[torsen] Re: Torsen torque split clarification

OorQue at aol.com OorQue at aol.com
Mon Dec 4 16:41:26 EST 2000


>> It was _my_ opinion that if you adjusted the friction unequally (front vs
>>  rear), you could achieve a static torque bias to the rear so that more
>>  torque goes to the rear under all conditions.  That might stop the
>>  instability of the centre torsen a little.  I'd sure like to try it. 

It's been a long time since I've even thought about Torsens but off the top 
of my head, I don't think this is a workable solution.  Assuming that it 
works -- and I'm not sure that it will -- I suspect it will wear out the diff 
in short order.  The extra heat that it will dump into the gear oil won't 
help things, either.  Porsche installed an oil cooler on the Type 016 they 
used in the 944T and that one didn't have a center diff!  Audi also installed 
an oil cooler on the V8 tranny as well...
 
> I think you are looking at it backwards, Jeff G looks at it correctly IMO.  
> If you have a baseline chassis that is of fwd U character, looking at the 
> above example, you could have 43r/57f +/-21.5 of predictable U, with a 
larger 
> BR.  Drive the car like a fwd car, stop trying to make an fwd car be a rwd 
> car (thinking of how much money audi spent moving the A4 btcc engine 6in 
> rearward comes to mind)

This was also my conclusion.  If you think about it, both the Ur-Q and Type 
44 chassis were originally conceived as FWD and AWD was added to them a few 
years later.  IMO, both cars behave very much like the FWD versions (except 
they have better traction!) and with the engine hung out past the front axle 
line, I doubt that anything short of redesigning the car from the floorpan up 
-- remember the Trans-Am car? -- will ever allow it to behave like a RWD 
without giving up overall grip in order to achieve a more RWD-like front-rear 
balance.   

> Bensingers comments on "...reckons O to be at xx/xx split..." means that 
you 
> have O and U in turns.  This enfatuation with O in a U chassis car doesn't 
> make any sense to me at all, you might have reduced the traction argument U 
> slightly, but you haven't really addressed fully the problem of turning 
> causing changes in torque allocation.  In your example, raise HP or lower 
Cf, 
> you are back to square 1 in terms of effective resultant chassis behavior.
 
Like I said, it's been a long time since I last thought about these matters 
but as I recall, I eventually came to the conclusion that the locked diff, 
while not being the ideal solution -- active diffs definitely appear to be 
the way to go -- is nevertheless the most driveable one due to its 
predictability.  This is especially true for those drivers (like me!) who 
trail brake into corners and/or left-foot brake since the Torsen doesn't 
respond well at all to this sort of driving style. For people who drive up to 
a corner, get off the gas (thus causing the amount of torque to be 
distributed to drop off markedly), turn the wheel and then get back on the 
gas, the Torsen probably works well enough ... and since this also describes 
the way that most sane people drive their cars on the street, they work well 
there, too.

In general, it's only on the track (or during unexpected emergency manuevers 
where my track-honed reflexes take over) that it causes any problems.  I 
wouldn't install one on a racecar because of its lack of predicability and 
given the option, I probably wouldn't have opted for one in my street car, 
either.  However, what's done is done and since my 200q only sees street duty 
these days, it's not worth the cost/effort to backdate to an early-style 
transaxle.

BTW, my "oorque at aol.com" email address will be going away as of Dec. 10th; 
anybody wanting to reach me after then should use "audidudi at mindspring.com" 
until further notice.

Jeffrey Goggin
Scottsdale, AZ 



More information about the Torsen mailing list