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

RE: Limited slip diffs



George writes
> The main difference between a Torsen and a regular 
>limited slip are how it accomplishes the task.  A Torsen uses a special 
>fluid inside the special diff.  A regular limited slip uses small clutch 
>packs and springs.  Some high performance ones use gears that bind together 
>under slippage conditions.

A little oversimplified and not really correct.  A torsen is a worm gear 
diff, the fluid type is not relevent to the operation and bias ratio of the 
differential.  See the website I posted for help on this.  I believe you are 
referring to a VC, that is a sealed unit that uses a special fluid with 
shearing v. locking properties.  Several might take exception to the concept 
of a "regular limited slip".  

>Application of the brake trick at higher speed will only result in the loss 
>of speed, not increased traction.  

Not true, if it works for slow speed, by definition it must work at high 
speed.  We are speaking of accelerating at a Bias Ratio, and using the brakes 
to affect the Bias Ratio.  Speed is not relevent to the exercise, only cf and 
slip angle are.  The brake trick at higher speed, can result in the loss of 
speed due to spin.  Ck archives from Mr. Eaton for some SAE references to 
this concept.

>The brake trick is primarily to move a 
>nonmoving vehicle, not to stabilize an already moving one.  

You might want to explain the concept of trail braking a torsen differential 
(or any other differential) during a turn for help on the "brake trick".  
Many racers find it very effective.

>Note the 
>e-brake trick doesn't work in a front wheel drive car, for what should be 
>obvious reasons (except Subaru which strangely use the parking brake on the 
>front (drive) wheels).
Saab is with you too.  

>I am going to restate the diff explanation in terms of a torque applying 
>wheel here, both this explanation and the one in the post below, are with a 
>two wheel drive, rear wheel drive car.  The same principles apply to the 
>center diff in a four-wheel drive car.
Except that in a two wheel drive car, the slip angle accross the axles is a 
constant.  In awd, slip angle differences aren't constant, and can affect the 
Bias Ratio of center LSD's.

>Lets say the ground on one side of the car can support 50 ft-lbs of torque 
>to move the vehicle, and the other side can support 150 ft-lbs.  A vehicle 
>with an open diff comes to a stop on this surface, and then proceeds on. 
>The wheel on the loose surface side transmits 50 ft-lbs, and the other side 
>transmits nothing, and the rest is wasted in spinning the free tire. 

No, and the reason for my post.  If one side can support 50lb/ftlbs of torque 
to move the vehicle, and the other 150lb/ft.  An open diff comes to a stop, 
then resume per your cf.  Both wheels transmit 50lb/ft of torque, the excess 
Trg will spin the 50lb/ft tire.  The total traction cannot not exceed double 
that of the wheel with the least traction (50lb/ft). In the above example, 
100lb/ft goes to the ground. You are confusing Trg with traction and torque.

> Now a 
>vehicle with a limited slip approaches the same surface and does the same 
>thing.  
>The spinning tire still transmits 50 ft-lbs to the ground, but an 
>addition 1/3 of 50 ft-lbs is transmitted to the other (non loose side), and 
>the vehicle moves a little faster, and with better stability.

Again oversimplified, you haven't assigned a Bias Raito.  From your post we 
can see that you claim the other wheel can support 50 + (50*.333) = 
66.6lb/ft.  Which means that you have a pretty low Bias Ratio.  You may want 
to read the Torsen paper further, it explains Bias Ratio well.

 >Now an car 
>with diff locks rolls up to this surface and engages his rear diff lock, 
>The axles are locked together now, and the firm side will transmit 150 
>ft-lbs to the surface, and in addition the loose side will transmit 50, and 
>then spin a little (but no faster than the other axle).  Thus in this 
>example, a open dif will transmit 50 ft-lbs, a limited slip will distribute 
>67 ft lbs, and the locker will transmit 200 ft-lbs.

Nope.  See the paper at  http://www.mindspring.com/~audidudi/Torsen.htm
You are only correct (given your own example) in the locking exercise.

>Now do this with a wheel off the ground.  A open diff will just spin the 
>wheel - 0 torque to the ground.  A limited slip will just spin the tire off 
>the ground - 0 + 1/3 of 0.  The locker will transmit the 150 ft-lbs from 
>the side on the ground+ 0 for a total of 150, and the vehicle moves.  The 
>application of the e-brake provides a little torque for the limited slip to 
>work with.

Not with you.  Watch your toes in example 2.  

George, your own examples indicate that you need to correct some of your 
thinking regarding LSD differentials and their operation.  Searching the 
archives could help.  The only website I have for you is the one I listed, 
but know there are plenty of others.  

HTH

Scott Justusson