RE: How do you cause a Torsen spider bite? - Long

[QSHIPQ@aol.com]

Eric writes:
>Ok, I have been following this thread now (to the best of my limited
>engineering ability, at least) for quite some time.  I still have not
>figured out what exactly causes a bite to occur; that is, what would I
>have to do to create the bite?

*ALL* it takes is the right set of variables - the understatement of the 
millenium.   Maybe a read of my "hot turn in a quattro", or Jeff G's 
description of the "bite" can help.  

>I have been to the track at Texas World Speedway several times now, and
>cannot recall any Torsen-based Audi spinning due to unpredictable
>behavior (some A4s have spun, but that has been due to driver error,
>and I won't name q-lister names.... ;).  I have driven my father's
>A4tq on that track, and never felt like the car did anything unpredictable.

A couple of thoughts.  1) I don't believe in driver error.  What was the 
error specifically?  I got an email from the Road America Qclub eventmaster, 
50 spins in a single day, but remember it was raining, but that's a lot of 
"driver error" especially when you look at the cars center diffs.  Many write 
off the error, when in fact the torsen is what got you into the trouble to 
begin with (once you start a spin, torsen is your worst enemy).  The easiest 
way to think of this whole phenomenon is this:  If you are accelerating and 
turning, the tosen is allocating torque at a bias ratio "other" than 50f/50r 
due to slip angle differences across the center axle.  The real "bite" 
question is, whether this allocation is "appropriate" for the given external 
variables in the turn and the variables introduced by the driver (steering, 
left foot brake, accel), and whether the allocation and conditions are right 
to make U-O-U character "significant" to chassis dynamics.  Some archive 
reading will show that those variables are pretty infinite.  When you read 
any white paper on the torsen operation, they all are for "given" sets of 
variables (get Dave E's referenced SAE 885140 for more on the 80q chassis).  
Variables change, so does the appropriate (or inappropriate) allocation of 
torque by the center diff.  Remember slip angle differences across a center 
axle is *not* a traction argument.   A spinning wheel, or one that is about 
to spin is.  Slip angle and traction are two variables, however, the torsen 
device uses either and/or both as identical inputs to torque allocation.  As 
can be read in 885140 the loss of traction is the torsen priority over slip 
angle differences, but the reality is, you have both working in a given turn. 
 Maybe for you, maybe not, maybe together, maybe not.  

Think of these Torsen truisms:  
*  A torsen center diff car sliding sideways with both wheels straight ahead, 
*doesn't* have a traction problem, it's the same Bias Ratio as a straight 
line.  
*  A torsen car that's turning on dry pavement doesn't have a traction 
problem either, but *all* torsen cars will allocate torque at maximum Bias 
Ratio rear at a given radius while accelerating. 
*  In car that is sliding sideways with the identical opposite lock steering 
angle as you entered the turn, the torsen allocates the same BR, until a 
wheel spins faster (or is just about to) than another, then it reallocates 
torque regardless of slip angles.

Any confusion in the above?  The torsen is too dumb to know the difference 
between relative slip angle and traction.  And it's too dumb to know that you 
know how to control a slide.

>I am certainly not at the driving level of Scott or Jeff, but I am
>confident that I was at least at 9/10ths of what the car could do.
>Likewise, several friends in A4s and one in a Coupe quattro have driven
>that track on many occasions without complaint of bite-like behavior.

Interesting, I drove Brendan Rudacks A4t at Pueblo with lister Ben Howell in 
the car, and I got bite within 2 laps (four distinct U-O-U-O) w/o 9/10ths.  
Brendan got bitten several times that same day.  I haven't sat in a car with 
you, but some of what Jeff and I discussed is the fact that racers use a 
specific technique in turning, that may constitute a more *bite* prone 
outcome.  I don't know what your 9/10ths is, but I've been bitten at 7/10ths 
on the dry many times.  Bottom line is this, if you follow Jeff's "bite 
description", you will get results.

>So my question, in my quest of trying to understand this issue, is
>     Why _weren't_ we bitten?  What are we NOT doing to cause the bite?

Again, there are plenty of archives regarding this (I called it a statistical 
non-event).  My guess is that (again, I've never driven with you) you allow 
for the torsen intervention by not taking the turn as hard as you 
should/could.  If you brake, turn in, then accelerate thru the apex, you 
should be able to unwind the steering wheel while adding throttle.  This is 
the classic race turn.  You are decreasing relative slip angles, which 
results in the ability to add traction.  If during that turn, you have to 
"re-turn" the wheel you may not ever get bitten.   But others will also be 
faster thru the turns than you.  If you are using the classic race turn, 
there really is NO ability to re-steer, you have used up those slip angles 
with traction, you overload the tires, you spin.  

The problem is, while you are executing the classic turn, the torsen is 
*also* reallocating torque based on slip angle, AND reallocates again - 
suddenly and with extreme prejudice - if you lose traction on any given wheel 
while turning.  That is also in 885140.  

Bottom Line:  It *can* happen.  It doesn't happen all the time.  Cf plays the 
biggest role in whether the bite hits you enough to spin or ditch, cf is the 
same as adding HP.  That means all torsen cars can get fooled into 
reallocating torque based on slip angle, potentially upsetting the chassis 
dynamics while turning.  Don't let the comforting feeling fool you, since the 
torsen was introduced (and this discussion is 3 year old too), there is only 
"rumor" that audi ever used it in any of their racing efforts.  For good 
reason, it's an absolute traction device that gets fooled by slip angle 
differences.  That's not a comfortable feeling to any race driver.

My advice?  Take a ride in a car with someone that has experience in the 
phenomenon.  I myself found that just the repeated lapping (same lap times) 
got the bite, my only explanation being hot tires changing the effective cf.  
Wish I could help you more, I know I could show you.   But I think the real 
answer lies in the understanding of the device while turning.  Dave E might 
be able to give you more insight soon.  Or you could get 885140 and read it 
carefully.  The bite is clearly an accepted phenomenon, just few authors of 
sae propaganda papers want to brag about the "dark" side of any engineered 
device.  

HTH

Scott Justusson