Thanks for the links. I will check them out tomorrow, as I continue to think through everything, especially after my experimentation today.
From my efforts today, I realized that moving the side shims changes back lash a lot. My issue is, I can't get the backlash in spec just by moving the side shim(s) as the manual instructs.
As assembled at the factory, this differential has one left side shim and two right side shims. To reduce backlash, you move shims from the left to the right side. Per the manual, you are not to change the overall thickness of the side shims. Mine are: left = .3mm (0.0118") and right = one .4mm (0.0157") and one .5mm (0.0197").
With everything assembled with factory shims in all locations (including the .126" pinion shim), and pre-load on the pinion bearings set to specification, I had about .020" of back lash. Moving my one left side shim to the right reduced back lash to about .014-.015". With no other shims to shift to the right side, I first became aware of my problem - replacing the pinion bearings changed the pinion to crown relationship from the factory setup. I went ahead with getting some pics of the wipe pattern on the coast and drive sides of the crown gear:
Contact appears to me to show heavy on toe (inboard edge of tooth) on coast (first pic), and heavy on heel (outboard edge of tooth) on drive (second pic).
Today, I put three shims (supposedly .001" each) in with the .126" pinion adjusting shim. In theory, that moved the pinion .003" closer to the crown gear. I moved the shim that was originally on the left side, back to the left side. I measured backlash at about .018" to .019. I went ahead with getting another set of pics of the wipe pattern on the coast and drive sides of the crown gear:
It looks to me like the contact on the coast side is close to the "correct" position. However, on the drive side, it looks like the contact is still too much on the "heel". After posting those pics earlier today, I went back out to the garage, and moved the left side shim to the right, and measured backlash. It was about .010" to .011". That was very interesting - it seemed like I had improved things overall by bringing the pinion gear about .003" closer to the crown gear.
As a last iteration today, I cut some more shims. I took the pinion out again, and stacked a total of eight .001" shims together with the .126" factory shim. My thought was that perhaps moving the pinion .008" closer instead of .003", I'd be able to keep the one shim on the left.
This time, when I got everything back together, when I torqued the pinion nut to the same 124-ish ft lbs, the amount of effort required to rotate the pinion was much higher than the times before. I backed the nut off again and retorqued. Same thing. Somehow, the pinion turning torque has become excessive - it is difficult to turn by hand. I am not sure what happened but I was using the same torque wrench that I had issues with on the front strut nut. Anyway, I will attempt to sort that out tomorrow. I see the torque spec in the differential manual that was uploaded to the resource section today is 101-123 ft lbs., while my factory service manual shows 122.9-144.6 ft lbs. Dang it.
Anyway, I measured backlash, again with the side shims back in their factory locations. This time, I got about .014"-.015". So, with the side shims in their factory locations, I appear to have progressed from about .20" with the .126" pinion shim, to .018"-.019" with the three .001" shims added, to .014-.015" with the eight .001" shims added. And when moving the left shim to the right in combination also, I measured backlash at about .014-.015" with the .126" pinion shim, to about 010"-.011" with the three .001" shims added, and about .007"-.008" with the eight .001" shims added.
It took most of the day for me to work on this. In conclusion, stacking a bunch of home made .001" shims in addition the existing factory shim which is between the rear pinion gear and the rear pinion gear bearing, does not change backlash a lot. But it does change it. By comparison, moving side flange shims has a larger effect on backlash.
Either tomorrow, or some day next week, I'll pick back up on this again. It helped to write up all my observations, to get what I found out today on paper... so to speak. I need to correct the bearing preload. I hope I didn't damage a bearing. I wouldn't think that is likely though. I will back the torque setting down to about 100 and see if that puts the preload back in spec as tested by checking turning torque with the Quaife unit out of the case.
If I can get that in spec again, I'll check the tooth contact pattern and see how it looks. I am somewhat excited that I was able to get backlash within spec. I think this is decent progress.