[torsen] Re: [V8] differential/torque sensing question

Keith Maddock Keith.Maddock at trw.com
Wed Mar 19 05:10:49 EST 2003


Its OK to suspect an error :) I'm always game to improve my knowledge.

I think we were both slightly off..

I'll forward you the two papers, let me know what you conclude...

I conclude:

The "decreased bias ratio effect" is only in conditions when input torque is reduced.
(coasting and braking with clutch depressed)

The " full bias ratio effect" is good for both forward and reverse and braking as long as there is appreciable engine torque, be it positive drive torque or negative drag torque.

Reference:
SAE 885140,  2nd Paragraph, Page 2
"The main feature of a Torsen differential is
that it is capable of giving a limited slip effect
which varies according to the applied engine
torque. Under power, the torque split is varied
automatically to suit the adhesion available at
the front and rear wheels. When engine torque
is decreased, the limited slip effect is also
reduced to give a safe brake force balance
without ABS and to allow individual regulated
braking cycles for each wheel with ABS."


Then also read  section "VEHICLE BEHAVIOUR WHEN BRAKING IN A TURN", Page 5/6, also Figure 8

This section is a bit of a muddy read, but clearly states/shows that under braking, the engine drag torque is split 75% to the rear and 25% to the front.

―--------------------

RE: "Torsen is at most capable of transmitting 80% of torque to the axle with
  most traction."
Semantics here... perhaps I should capitalize  "AT MOST"
or should say "current engine torque output" instead of "torque" perhaps implied "max engine torque".

In the case that 2% torque to the rear, 8% torque to the front, where's the other 90%?
Its still 80% of the current output...

So I'll restate it.. 
"Torsen is AT MOST capable of transmitting 80% of current engine torque output to the axle with most traction"
Does that work?

―--------------------

I do agree with the theory that newer Audi's have improved weight distributions over the older ones, I also have no hard evidence.

―--------------------

Good point about the handbrake.. 

This is essentially "Poor Man's EDL" across the center differential, this will increase torque sent to the front axle in both the Torsen or standard open center differential cases.

A lot of 4 wheel EDL systems  installed in open or torsen center diff cars basically do this for you automatically....
(This is how I make a living incidentally)


Cheers,
Keith
Currently 93391 Arvidsjaur Sweden










Keith Maddock, TRW Automotive,  Koblenz, Germany
Slip Control Systems, Systems Design, Traction Control
+49 (0)261/ 895 2474    -   -   keith.maddock at trw.com


>>> "Andres Kroonmaa" <andre at online.ee> 10:05:15 19.03.2003 >>>

 uh, I'm sorry to suspect an error, but Torsen design is specifically
 made so that bias ratio CAN be different for onpower vs coasting. Its
 done via different shape of gear toothing and gear friction with washers.
 Forces change direction for different scenarios
 Whether that feature is implemented is question about specific torsen
 unit. Given that abs is sensitive to torque shuffling, I believe
 they really do implement it. If you know that this is not true, maybe
 you can quote some authority?

 "Torsen is at most capable of transmitting 80% of torque to the axle with
  most traction."
 This alone is misleading. Torsen is able to transfer 4 times more
 torque to axle with most traction with reference to axle with least
 traction. This means that if slipping axle is able to support 2% of
 engine torque, best axle will receive no more than 8% of engine torque,
 rest goes into wheelspin. In noslip condition this will be 20/80 split
 and normal differentiation.

 I don't know about older audis or V8s, but newer one don't have that bad
 weight distribution. 60/40 about max. But would need to backup that too.
 But anyway, if you assume that reverse bias is 1.5:1 instead of 4:1, then
 your explanation seems even more correct.

 Interesting thing about torsen is that when stuck in such scenario as
 described, applying little brake causes immediately very notable amount
 of torque to be available at other axle. Specially handbrake. This causes
 torsen to sense traction in rears, and after multiplicating it by bias
 ratio, apply to fronts. I imagine that if Mike had applied mild slipping
 handbrake, he'd get out of situation with style.

 11317 Estonia



More information about the Torsen mailing list