Creating confusion - where's the haldex now?

[QSHIPQ@aol.com]

Dave E writes:
>bingo scott, you've got it.  same torsen, different chassis.  some affected,
>others not.  as the zexel paper states.  let me see, spring rates on the
>type 44 vs the type 85?  weight distribution?  torsional stiffness?
>suspension kinematics?  all different and will never be the same. 

Problem:  Neither will cf.  No one has made the matrix complete, "optimal 
will be a compromise" really rings too true.  In terms of torque split, we 
could certainly argue that a lower swing in torque might be transparent to a 
couple of chassis (if the maximum torque shift rear still creates U, you 
can't have U-O-U..~  in a turn).  However, part of the "optimal compromise" 
is to straight line traction, without question.  I'm having difficulting 
understanding how, once all the chassis matrix is done, slip angle 
differences fooling a torsen are "corrected" in any way.  Torque shift still 
happens, from 78f to 78r when the device is fooled.  How exactly can't the 
device be fooled, is my question? Please identify that matrix, so we can all 
correct our "other"' cars.

 >all
>stated as important by zexel.  hence you have to accept that the bite will
>never happen on some chassis where it can happen on another.

...  "For given road conditions" is the key to discounting the above.  You 
can't have both absolute traction in a straight line AND have it in a turn 
AND have fully functional ABS and address all cf.  So proven.  Given audis 
experience with the device, ABS doesn't make a difference to the argument 
against the device in racing (same cf) or rallying (differing cf)

>i also find it interesting that the engineers who modelled the torsen
>required 6 different variables in order to model it's operation (torque,
>weight, f/r weight distribution, roll, suspension stiffness, tbr) - most of
>the variables obviously *external* to the device, but obviously contributing
>to it's operation.  

No question, I don't disagree with any of the above, but in terms of torque 
shift you can look at torque and cf as the same thing, can't you?   Does this 
preclude the device in any chassis from inappropriately shifting torque, 
upsetting the balance of the chassis?  No, not unless you can say under all 
conditions the car will understeer, more or less is acceptable, but always 
understeer.  To make that "optimal torque split" you decrease optimal 
straight line traction.  Marketing rules, that didn't happen.  Why?  Cuz any 
less of the maximum torque split would have affected ABS operation.  A major 
goal in Ingolstat.  The "compromise"?  Slip angle differences across the 
center diff in a turn causing inappropriate shifting of torque from front to 
rear and back of 56%*Trg.  

>they believe that any less than those 6 degrees of
>freedom leads to an incomplete understanding of the device.  needless to say
>these factors are all different between the type 85 and the type 44.
>as phil has found in practice.  possible bite on the type 44, no bite on the
t>ype 85.happy to point you in the direction if you want to have a look at 
their
>maths...

No thanks (reread the part of "for given road conditions" = for a given cf, 
it might help you here), the only math that is important, is the claim that 
78r/22f is oversteer, and 22f/78r is understeer.  One can't deny that the 
device is capable of either split in ANY torsen audi chassis.  How does one 
design the "6 degrees of separation" (tm- DE), to make that an untrue 
statement.  What I find interesting, is a claim of a 50/50 locked diff 
("capable of transfering 100% torque to either axle") will either understeer, 
or understeer more, not a single report or race driver quoting "oversteer" 
with a locked quattro diff.  I think somewhere in your "white papers" you are 
missing some basic chassis understanding.

>perhaps audi decided that the type 44 would not be used in the same way as
>the type 85?

Perhaps the device is indeed designed to be an absolute traction device with 
ABS ability, turning compromises so accepted.  If the above is true, why 
don't different chassis use different splits.  "Optimal" indeed, screw the 
turns, that's what ABS is for...

>also, your preoccupation with torque shift is a little difficult for this
>lister to understand when you favourite locked centre diff will shift torque
>up to 100% (not 56% as the torsen can).  perhaps you should re-think your
>maths for this case?

That's not a math issue, that's the "extreme" issue for the exact length of a 
wheelbase in a straight line, your trademark.  Please provide ANY 
documentation supporting a locked diff in a quattro, throttle oversteering 
during a turn.  Most audi racers, and everyone here with btdt would be in awe 
of *any* support to this claim in ANY locker chassis.  I've been to Steamboat 
for 5 years in a row now, not once have I seen nor heard of any locked 
quattro driver talk about "oversteering" into a spin, with either the center 
locked and or center/rear locked.  I've been in many torsen cars at the same 
venue, with LTO AND WOTO, what am I missing?  Let's be clear, cf at steamboat 
is the lowest possible on the scale.  Missing math, or basic misunderstanding 
of the effect of center diffs on chassis dynamics, or just a  "marketing 
swipe" at the 100-0-100 concept?

>as for the handling of the sport quattro (with locked centre).  what do you
>think contributed to it's twitchiness?  perhaps we should say that because
>the sport q was very twitchy with a locked centre, all quattro's with a
>locked centre's are twitchy?

Please read Stig's comments on the use of the torsen in rallying.   Amazing 
that the A1/2 quattros with the same center, front and rear diffs and HP, 
didn't get this "twitchiness".  Seems pretty obvious to me, so does the fact 
that shorter wheelbase torsens aren't part of the "correct" matrix in terms 
of U-O-U.  One could look at the other "factors" you listed, and certainly 
argue, that audi sport either tried to address them, or did, and found no 
"optimal design" in the torsen application.

We are back to well chartered disagreements.  Might I suggest we spend some 
time on VC diffs and VC with center diffs.  Your misunderstanding of the 
Torsen is well documented.  Maybe if you understood the center diff VC, this 
all might make sense.  Try Mitsu and Toyota awd systems for great 
applications of the concept.

Love to hear from any locked diff driver about the WOT/LTO oversteer with the 
center diff locked.  Then we'll get a conference call with John Buffum, "Hey, 
John, in your book, here's what you shoulda done..."

Scott Justusson