[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Torsen post for Gary
In a message dated 10/20/99 2:49:02 PM Central Daylight Time,
> My question is very basic, really. Stan says that it
> differentiates at the BR, 885140 seems to say that the torque
> split is variable when differentiating. You say they're both
> right. How's that? I don't care about the inputs, the Torsen
> doesn't care or "know" where they came from (slip vs traction),
> I'm just concerned with the device's outputs.
Stan's paper is for rwd and fwd applications arguing traction, specifically,
where relative slip angles (to the wheels with a common axle) are a constant.
They are a variable in a center diff.
Look at it this way Eric, Stan says that differentiation means torque split
is maximum BR in a single axle. What this means is that indeed torsen slip
is figured into the turning radius of the car (your point). The problem
becomes when you have crossed the axle where torque is allocated between a
set of steering wheels, and a set of non steering wheels. Now, Relative slip
should be figured (but we know it's not, cuz that's a major wheelbase
argument, remember the Urq and the v8 use the same torsen, I might jump and
say *all* are the same BR) so that each audi torsen car can turn without
changing torque allocation. That's just not the case, 885140, or just turn
your wheel to full lock and accelerate.
It should be noted that even if you can show an rwd or fwd axle that does
allocate torque based on slip angle differences in a single axle (BR max by
turning radius), you need to go to the "next" step, what is the resultant
chassis behavior. In rwd U or O won't change regardless of TBR, and in fwd U
or O won't change regardless of TBR applied. In a Torsen CENTER application,
U and O change exactly because of TBR applied.
The difference between a locker and a torsen, is that a torsen can allocate
torque output to the rear, the locker can't. This is KEY to understanding
slip. The reason a locker U's all the time is that the locker will spin the
tires that are overloaded (in this case the ones with less weight and
turning), cuz it can't change allocation torque, it has no Bias Ratio to do
so, all inputs result in the same output. So slip front to rear increases
disproportionately in a locker, the fronts accelerating will *always* lose
traction first, cuz they have to support steering and traction loads with
less weight than the rears. This applies in a straight line, a turn, a
slide, a correction, any cf (constant). The only time it doesn't satisfy
that rule, is when you have differing cf, or raise a wheel. However, if you
only raise one wheel (as in cornering on high cf) the steering ability just
decreased by 50%, which then increases the understeer.
Eric, all this information is available in 885140.