the stock diff is a speed-sensing unit. it has no accel/decel values as it has no ramps where the diffaxles run up and down generating pressure depending on the torque applied, as in a torque-sensing clutch-type lsd design.
with the stock unit you necesserily need different wheel speeds over a certain amount of time (meaning: one spinning wheel) to activate the sheerpump which generates the pressure on the clutches. the engagement is smooth... as is the disengagement.
so one wheel spins for a short while, diff starts locking smoothly (and can lock up to 100% which means no matter how big the torque difference left/right will get, it remains locked), then after locking a while there is of course no more difference in wheel speeds and the sheer pump will disengage and releasing the pressure on the piston. so the diff opens up smoothly. if there is still a huge difference in traction l/r it will again start from the beginning... on and on...
so with this unit you will have no corner entry and mid understeer at all (except the amount of preload it is set to.... this is what you feel when turning in low speeds and a lot of customers are complaining about... also here in this forum).
but on corner exit, when hitting the gas quite hard, this unit will not lock immediately but let the inner wheel spin for a while. this is when your opponent with a clutch-type lsd will gain several feet... and you have no "direct connection" (via the gas pedal and the currently applied torque) to your lsd as everything happens timeshifted in a speedsensing unit.
shimming the preload spring is pretty silly in my opinion. i know this is done often but for street and track driving it definitely is the wrong direction of modification. for a drifting only car you can drive high preloads.
and keep in mind that this will always lead to higher wear.
concerning nitriding i am no expert. but for the clutches i would not change them out of manufacturer hardness spec. as the inner clutches are molybdenum sprayed i am not sure if you can simply nitride them?! some guys apply some surface finishing (rem polishing) on it. this has no disadvantage and i could imagine it will extend life and reduce oil pollution (a very important point if you drive a porsche with a gearbox-integrated diff).
i only nitride diff-items that were machined and then have no sufficient surface hardness any more.
anyway, i wouldnt waste time and money in optimising a non-optimum stock lsd-unit. if you have no complains about the stock unit leave it like it is. if you want a higher performing, quicker reacting unit, change to one of the above mentioned.
the drexler gt4 has two rampsets on the pressure plates: 40/50 and 50/65 degree ramps (dont mix this with locking rates given in %!!!)
drexler calculates the angles like zf did (based on the vertical zeroline) and this is the opposite way as os giken does (they take the horizontal line... so you need to subtract from 90 to get the "other" values). so in osg language this would be 50/40 and 40/25.
drexler doesnt give lockup rates (as any other aftermarket lsd manufacturer does not either). i calculated the following:
40/50 degree ramps: ~85/70%
50/65 degree ramps: ~70/40%
in standard it is setup for more decel lockup and i think the lower, higher locking values.
i am currently driving 40% accel, 70% decel. this provides to few lockup on accel for me. so i am thinking whether to rotate the pressure rings so i will have 70 an accel and 40 on decel or go with the 70/85 that the other rampset would provide?!
Last edited by driftflo; 01-09-2013 at 05:15 AM.