[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: A New Look at Torsens

Dave E writes:
>no, while the torsen starts with a static 50% front/rear bias ratio (not
>true for the locked centre - see below), the torsen will progressively shift
>torque to the rear until the bias ratio is reached and then increased slip
>will be rewarded by wheel spin.  don't forget that the accelerating vehicle
>has a rearwards weight bias as well, so the torsen will result in
>significantly less understeer than a generation 1 car.

Up to a point.  The problem is Dave, that that Bias Ratio isn't correlated to 
any type of specific chassis design.  It's correlated to Slip Angle 
differences only.  Remember, you need to interpret all the actions of the 
torsen in terms of U and O.  If you don't, the above gets confusing.  Bias 
Ratio rear varies with EVERY turn, and with every slide or skid.  Lets be 
clear here, the maximum (what you mean to say above) Bias Ratio doesn't have 
to be acheived to get O.  Nor does the maximum Bias Ratio front have to be 
acheived to get U.  The torsen *might* have less understeer than a gen 1, it 
might have *more*.  Depends on a bunch of factors.  How, when and where you 
turn  is included.  Variables that just don't create U-O-U while turning

>the open centre will fix torque split @50% and when (front) traction is lost
>(spinning inside wheel), the torque to the rear will decrease to the amount
>supported by the front (i.e. keeping the 50% front/rear distribution).
>the fixed centre will have a static front/rear torque split mirroring the
>weight distribution (fr audi's usually 60% front, 40% rear) resulting in
>more initial understeer but, as weight is transferred backwards, so is
>torque.  it mirrors the torsen quite closely until the point where
>front/rear slip differences becaome enough to cause the torsen to hit the
>bias ratio.  at this point the locker has no option by terminal understeer.

That is unclear.  The first problem is that we need to assume acceleration 
for a given radius.  The locked center will understeer, because the axles are 
locked together.  

>the torsen will not allocate any more torque rearwards at this point, and so
>has less understeer than the locker.

It *CAN* have less understeer, but if that rearward torque shift overloads 
the traction in the rear, the torsen will have MORE understeer than a gen 1 
locked.  Now we are getting close to agreement.

>a fundamental misunderstanding in this debate is that the locker supports
>100% torque shift, wheras the torsen will limit torque shift to around 75%
>(depending upon design).  the open centre can't shift torque at all, but is
>locked at 50% to each driveshaft.

"A fundemental misunderstanding in this debate is that the locker supports 
100% torque shift..."  Dave, on the same cf this is incorrect, and btw, ALL 
torque charts and methodologies in your torsen reference paper make cf a 
constant.  The above is only correct if you *want* to make a wheelbase v cf 
argument.  Not the way *any* paper has been presented to date, your 885140 

I'm most intrigued at the conclusions.  I don't disagree that they *can't* be 
true Dave.  Just disagree that they are *always* and *necessarily* true.  In 
the case of the torsen, those points affect the chassis dynamics while 
turning, a predictability problem the locker just doesn't have.

Since I answered the questions posed by you, would you please return the 
At your convenience, thanks

Scott Justusson