Diff rebuild issues

[Warb]

Having stripped the rear diff, I carefully mic'ed the various shims and started reassembly. I made up the relevant bearing pushers, together with a dogbone to measure the pinion height as described in the workshop manual.

With new (Timken) bearings, and exactly the same shim thicknesses as were removed:

Problem 1: When the drive flange nut is tightened, the pinion starts to bind. To get the correct preload on the bearings (measured on a torque wrench) requires 40 thou more shims than were originally installed (60 thou vs 20 thou)

Problem 2: With the original thickness of shims added, the pinion head with new bearings is 20 thou too high - it would require 20 thou of shims to be removed to achieve the correct (book) height.

All bearing are seated correctly, and square, and all shims have been measured very accurately to replicate the original thicknesses.

Given that the pinion is 20 thou too high, and the preload is 40 thou "low", my suspicion would be that both bearings are about 20 thou "thicker" than the originals. Is this likely?

Is the best approach now to leave the pinion "high" and hope (of course I didn't measure the height before I took it apart...), or to correct the shims to give the "correct" height?

[PhilipA]

The diff pinion height should not change if you are using the same pinion height shims and the same pinion .

I would tend to leave the pinion height the same as it was, if you are reusing the same pinion gear or the diff may become very noisy.

The bearing preload shims are different to the pinion height shims as you no doubt know. In this case I would just set up the bearings to the correct preload ., but you had better be very very sure that you have the inner and outer races both seated properly.

IMHO these are 2 different issues, that should be treated separately. I have seen so called exact replacemnet bearings being significantly different to the original ones.
I will give a caveat here that I have only reassembled 2 diffs, and one (Porsche 924 turbo) I had end up having reset by and expert as the bearing preload depended on the cases matching and the bearings were different , and more recently I rebuilt a Rover diff successfully.
Rovercare is the expert on this so perhaps PM him.
Regards Philip A

[Warb]

But what was the original height? If the bearings are indeed different then using the same thickness of shims does not achieve the same pinion height, it only achieves the same "base of the bearing shell in relation to the case" height, if you see what I mean.

I believe the two issues are directly related. The preload shims add space between the two inner races by extending the length of material between them. The intention of the pinion height shims is to adjust the height of the pinion, but as a side effect they increase the distance between the shells. If the races are kept at the same spacing and the shells moved apart, the preload will increase.

In this case, the height of the pinion has increased (assuming it was right to start with!). That suggests the bearing is "thicker", which moves the pinion further out, and therefore increases the preload. Additional "preload" shims are then required to further separate the two races and reduce the preload to the correct level....

In the morning I'll press the pinion head bearing back out, re-measure the shims and maybe try to measure the bearings as well.

[chazza]

Problem 1: When the drive flange nut is tightened, the pinion starts to bind. To get the correct preload on the bearings (measured on a torque wrench) requires 40 thou more shims than were originally installed (60 thou vs 20 thou)

Problem 2: With the original thickness of shims added, the pinion head with new bearings is 20 thou too high - it would require 20 thou of shims to be removed to achieve the correct (book) height.

Is the best approach now to leave the pinion "high" and hope (of course I didn't measure the height before I took it apart...), or to correct the shims to give the "correct" height?

Is it a Rover diff, or a Salisbury?

If you suspect the new bearings, take them out and measure them - you are probably correct in suspecting them.

Don't leave the pinion high, unless you have blued the teeth and checked that they are bearing on each other in the correct place.

When you say that you have checked the pre-load with a torque wrench, I can't see how that would work if it is a Rover diff; the recommended way is to use a spring-balance, a pulley wheel and some twine, so that the higher start-up load can be ignored and the running-load measured on the scales,

Cheers Charlie

[Warb]

It's a Rover diff.

I have a dial type torque wrench (0-30in.lbs). It gives a direct reading of the torque, so you just turn it at a constant rate and it tells you exactly what torque is being applied at any instant. It also has a "maximum" needle to indicate the maximum torque applied, but in this case (as you say) that just indicates the start-up torque.

A spring balance works, but requires a lever to be bolted to the flange so that the string can be fixed exactly 12 inches from the centre of the pinion, otherwise the reading on the spring balance is meaningless or at the very least requires some calculation to correct for the length of leverage.

And, of course, I don't own a spring balance that I would regard as accurate, but I do have a "certified to within 4%" torque wrench.....

[PhilipA]

Well I have been trying to recall exactly what goes where , and yes the pinion bearing depth will have an effect on pinion height.

I guess I would be first checking that both the shims and bearing outer to shim are very clean with no burrs.
I would be tempted to use the original height setting shims , and then just adjust the preload shims.
You cannot expect a brand new bearing to be the same overall height as an old one, as they would get shorter with wear as they are tapered rollers. So the pinion would sink over time by the wear in the roller bearings, and the preload also be lost.

Regards Philip A

[Warb]

You cannot expect a brand new bearing to be the same overall height as an old one, as they would get shorter with wear as they are tapered rollers. So the pinion would sink over time by the wear in the roller bearings, and the preload also be lost.

Very true, but with the new bearing and exactly the same thickness of shims as originally fitted, the pinion is higher than it should be - that's higher than the "new assembly" spec., not just "higher than the old one"! And it's significantly higher - 20 thou out of spec, when the spec is +/- 1 thou.

It seems to me that either the bearings are wrong, or there is a burr somewhere that has prevented them from seating (though inspecting them in position doesn't reveal anything in the realms of 20 thou!), or perhaps the diff had been played with in the past and the set-up was utterly wrong. It wouldn't be the first thing on this truck that had been badly botched!

I'll take it all apart and check for problems, measure the bearings and then make a decision. If I can't find anything wrong I might install the crownwheel and check how everything lines up!

[Blknight.aus]

Blue the diff before you strp it down. If you dont have bearing blue you can apply whiteboard marker.

[bulletproof]

The pinion height is quite critical and should be according to the workshop manual .

The mathematics of it is that lines drawn at the angle of the pinion teeth with the pinion set at the right height should intercept each other at the centre line of the crown wheel regardless of any different bearings and shims used.

Because you have a worn diff the teeth could have worn into each other in a setup that may not have been quite correct so you will need to check teeth markings with Blue to make sure they are meshing half way up the teeth and adjust with small shims until it is correct.

As long as the bearing are definitely pressed right home even if the bearings are slightly different they will still work as long as the pinion height remains correct.

Cheers Richard

[Warb]

OK, it looked rubbish so I stripped it again. The new pinion head bearing, a supposedly identical Timken unit, is about 20thou deeper between "operating surfaces" (not the cage!) than the old one. I do not believe that there can have been 20thou of wear on the bearings, so I can only think that the old and new bearings are simply a different size either by change of design or by tolerances (although 20thou seems a large "tolerance" difference!).

So, for anybody who is contemplating doing this job (and it's not hard) be aware that it is NOT POSSIBLE to simply replace the bearings and assume that everything will be good and the old shims will give the right pinion height. At the very least the relative thicknesses of the bearings should be checked!