Perhaps Paul could explain why the PnG/GTR2 diff appears to me to operate in the opposite way to the GPL diff.

In GPL I was aligned with the fact that if I increased the amount of diff lock (lower angle) then I would increase the amount that the car would turn inwards at mid-corner when the power was applied.  Due to the outside rear wheel driving it around.

When I moved to PnG I went all the way up to 100% power side trying to achieve the same result and failed to notice any such thing.

Then Manteos provided me with a Cobra setup that had 20% power side diff and there it was.  I now don’t have a clue how the PnG diff is supposed to be working but if I want it to turn under power then the setting on the diff needs to be low.  It’s as if PnG never models the inside rear losing traction. 

I can only do the last stint 8 really for lmp1. I think my last race at Donington is 5ish the day before so i'll need a lie in to recover 

2. It is not fine for a normal road car unless it's hideously underpowered.  You just don't notice most of the time when you're just tootling along.

Your discription of the diff was spot on and nice and straight forward, unlike mine. The one piece that I thought might confuse folks though was the above as it could easliy be taken as increasing drive ramp angle from say 60 deg to 80 deg will increase the locking factor. This wouldn't be correct as a higher angle is actually less locking factor. I know you probably didn't mean
for it to read that way but that is what I would read it as the way you have it. The ramp angle for both drive and coast is inversly proportional to locking factor.

Good write up Niek, the one area that I think your explanation doesn't seem quite right though is the infamous diff.

The Coast and Drive ramp angles work the same in that the lower the angle the more locking the higher the angle the less locking. What differs between Coast and Drive to the feel of the driver is the effect it has.

Higher drive ramp angle (e.g. 70-85deg) = lower locking on throttle application, which allows the car to turn more readily when under the limit of the tyres adhesion but if taken too far allows the inside wheel to spin this can have the following symptoms: on slow speed corners loss of traction and hence poor acceleration out of the corner, on high speed corners it can result in understeer on mid corner and corner exit with throttle application.

Lower drive ramp angle (e.g. 45-60 or lower) = more locking on throttle application, which resists the car turning when under the limit of the tyres adhesion but when approaching the limit of tyre adhesion allows the driver to 'turn the car with the throttle". When taken too far it can result in rapid throttle oversteer. Lower drive ramp gives better acceleration and enables the car to turn on medium to high speed corners easier without having to lift off the throttle.

To set drive ramp angle I usually start with three plates and 70deg, if the car is understeering on the mid to corner exits of high speed corners or the inside wheel is spinning up under acceleration I will reduce the ramp angle a click, if the car is entering throttle on oversteer to easily I will increase the ramp angle a click.

Higher coast ramp angle (e.g. 45-60 or higher) = lower locking power under deceleration, this allows the car to turn more readily when under the limit of the tyre adhesion but if taken to far allows a single rear wheel lock up which can equal instability in the braking zone often ending in a spin.

Lower coast ramp angle (e.g. 20-30) = more locking under deceleration, which resists the car turning into a corner when under the limit of the tyres adhesion. A lower coast ramp angle increases braking stability but if pushed too hard can be unforgiving as both rear tyres will loose traction together (not good).

To set the coast angle I first make some assumptions of the car. The GT doesn't like trail braking as much as some other cars so I tend to brake in a straight line as much as possible and be trailing off the brake very rapidly or be off the brake fully for turn in. With this in mind I aim to make its brake performance in a straight line very strong and hence start with the coast ramp angle at 30 degrees with three plates. This I have found is a good setting for most tracks, in some cases I may raise the ramp angle if the braking zones aren’t aggressive and there is a lot of high speed turns.

On a side note on plates - my observation is that adding a plate increases the locking factor by about the equivalent of 1.5 to 2 clicks lower in ramp angle meaning 70 with 3 plates is approximately the same as 60-65 with two plates. With preload and plates I don't feel there is much difference in preload when adjusting plates and therefore don't adjust preload if changing amount of plates. Another interesting observation with number of plates is that the less plates you have the diff reacts more quickly and aggressively, with more plates the diff reacts a little slower and smoother.

Preload - preload is a tricky one to understand what is best. Obviously higher preload means more locking when the car is neutral (i.e. not accelerating nor decelerating). So the natural thought is that a higher preload is bad as it resists turning mid corner right? So you should have it as low as possible? There is much discussion about this even on real life forums when it comes to building diffs, most diff builders will say as little as possible is better. However there is evidence that some top level teams use high preload in there diff builds and hence there is an alternative view that a high preload diff is better. At the end of the day it will come down to driver style and personal preference.

My preference is for higher preload and as such I start building sets with a preload of 4,800 kg and often use much higher than that. Why would high preload work you ask, I think it works well for me as the premise for building GT sets is that I do most of my braking in a straight line and as such the trajectories through a corner may not be as smooth as someone who trail brakes more or coasts into a corner more. Hence the handling characteristics that I am looking for in the GT are: strong stability in the braking zone, a tight car that can be thrown into a corner aggressively (understeery in neutral phase) and a car that can be steered with the throttle.

With that in mind my starting points for the diff in the GT are Drive/Coast/Plates/Preload 70/30/3/4800 if the car is too loose mid corner I will raise the preload if it is too tight I will reduce the preload. This judgement is taken when in the scary mode when the tyre noise is telling you that they are scrubbing, if you’re not judging it in that mode then you will be adjusting it incorrectly.

So the above on preload should not replace what Martin has written but it is an alternative view of how to set up the diff.

One thing to note is that obviously preload has an interplay with coast and drive locking and as such my starting points with building a diff would likely be different with lower preload as a starting point. In other words if you take a diff setup for a given track from Martin and whack high preload on it will likely not work nicely. If I change preload by a large margin I normally also adjust drive and coast as well.

Jeez I didn't intend to turn this into an essay...   I hope someone gets some use from it and it hasn’t put you to sleep!
This message was edited 1 time(s). Last update was at 4/1/2012 4:42 p.m.

Though infact it does say that coast setting is opposite to the power setting!

In iRacing, power side is lower numbers for less locking (same for clutches and preload), but coast side is higher numbers for more locking.