Loose Diff - setup for motorsport
 

I've heard it mentioned a few times that you can set up a 'loose diff' for motorsport. I would assume this means setting it for less friction but sacrificing life on the bearings / wear on the ring and pinion. Googling nets me nothing so I'm just posting to see if any of you know what this 'loose' setting is referring too.

When you preload the pinion, it's set to something like 12 in lbs of resistance which I assume increases with speed. Halving this might gain you some extra torques at the risk of the pinion wandering a bit on the bearing. I would assume you wouldn't muck with the preload on the carrier.

I would guess this might net you 5NM of torque at high RPM... maybe less.

alternatively could it be in the backlash settings running a larger gap, but I would assume this would destroy diffs fairly quickly.

Or am I on the wrong path entirely and this is not a real thing?

Edit: This was definitely not in relation to clutchpack style diffs and their preloads. This was specifically regarding the ring and pinion setup.

I would think the pinion expelling the ring gear at random would be a risk way to large to take for whatever you think you'll find.

Plus I drove on a loose as hell 1.6 VLSD, the slack sucks hard coming in and out of overrun, I wouldn't want to increase that, not on a car I want predictable in corners.

On a drag car, yeah maybe. Those idiots ice their oil and crap though, more money than brains in some cases.

As long as the ring and carrier are held in place via the stock preload (ie very tight), and it's just the pinion itself that's a little looser, I wouldn't have thought a little movement would be that devastating. It is a smaller gear and I only do 1000km / year in all my events combined, plus my car is only hitting moderate power (200HP). I think the looseness you're referring to is the backlash setting, and yes, I don't want that sloppy feeling on and off the throttle.

I don't think it's worth attempting with unless there is decent end result, the reason for this post.

I'd think for every extra bit of preload lost, the pinion will be able to climb the ring gear more in acceleration. That there's what'd eject the ring, no matter how it's held down.

I suspect it's why so many VLSD's/1.6's die, if you are aware of it and keep it loaded smoothly they can last long enough.

Maybe I'm mistaken, but I think setting backlash is a precise mix of pinion preload and ring gear shimmage, am I wrong? Isn't there a crush sleeve that basically positions and preloads the pinion bearings?

My dense ass just realized you don't really care, I'll back out now!

I appreciate the discussion... seriously. I don't know that much about diffs and what destroys them but I get what you mean regarding the climbing the ring gear.

Over the past few days I've been reading up on how to set the backlash as my diff is currently in bits whilst I switch ring and pinions. There are precise ranges for all the preloads, backlash etc that need to be within spec. The crush sleeve sets the preload on the pinion bearing. By tighten up nylock nut on the pinion you're crushing the sleeve until you reach the required bearing resistance. I've read on the RX7 forums that the crush sleeve can collapse with multiple hard launches and basically does what you're saying and lets the pinion move and fail. There is speculation that this is the major failure point of the diff. To prevent this, there is a solid shimmed sleeve you can buy / machine and that is what you'd need to do to run a slightly looser pinion without it getting looser over time.

In drag racing they set them up on the tight side. Keeps the gears from climbing around as much but increases bearing wear/shortens life of diff (might only last 100k miles instead of 250k miles) for racing, don't matter nobody puts 100k miles on a race car before an overhaul.

When I think of a loose diff I think of clutch pack lock up characteristics, maybe that's just me though.

Last night I set up the new ring and pinion. Setting 12 inch pounds is actually very hard. It's so little torque (1ft lbs) that it's hard to measure on any equipment I have. I ended up using a fish weight scale on a 1ft long bar to measure the friction of around 1 lb. This isn't a lot so halving it would make no sense. For anyone interested, compressing that crush sleeve is requires a lot of force.

Other than that, I tightened up the carrier bearings to the preload and set the backlash by hand based on how much wiggle it had before I pulled it apart. I then used some bearing blue and checked the tooth load point and it was perfectly placed on the face. I'm going to check with a dial gauge but it looks like I fluked it first go.