(LAC) Forgive me father, for I am about to sin

Roy Wendell erwendell at mac.com
Mon Feb 26 16:06:41 EST 2007

On Feb 26, 2007, at 3:42 PM, Taka Mizutani wrote:

> Thank you for the clarification.
> I actually *like* how you can have the car rear-biased and loose when 
> you want to have
> fun and then run it 50/50 in the snow for conservative driving. That 
> seems to work very well
> in that you can get the best of both worlds.
> I think with the yaw sensor addition in '05 (or was it '06?) and the 
> 40/60 bias of the current
> car, the DCCD would even work better and less tail-happy when you 
> don't expect it.

I'm actually the most familiar with the '06 having driven one quite a 
bit. When autocrossing it's the most rear wheel drive feeling system 
I've yet driven although the Nissan ATTESA? is the only system that can 
truly do that. When in auto in the snow though it seems to interpret a 
turned steering wheel and power as "unlock the diff and spin the rear 
wheels like mad". Fun once you get used to it. It doesn't seem to have 
any concept of the idea that the lack of traction means that you don't 
transfer any weight and therefore traction to the rear tires under 
> As a follow-up, when you say the VCs in Subarus fail, how do they fail?

The VC unit locks up solid thus making the differential redundant. The 
car hops/skips/clanks whenever you try to turn just like a part time 
non center diff 4wd vehicle. On one I swapped out the center diff (with 
built in viscous unit) to the tune of $500 plus labor. On the second I 
just ripped the guts (plates and fluid) out of the viscous unit thus 
making it open diff like first gen quattro.

> Do they fail in the same manner that DSM xfer cases fail- seal failure,
> followed by loss of fluid, followed by melting of gears and complete 
> driveline
> lockup? I've had that happen twice to me, both times I got away very 
> lucky
> w/o any loss of control of the car, that's one reason why I won't 
> drive a DSM or really
> any Mitsubishi. That's a fatal design flaw.

Nope, although that's more a matter of an improperly/under designed 
case than the differential itself.

This locked mode of failure of the viscous unit still has me a little 
puzzled as by most accounts it's more common that they fail to lock. On 
the first one I disassembled I found some surface wear of one of the 
plates but nothing like the ball of metal bits I expected to find. 
Quite a quantity of black silicone goo came out as well but I really 
had no way of measuring the exact quantity. As I understand it the 
ratio of goo to air space inside the unit is very critical.


More information about the quattro mailing list