63 replies to this topic

[jack be nimble]

What does the factory allow? 1% mismatch? That's 1 turn out in 100 tailshaft revolutions

[doommachine]

That's a good suggestion, but tbh removing the transfer is easier than pushing the car 50m, I live on a main road

 

I think that if I lift two wheels off the same side of the car off the ground and rotate the wheels I can also display the mismatch.

 

If the rear wheel turns at a lower rate than the front I can say, okay, the VCU is slipping and this is normal.

 

BUT if I turn the rear wheel and it makes the front wheel turns at a higher rate, well, you can't add more energy into the system now can you.

 

If the transfer was 1.074 ratio you would expect the front wheels to turn at a higher rate as the effective ratio of the rear diff would be somewhere around 3.6. I drew this all on a piece of paper at work but now its midnight I'm not sure, but this sounds reasonable right?

 

Edit: Theres about 1.5% difference between the two transfer ratios so I'd expect 10.1 turns from the front wheel if I turn the rear wheel 10 times. I'll try it out after work.

Edited by doommachine, 27 August 2017 - 02:41 AM.

[jack be nimble]

But you are involving the spider gears too. If they are the same gears front and rear, I'd say its OK. But if not...I don't know.
I think you are correct.

Try your way and let us know.

You could prep it all, and take it somewhere quieter

[doommachine]

Hmm I get that the VCU and the spider gears in the centre diff will cause some loss between front and rear but it should be impossible to make the front wheel gain even a fraction of a turn relative to the rear. With the correct ratio transfer the rear ratio is about 0.02% higher.

 

I'll see how it goes. If I gain nothing I'll pull the transfer and inspect. The grey sealant doesn't necessarily mean its in great condition.

 

I can pretty much rule out the rear diff as being broken because I had it apart when I pulled the AWS/ installed evo LSD. Ring gear and pinion both look undamaged/unworn, the oil that came out of it had no metal particles. Rear backlash is reasonable although even if it was sloppy it surely wouldn't stress the centre diff - just itself, right?

 

But it must be something attached to the centre diff causing more stress. Transfer wrong, rear diff ratio wrong. I can't see that a wrong rear diff ratio would not have a centre diff exploding for over 6000km, even a 4 spider...

 

EDIT:

Hey so.. why don't I just get all 4 wheels off the ground, put a chalk mark on each tyre, pull the plugs out and rotate the engine? If there's no resistance all the wheels should rotate equal amounts, right?

 

Also I've been googling americans swapping manuals into auto 3kgts and not swapping the diffs and driving for miles without blowing the centre diff. So, guess I can't rule out the rear diff being the wrong ratio.

Edited by doommachine, 27 August 2017 - 05:41 AM.

[Cranked]

I have a welded centre so I don't really know

[doommachine]

Well, I think I've confirmed the issue. Rear diff, while it has a 3.545 cover, is 3.9 ratio.

 

One turn of the wheel at the back equates to a faster rotation at the front.

It takes 10 full rotations to get the mark on the front wheel to get to its original position - therefore the rear ratio is 10% higher than the front (the front wheel has travelled 11 rotations for every 10 of the back).

 

Amazing to think. Entaran blew a centre diff, spent something in the realm of $5k to get it rebuilt and the shop who installed it didn't check to see if the reason it was binding was different front and rear ratios.

 

To the for sale section with me.

[bazeng]

Good job finding the issue ..

[doommachine]

Yeah I dont feel that good about it to be honest Baz, the diff cover says 3.5. Why would you swap the covers???!!

 

I guess the moral of the story is always check everything.

 

Just to confirm - 11 turns of the wheel results in just under 22 rotations of the tail shaft - my understanding of diffs means that the ratio is thus 22/5.5, or about 4.

[bazeng]

I would instead assume that the car once possibly had a 3.9 box so the stock diff was used with a matching crown wheel / pinion.

I'm assuming it is a 3 bolt rear?

If it was a 4 bolt rear it may even be the inners of the diff swapped out to suit a 3.9..
I guess who will ever know except for Laurence himself... Should let him know! It may give him closure..

[doommachine]

Yes its a 3 bolt housing, but I recently swapped over to an evo LSD / 4 bolt axles.

 

I shot him an email, good idea.

[VR42LT]

Divide the crown wheel teeth by the pinion teeth to get a 100% confirmation on ratio.

I have posted on your wanted add.

[doommachine]

I might possibly be  alright, as always, 265t to the rescue.

[BYBY5L]

Good work finding the culprit! Booo to the shop who didnt pick that up though.

[doommachine]

Hmm...

If I run 205/50/16 tyres on the front and 225/60/16 on the back.. I mean short of the annoyance of carrying a spare this pretty much solves the problem right?

[jack be nimble]

Everyone will tell you that's not allowed. But I think if you have done the maths properly to get your rolling diameters to equal the ratio mismatch, I can't see how it won't work. That's what I was going to do with my ute. 3.5 rear ratio, with few % bigger tyre to make it match. I was going to go as far as adjusting rear tyre pressure to get the match closer still.

Do it! Staggered wheels look better imo.

[doommachine]

It's pretty tempting haha...

[bazeng]

There is a GTO that does this in America. I've done this in my Stagea.

As long as rolling diameter is the same you will be OK.

[jack be nimble]

He is saying to make the rolling diameter different to make up for the different ratios. 3.45 - 3.9 is around 12%, so run 12% bigger tyres on the rear.

[doommachine]

Just had my rebuilt LSD returned from Neat gearboxes.. have yet to install but I am deeply impressed by the turnaround time, as well as being told the backlash specs and breakaway torque of the centre after the rebuild.

 

I have to replace the wheel studs and I'm going to replace the front ball joints and tie rod ends before I get it all back together, but when eventually I get that done I'll let you all know how noise-free it now is

 

Thank you to Paul for finding a 3.5 crownwheel and pinion for me.

Edited by doommachine, 26 September 2017 - 09:20 AM.

[doommachine]

Changed my mind since theres a delay on the tie rod ends and just got it all back together. It drives amazingly well now, the car drives noticeably smoother now - likely since the wheels aren't pulling at two different speeds

 

Can't exactly talk about it's skid performance yet since I'm running wastegate pressure while I tune my closed loop MAF comp.

 

Oh, another thing which came to mind regarding having the ratios different. I doubt it will come up for anyone in the future, but something to consider. Theres an experimental road surface about 100 km out of the city on the way to Keith, which when I hit at 110 caused the front end to weave about uncontrollably. My assumption was that the coilovers were so hard they were causing the front wheels to bounce off the surface.

 

Entaran noted that the 225's he had were "like ice" in the wet. But, in hindsight, it was likely the front wheels trying to catch up with the rear. Scary things.... if you're buying a modified AWD check the ratios!

Edited by doommachine, 09 October 2017 - 01:50 AM.