Oh well, that's what I get for trying to post during a meeting.

The Torsen/Quaife worm-gear diffs have their own set of strengths and weaknesses. The diff does not control wheel speed, it instead transfers torque away from the wheel that resists less. It will in fact tend to make the wheel that resists more spin faster, until the limit of its torque-biasing ability (most around 3.5:1) is reached.

This tends to make torque-biasing-diff-equipped cars feel 'looser' than cars equipped with conventional friction diffs, probably more so even than the Viscodrive type (though I've never driven one of those that hard) and you can often run less rear roll stiffness.

Most torque-biasing diffs (Quaife, TrueTrac, most Torsens) do not have friction elements, and if one wheel offers no resistance at all (off the ground or glare ice), the diff will go 'open'. Some, mostly targeted at serious track use (Gold Trak/Black Gold, Torsen T-2R) have friction elements to provide a little internal 'drag' to prevent this. Reputedly the Quaife can be shimmed to create this effect as well, but I've never seen it.

One other characteristic can be an issue in track applications: they generate heat. Worm gears generate wiping friction, and the torque transfer is effected by gear-stacks tightening against the case. So...in continuous hard use - that is, half-hour track sessions, not back-road blasts - you may need a pump and cooler to keep the diff from welding itself into an expensive spool.

I've got a Quaife in my SHO (I had the second one to actually hit the road, some would say "a $2500 car with a $1200 diff", there's probably a hundred or so out there now) and a Gold Trak in my track Mustang and I've been ecstatically happy with both. I suspect that a Quaife in the M5 would help reduce its tendency toward understeer and turn-in 'push'.

Last edited by JEM on 24th March 2002 at 06:11

Very interesting question! Thinking out loud here....Both systems react to wheel slip - the quaife by transferring more torque to the non-slipping wheel, the DSC by reducing power. If the quaife senses slip and ramps up this torque transfer sooner than the DSC kicks in, it should allow more power to be transmitted without slip, and the DSC would stay out of the picture. If the DSC reduces throttle sooner, the slip stops and the quaife can't do its job. I really don't know which of these cases would dominate. On the other hand, you can turn DSC off, and then the quaife is free to do its job, and you should come out ahead.

John -Let me test my thinking with you. My understanding of the diff's impact on turn-in says a different phenomenon is going on than mid-corner (you've mentioned both). The former has to do with resistance to slip (which is needed as soon as the car turns) in "overrun" - i.e., wheels driving driveshaft, as opposed to other way around. I know some diffs are designed to run free in overrun, which I assume would make turn-in easier. And I assume the clutch-type stock unit resists the acceleration of the outside wheel a bit, creating more drag on the outside and hence some understeer.

In mid-corner (some power on) both diffs xfer torque to the outside wheel because the inside wheel is lightly loaded and begins to slip. If the quaife gets more power to the ground than the stock unit, then two forces come into play that offset each other to some extent: 1) Better hookup means more rearward weight transfer which increases understeer and 2) More driving force coming from the outside wheel counteracts understeer.

SO): 1) Do I have this right? 2) What does the Quaife do in overrun? 3) would you say that the mid-corner behavior will only differ if the required torque transfer exceeds the stock unit's ratio? (i.e. tighter turn, more off-camber, whatever) Or is there more to it?