Yeah, this one is the same basic idea, but the planetary gears are set up differently. The idea is the same, but the way it is setup allows more planetary gears to be added, which will improve strength.

Just looking at it, it looks like the forces acting on the planetary gears will also be more manageable. In the T-1, it looked like you could snap the pins if you applied too much sudden power. This would be makedly harder to do with the design of the T-2. It also looked like the T-1 was applying torque to a relatively small survace area on each axle.

Anyway, remember I said that the whole system is spinning, and the only time the planetary gears turn is when the axles are moving in different directions. This is also true with the T-2, but the planetary gears are set up in parallel to the axle, instead or perpendicular. The planetary gears are locked to each other on the 'tire side' of the gear that is attached to the axle.

The problem is that the first picture is really confusing.

Look at this one:



See the first two gears you can see on the right side? The closer one drives the further one with the small gear at the top. The wider teeth at the bottom are connected to the axle. If one axle is turned with respect to the other, the axle drives the 'fat' side of the planetary, the planetary drives the 'skinny' end of the neighboring planetary, and the 'fat' end of that planetary drives the other axle in the opposite direction.

When the differential spind, the driving force is applied to the planetary gears. These planetary gears then apply force to the axles.

The friction preventing rotation is between the 'fat' end of the planetary and the axle, just like it was in the T-1. Torsen can control the amount of friction by changing the angle of the gears on the axle.

As, RT pointed out, basically the same concept as the T-2, but with a better location for the planetary gears, and a more compact design, which comes in handy for smaller cars.

 

Geez.. I forgot the original question.

Yes, it works at all speeds. It is more pronounced when there is more torque applied to the diff, so it will only be in effect when accelerating and at speed. The faster you go, the more the diff will dig in. The whole effect is effectively disengaged when coasting.

 

For those in the dark about diffs in general I present the following waffle:

A differential allows the torque to be shared between the two rear wheels. They are required because when a vehicle turns, one wheel (the outside) turns more than the other.

The problem is that if one wheel loses traction all the torque goes to it and is wasted. You're bogged basically.

So some clever people came up with a way of locking the two wheels rotation together. Some of the cruder methods used are the CIG locker. This is where you weld the guts of the diff together. Still used in some forms of racing.

Other systems use a screw system that allows a wheel to spin a certain number of times, all the time winding together a series of clutches. When the clutches lock the diff is locked.Not sure how it becomes unlocked though?

The Torsen T2 uses planetary gears which although allow torques sharing, also resist it to some degree because of the type of screw gears used.

Another type of diff which I first saw in Nissan Pulsars in the early 90's is a viscous coupling between one drive shaft and the diff cage. It's basically two impellers in a viscous fluid (transmission fluid). Because of the nature of the fluid there is a resistance to movement between the two impellers so the diff is to some degree locked (much like the Torsen setup).

 

I don't know anything about clutched diff's, but according to the Torsen document:

"Many limited-slip differentials provide for pre-loading friction clutches to oppose the transfer of torque between drive axles. This firctional pre-load represents a particular minumum magnitude of resitance which must be overcome to permit any relative rotation between drove axles which may interfere with the operation of anti-lock braking systems. Also, since frictional forces are continually active to resist differentiation, the friction clutches tend to wear, resulting in a deterioration of intended differential performance."

"Another known approach to modifying the operative connection between drive axles is to provide for resisting differentiation as a function of the speed difference between drive axles. It has long been appreciated that undesirable wheel slip is associated with very high rates of differentiation. Differentials have been designed using fluid shear friction, which respond to increased rates of differentiation. The obvious problem with such 'speed sensitive' differentials is that undesirable wheels slip has already occurred well in advance of its detection. Also, the fluid shear friction designs generally rely on the changes in fluid temperature associated with high differential shear rates to increase resistance to differentiation. However, similar temperature changes may be associated with extended periods of desirable differentiation, or may be influenced by changes in ambient temperature, so that resistance to differentiation may bery throughout ordinary conditions of vehicle use."

So, clutch based diffs wear out quickly, and fluid based diffs are not consistent and have slow reaction times. The Torsen diff should virtually last forever, and is not affected by temperature or prolonged use.

Of course, they are selling something....

 

Hey Sparky, nice job! Thanks!