14 replies to this topic

[Chris Wilson]

Not sure where top place this post, so please feel free to move it!

I am as much, or more, interested in chassis technology as engine technology, and I have often wondered what would be required to give meaningful aftermarket ecu control to an active rear diff in a 2 wheel drive car used on circuits? I have a 2 WD Skyline that I want to add an Skyline R34 V spec active diff to. This uses an electric pump and ecu control to alter the plate pressure in the diff. Has anyone any links or info as to practical control of such a thing? Could a Motec M800 run it, or would I be looking at something totally different? Initially I would like to know how the Nissan ecu controls it, I know the 4WD cars have a 3 axis G sensor and steering angle input sensor, plus some brake effort / ABS input. Looking to reduce tight corner, off power understeer whilst retaining good corner exit traction and straight line traction. Thanks.

[kikiturbo2]

what type of rear diff is it? does it have one hydraulic clutch, as in "active" diff lock, or is it an AYC or active yaw control diff that can send power to the left or right and has two clutch packs.

On some mitsubishi evolutions we have the AYC rear diffs but as I am aware there are no programmable diff controllers, we can only use what is already in the car or buy an upgraded controller from Ralliart.

However, the mitsubishi centre diff has an active lock, with one clutch pack, and there are programmable diff controllers available from motec and other manufacturers, to be used in GrN Rally...

[Greg Locock]

Not sure where top place this post, so please feel free to move it!

I am as much, or more, interested in chassis technology as engine technology, and I have often wondered what would be required to give meaningful aftermarket ecu control to an active rear diff in a 2 wheel drive car used on circuits? I have a 2 WD Skyline that I want to add an Skyline R34 V spec active diff to. This uses an electric pump and ecu control to alter the plate pressure in the diff. Has anyone any links or info as to practical control of such a thing? Could a Motec M800 run it, or would I be looking at something totally different? Initially I would like to know how the Nissan ecu controls it, I know the 4WD cars have a 3 axis G sensor and steering angle input sensor, plus some brake effort / ABS input. Looking to reduce tight corner, off power understeer whilst retaining good corner exit traction and straight line traction. Thanks.

The only relevant experience I have is from writing and simulating ESC code for road cars.

First thing I'd do is fit a yaw gyro, and strain gauge the rear wishbones so I could pick up tractive forces. It sounds like a simple device to control electrically , and a rather hard one to optimise (ie write a control strategy for).

I'd guess an open diff under braking and locking it when accelerating would be a start. If you can get the track time then a knob for the driver to play with might be the quickest way for homing in on a good strategy.

[cheapracer]

I'd guess an open diff under braking and locking it when accelerating would be a start.

Run a wire from your brake light switch

[kikiturbo2]

Run a wire from your brake light switch ;)

or just install a Torsen diff..

[Paul Vanderheijden]

or just install a Torsen diff..

Perhaps I do not fully understand Torsen differentials, but in my experience in order for this type of differential to work correctly you unload a wheel. So yes, on braking it might be an option, but even then some level of limited slip may be of use, as it helps in minimize/delay single wheel lockup when braking near to the limit of adhesion (in racing this is the idea I presume) . However, in the corner if the inside wheel unloads, then all forward traction is lost, as there is no sense input to bias the differential from the unloaded wheel.

Perhaps it is the turn-in understeer, with a limited slip, that the writer is trying to limit or negate?

I will stand corrected if I am interpreting this incorrectly.

Edited by Paul Vanderheijden, 04 October 2009 - 18:28.

[kikiturbo2]

you are correct of course...
Torsen is good in a way that it locks in a nice manner.. and is open under braking (although looking at the way it works I'd say it will "lock" under engine braking..)

you can always go for a 1 way plate diff, that one will not lock under braking and will be locked under acceleration..

I have driven various Torsen equipped cars and for road cars I'd say they are perfect.. They of course will not work well with limited droop and stiff suspension.. which you will find in race cars.. Also, a plate diff, esp a 1,5 way plate diff ("half" locking under braking) is nice as it reacts immedeately...

I am currently driving a 4WD car with an active lock in the centre diff and active diff in the back (that can transfer torque left and right at will) and I am amazed at the vastly different vehicle response you can get with different diff programming..

[Fat Boy]

Looking to reduce tight corner, off power understeer whilst retaining good corner exit traction and straight line traction. Thanks.

Slow your hands down.

[gruntguru]

Slow your hands down.

Fiddle brake.

[Tony Matthews]

more power

[gruntguru]

More torque.

[Fat Boy]

Tony / Grunt...he said "off throttle"

Along those lines, to the OP. When you are 'off throttle', that doesn't mean 5%, it means 0%. It also doesn't mean jabbing at the gas. This is a bigger deal than it sounds. Also, you have a car that will accelerate like a rocket. Make sure you have enough speed scrubbed off before initiating the turn. NEVER ADD STEERING AND THROTTLE AT THE SAME TIME. Have a little patience. Let the thing rotate. Then full throttle.

Edited by Fat Boy, 06 October 2009 - 02:38.

[cheapracer]

NEVER ADD STEERING AND THROTTLE AT THE SAME TIME.

The golden rule - more is less, less is more (throttle/steering).

There is no mention of if it's an open diff now or not, if it's open now and it's doing what it's doing it's probably just an overly stiff bucket that needs sorting out first and that no diff is going to cure.

I'll take the feel and gradual take up of a well set up clutch pack LSD over a Torsen any day, Torsens are vulgar in feel but they last a long time, are cheap and do a basic job so are very viable for production cars and budget racing.

[StressedDave]

Coming back to the original question though, in terms of understeer reduction, other than switching the diff fully open when cornering with partial throttle, you've not got a huge option to 'improve' matters above and beyond what you've already designed into the car in terms of things like roll moment distribution, damper rates and the like.

If you were running a diff similar to the Mistubishi AYC where you can send torque from inside to outside, then you'd have more option for adding a yaw moment to change the attitude and behaviour. In terms of control theory, back in the early-to-mid 90's I had a research vehicle which used tank steering gear to have a controllable differential (it was 4wd using a H-configuration rather than the more usual I-configuration but that's by the by). We found that open loop control based on forward speed and steering angle was more than enough, although for racing I'd imagine adding Greg's yaw rate sensor and strain gauges would enable much more optimisation and driver confusion.

Edited by StressedDave, 06 October 2009 - 10:23.

[gordmac]

Tight corner power off understeer could very well be more to do with the driver than the car as some above have indicated. Go in a bit slower and smoother and have patience. Or use the handbrake!