AWD drivetrain info [rallycross]

[spacer]

Hi all,

I always enjoy reading non-F1 tech questions here on F1T, so hoping some knowledgeable people chime in.

I'm looking for decent information regarding the practical design and tuning of AWD drivetrains, but I can't seem to find any "broadly accepted" literature about it. For comparison, when reading into suspension design, there's always a book or 5 that everybody seems to recommend to read-up on to get a decent basis.
But, even better, perhaps there's people willing to add some useful insights/experience into this thread? Any info will be greatly appreciated!

My main point of focus is the effects and design choices of different differential types / torque split scenarios for an AWD rallycross vehicle. We've been racing in pseudo-rallycross classes for over 10 years (100% gravel/dirt tracks), but all of our experience is with FWD cars. Over the years we settled on a salisbury-type diff (metal plate LSD) with some preload and what most people call "1.5 way" ramp angles to allow sharp turn in and good corner-exit traction.

We're currently in the early phases of designing our new car/chassis for the 2019 season which will see us step up to the AWD class.
Our proposed chassis at this moment: front longitudinal engine, central seq. gearbox with centre diff/transfer case, front/rear diffs in separate housings. Ballpark figures 550BHP / 1100kg. Weight distr. approx 55/45 front/rear.

What we're uncertain about at this moment are particularly the centre and front diff types, as well as maybe any difference in final drive ratios between f/r?
Centre: torque biasing, Rear: Salisbury, Front: Salisbury is what my gutt feeling says.
-Perhaps front could be more benificial to use the torque biasing type?
-Perhaps centre diffs in gravel-only situations might as well be 100% locked connecting front:rear 1:1?
-Could it be benificial to run the rear axle at a slightly higher speed compared to front?
-What about viscous centre diffs?

As far as electronically controlled diffs, I'm of the opinion these aren't worth the large chunk of our time when concerned with developing a new car. Waaaay to much other areas where we're supposed to gain laptime for the first year(s) before adding in the complexity of active diffs. Unless we're missing out on *the* holy grail?

(I've been told the (non-WRC) driver-controlled diff settings in subarus and lancers only switch between 50/50, 66/33 and 100/0 torque split depending on which mode (tarmac-gravel-snow) the driver selects. Which would basically render it useless for our gravel-only purpose imho.)

Obviously, we'll test the heck out of the thing once it's ready to go vroom in 1-1.5 years time. But I wasn't planning on buying a lifetime supply of differentials....


thanks for any input!

cheers -Tom

[Greg Locock]

I'm supposed to be working on an electric 4wd off road buggy. My thoughts for traction control, since we don't have a clue, is just to set up pots that adjust the front to rear and side to side bias in terms of the LSD setting and leave it to the driver, at least for the first season. Of course that's much easier done electronically than what you are proposing., as the motor controllers can be set to run in speed following mode that is, no differential action, or equal torque mode, which is an open diff.

[factory_p]

If you want to go cheap, simple and reliable without losing much, I would ditch the center diff. I know several teams run without a central diff in WRX and still perform well. Your time and money would porbably be better invested working on some suspension development or your weight distribution.

[Zynerji]

I would buy a wrecked AudiTT VR6 Quattro with the haldex drivetrain, and start from there.

Computer already has the AWD sorted, and the wheelbase is short and sporty.