Effect of Differential settings on setup

Hi Tim,

Do any of the courses cover the interaction/relationship between diffs and chassis/suspension? Or how driving styles need to be adapted for different cars/setup/regulations (that's probably more the Driver Training Courses)?

I mean if you look at an R5/Rally2 car, they take tight corners like they're on rails but the regulations mandate that they have no center diff (just a disconnect from rear diff for handbrake). Now, ordinarily if you run a very tight (or no) center diff the car will want to go straight on and have a tendency to understeer in tighter corners. If you look at the gpA era drivers would have to 'Scandinavian flick' cars to get them to turn in to tighter corners.

More modern machines have longer travel (and probably softer) suspension setups and appear to have more dive in the suspension but the driving technique for these cars is not what you would expect.

Far from braking deep into the corner in the hope of unsettling the rear and getting it to turn in, the braking is done in a straight line before the corner and the driver (apparently) needs to get on the power early and drive the car through the corner.

So in the case of a tight or no center diff, what is done with the setup to combat the understeer this would usually induce?

Or, where preload and ramp angles are adjustable in any of the 3 diffs of a 4wd car, if the driver reports understeer during testing (not so much on event) would you (let's say on asphalt where traction may not be as much of an issue as on gravel) back off the diffs first or mess with ride heights and roll bars to try and redress the handling balance?

Thanks.

Chris,

Great questions that I would also like to hear a studied response to.

In my rally car (GD Subaru STI) I run the center locked with a handbrake interrupt, 1.5-way front, and 2-way rear. The front has a bit more preload than the rear so it helps pick up sooner (I can't adjust ramp angles) and prevents excessive oversteer. This setup works great on everything but a load speed corner where I either handbrake or just pull the handle enough to trigger the interrupt to unlock the center diff. I've considered reconfiguring the front to be in the 1-way configuration but I do like the added grip in high-speed corner entry. The problem gets worse when it's wet/muddy.

I'd love to reverse engineer an R5 to see how they do it. Dumb diffs all around but you're absolutely correct they're on rails. Looking forward to Tim's response.

Hi Sam,

On Gravel I think it's fair just to run the centre diff locked with handbrake release. I too run Subarus and I've compared on tarmac and having it locked does two things, it stabilises the car under braking and prevents locking the rears if your bias isn't perfect but it also promotes understeer. I found the same with a Mazda 323 GT-R I used to run when the centre viscous LSD got 'cooked' from handbrake use and locked itself up.

It's an 'unusual' choice to run more preload on the front diff than the rear on any surface.

As I said, R5 cars have no centre diff but they also have a lot more travel and I can't help but think that weight transfer and dive/squat/anti-dive play an important role in redressing the balance.

I too eagerly await some more enlightened input.

I was incorrect and misinterpreted a chart I had in the back of my head. I have a Cusco Type RS in the front and a Type MZ in the rear. Higher preload in the rear but the front comes on quicker.

That would just suggest higher ramp angles in the front diff.

I do mostly tarmac rallying, in recent years I have used a standard synchro 6 speed which I believe has an AP Suretrac front diff. When I previously ran the older 5 speed boxes I used both open and helical (torsen) front diffs and have to say it's only under some circumstances I even noticed the front LSD. In slippery conditions when the car was sideways but you kept the power on, the torsen diff did help to pull the front of car straight more; that was about the only time I noticed it.

The gearbox for my new car (Modena sequential) has all 3 mechanical plate LSDs so I think using the handbrake is going to be out of the question. I would have preferred to retain the DCCD but the standard drop gear splines are known to twist and it saves on a bit of wiring not to have the DCCD.