While Torsen diffs have reasonably compelling advantages for many people normal road driving I would like to point out that the Autodelta clutch type diff actually almost eradicates torque steer, and I imagine would work much better on bumpy roads than a torque biasing diff mainly because it can't transfer torque as quickly as a Torque biasing diff..
..but also because it won't "panic" when going over a bump (not neccessarily airborne..) and causing one wheel to get unloaded.
Remeber Jeremy Clarkson and Evo and other magazines raved about how usable the Autodelta diff was on the 3.7 cars, lots of magazines and Jeremy complained about the torque steer from the Quiafe diff in the Focus..
All torque sensing diffs transfer torque pretty much instantly, modern clutch diffs can be configured to transfer it relatively progressively with an initial configurable suddeness and progressive locking up to a configurable amount.
I'm only hoping that the Q2 diff is somewhere near as good (but can cope with more torque) as the AD diff..
The T2-R Torsen racing type diff linked to below, actually adds clutch type capabiltities to the torsen diff!
i.e. it removes the classic major disadvantage of Torsen diffs which is that their behaviour "collapses" when
a) one wheel is very unloaded, e.g. going over a kerb, in the air etc.
or, perhaps less obviously:
b) when the minimum split of torque that it can send to one wheel exceeds the available traction, in this case inside wheel spinup will occur and the behaviour will revert to that of an open diff until traction is regained.
Neither of these is an issue with clutch type diffs since the spin-up is restricted by the friction of the clutch plates, the greater the relative wheelspeed the more friction - this stops all power being uselssly directed to the already spinning wheel.
This is why the T2-R added a clutch pack, to add preload to avoid the classic "torque collapse" scenario.
I don't know if they make the T2-R for our cars but it sounds like a really cool diff but possibly very extreme, with a 4:1 torque bias ratio.
AFAIK the Q2 torque bias is 2.5:1 and I beleive the Quaife torque bias ratio is 2:1 (at least these were the figures for diffs for a Ford).
Anyway next week I will get to try the Q2 diff at the Ring, and have already tried the Quaife and Autodelta diffs on track so should have the full lowdown.
..and I'll get to try Peter K's Madeno Racing diff, which is suspected to be a Quiafe.
Cheers,
TB
Originally Posted by alfasud.ti
Because it produces steering wheel tugging and quirky behaviour when it locks, which is worse for FWD cars with McPherson struts (not a case with Alfas though). Several people claim that it is better for track use because it will never unlock itself if a wheel completely loses contact with the ground as a Torsen would. Maybe that's why Autodelta chose it for their 3.7 GTAm's but I think that it had more to do with copyrights from Torsen and Quaife and not being able to strike a good deal with either of them.
Almost. The Q2 is a
Torsen 2 type diff while the Quaife appears to be similar to a
Torsen 2R, judging by what little detail is revealed in its photos.
Well, the Q2 hasn't proved its reliability yet, none of them has many miles, but in my opinion, if it is as reliable as the Quaife then it is the obvious choice because it is much cheaper.
2006-present 147 GTA + Mods 
