tough to keep both
front and rear overpowered for long).

Torsen front or rear diffs would be nice, though (with maybe a 65/35 max
split?). I just dont' like the Torsen center. I despise it for the same
reason I despise slushboxes - I can't control it. Unless I can control
it, I can't be certain what it will do next, except in as much as I have
control of individual tire adhesion via steering/brake inputs. Since
front/rear torque split is so critical, I think it is best left locked
or cockpit-selectable.

Torsen control note: I've found, whilst executing 'perfect' doughnuts**
, that you can keep more power in the rear by getting on and off the
gas. As a result of the Torsen behavior in my above statement (the 90
deg rain corner) I find that blipping on and off the gas will
continuously re-fool the torsen into sending the power to the rear. It
seems to be biased to default to front torque if both ends are loose, so
one must effectively re-set the torsen by getting on and off the gas.

This is a neat thread, and while I've held my tongue in deference to
more knowledgeable techno-folks, I'm curious if my 'layman's
observations' corroborate the actual operation of Torsen: although the
default bias is 50/50, in my wet corner scenario above, the bias will
seem to go to the front and stay there until the car levels out and the
inside front wheel stops spinning, at seemingly 70/30 F/R.? Is it so?
True or false?

Regards,

Sarge

91 200q TAP
86 5ktq IA
former 90 90q 20V TAP

"Perfect Doughnut": you meet your intitial turn-in tracks perfectly at
the end of one 'rotation' and the resultant doughnut is perfectly round,
not oblong - repeat ad infinitum. I think this should be an olympic
event. We could have figure eight events, doughnut events, creative
righting (no four letter words, of course), etc.