3.8 differential shim issue

If up to me I’d set them up with just a slight amount of pre load - when you’ve got it assembled with some oil on the bearing surface, but without the oil seal and with the nut tightened to what you intend to use you feel a slight amount of drag as you turn the shaft inside the hub. I’m not talking about a bit of stickiness when you first try to turn it, but rather the slight drag you feel as you continuously turn it.

1 Like

Hello LLynn,
There is no reason for not setting the Output Bearing up using the pre-load specs of the XJS; effectively they are of the same arrangement. The XJS uses a crushable spacer, but that innovation is only to circumvent the need to determine the correct shims to use, thus speeding up the assembly process in production. The pre-load for the XJS is 25/36 in/lb torque, measured as the amount of torque required to rotate the the Output Shaft/Bearing assembly in the housing with the oil seal in place.

When using the crushable spacer, the correct torque can be achieved in the one assembly process; accordingly, the assembly and pre-load measurement is made with the oil seal in place. When using a spacer and shims, the correct pre-load is achieved through a process of trial and error, necessitating the reassembly of at least two times. Once with an excess of shims to give measurable End Float (from that you can determine the correct shim stack thickness to give Zero End Float - 0.001 to 0.002 for pre-load), then a second time with the supposed correct shims. If when checking the pre-load its found to be incorrect, shims will have to be added, or subtracted as required to achieve the correct torque to rotate the Output Shaft. Don’t assemble the oil seal until the final assembly. Accordingly, when setting and testing the torque required to rotate the Output Shaft, allow approximately 5in/lb torque for the absence of the seal and use the bottom end of the torque setting tolerance.

Regards,

Bill

2 Likes

thanks for the reply. I know you would have set up a lot of these

My apology for not understanding

presumably you mean it is ok to follow the Service Manual Specs ?

I aimed for as close to .001" endfloat

  • cause the FSM says to have endfloat
  • too much pre-load definitely damages them

this was fairly easy using honed dummy bearings Timken LM 48548,
$25 at the bearing shop but show as $11 on Amazon!

DSCN0591-crop

following the early shop manual specs will lead to a shortened life of the bearings, shorter seal life and a low brake pedal because the disc will wobble and push the pads back. Jaguar figured it out and corrected the specs on the later diffs with the same design taper roller bearings. If you read the manual correctly, it calls for end float on the front also. With end float, the bearing is not consistent with the surface of the race it rides on. Been setting them up for preload for over 40 years and all works well. If you really want to be a stickler for following the shop manual exactly, the late 70’s XJ12 calls for toe out on the front. This will eat tires in quick order. Later manuals corrected it. The suspension did not change, only the manual. Sometimes it is best to go with the later specs as Jaguar learned what worked best.

2 Likes

So you are saying to set both the front and rear bearings up with preload on Lynn’s 64 E-Type,. How much preload and how is it measured or determined?

–Drew

Indeed, on later years the manual called for preload.
And I’ve set all of them with a slight preload, output shafts, rear and front hubs.

Hello Aristides,
Rear yes. Standard Jaguar front wheel bearing set up, Definitely Not. The fit of both bearings on the Front Stub Axle is a slight clearance fit (sliding). Even with the Bearings set up with End Float, just about every Front Hub you will remove from a car that’s done a few miles, will reveal a wear band the width of the inner bearing race at 6 o’clock on the Stub Axle. This is caused by the inner race of the bearing spinning on the axle.

Unless some provision is made to prevent the bearings from spinning, such as a spacer between the Inner and Outer Bearing, or a Stub Axle with bearing journals that are an interference fit with the bearing’s inner race, preloading the standard bearing/stub axle arrangement will ensure that the bearings will spin on the shaft.

I provide both solutions. I manufacture stub axles with bearing journals that are an interference fit with the bearing. Mechanics don’t like this system, as it leaves the Inner Bearing and Seal on the Stub Axle when the Hub is removed, requiring another operation to pull the bearing off of the axle and when replacing the bearings on the shaft, they have to be pressed on. Following is a picture of an S3 Stub Axle I manufacture.

I also manufacture a spacer, the bore of which follows the profile of the stub axle, but with plenty of clearance on the taper, that sits between the Inner and Outer Bearing, just like the rear set up. Pre-load is achieved with shims and the Stub Axle nut can be torqued up tight to lock the bearing on the Stub Axle.

Regards,

Bill

1 Like

Hello Tony,
So, the reference you find from Timken (manufacturer of Taper Roller Bearings) and others recommend Taper Roller Bearing be set up with Pre-load. Later Jaguars, with Taper Roller Bearing on the Output Shafts have moved to a Pre-load set up, but still you would go with End Float.

For many years I serviced CNC machine tools. The spindles of these machines in a typical production environment, rev way quicker and for much longer periods than the overwhelming majority of E Types will see in their lifetime. All of the spindles have a bearing arrangement in Pre-load. Most have Angular Contact Ball Bearing races, supplied as a matched set, with built in Pre-load, arranged Back to Back, or Front to Front. Some have Taper Roller Bearings with the Pre-load set with a spacer between opposing bearings, in the same way the Taper Roller bearings on the Jaguar Output shaft are arranged. And these machines just go and go and go.

As a beside, its rather surprising to some, the small amount of lubricant used with the Angular Contact, Ball Bearing Races in these precision machines. Picture, if you will, the front, or rear bearing of the Jaguar, four speed gearbox. Now picture the amount of air space between the inner and outer race, not displaced by the Balls in the Bearing; not that much you may think. The correct amount of grease is only 10% of the available air space.

Regards,

Bill

1 Like

FWIW, the front wheel bearings of my 15-ton bus are set to 75 ft-lbs: they work quite for over 200,000 miles.

On Formulas Vee, the ball bearing front wheels run pre-loaded, dipped in oil (to reduce drag), about 10 ft-lbs. Never saw a failure.

Pre-load is not the bug bear it was once thought of.

Hi Bill,

thanks to the clarification offered by yourself and Dick M,
(and serious reservations on my behalf anyway),
I will now regard that section of the FSM to be incorrect, and go for pre-load

Hello Paul,
In my opinion, its never was.

What type of bearings and how are they arranged?

Regards,

Bill

1 Like

On the bus? Just standard taper roller bearings… just HUGE!

Hello Tony,
Its not really incorrect, just dated. Set ups and materials used in various Jaguar parts has evolved over time. For example, the End Float specified by Jaguar for S1 and S2 Rear Hub bearings is 0.002" to 0.006" (0.004" mean). That has been changed for the S3 to 0.001" to 0.003" (0.002" mean); a halving of the tolerance and the mean setting.

This is another bearing set up I normally set in Pre-load; but that’s another story.

Regards,

Bill

1 Like

Hello Paul,
Is the 75lb/ft torque on the nut? Surely there must be a spacer between the bearing to soak up some of that torque?

Regards,

Bill

Indeed there is: the bus wrench told me the bearings themselves are, when torqued down, noticeably “sticky” to turn (no clue as to what breakaway torque is, but I can find out).

I was fairly astonished when I saw him setting the preload, though!

I’ve always set wheel bearings at zero lash, preload.

Yes, you are absolutely right, and indeed my stub axles had grooves from the bearings spinning…
I assembled them with Locktite, special for bearing assemblies, to prevent this from happening again.
I believe this set-up will prolong bearing life. Time will tell…

Aristides

Drew, looks like others have answered your question as it was late last night when I signed off. I will add another observation. On the XJ6 and XJS, loose front wheel bearings (ie: end float) will lead to the inner bearing rocking slightly under load when the car is driven. This wears the bottom side of the spindle at the inner bearing position to the point that you cannot adjust out the free play in the front wheel bearings. This causes shake when braking as the disc is not held in a consistent rotation plane. If you check your rear wheel by grabbing it at 3 & 9 positions and moving it. If it moves, you have free play or end float. This really causes the car to just not drive right. Nothing like an E-Type that is set up properly and tight. One set up with built in slop will not feel as nice when driving.

2 Likes

Oil may actually work better. Sometimes the grease just gets pushed to the side and stays there while the rollers run dry.

What great, short and well thought out explanation of the crush collar process. Don’t recall ever seeing such a simple and coherent explanation. Can’t comment on the absolute truth - but it sure makes sense to someone who has done both systems on cars. Bravo.

1 Like

One issue which will apply in only a few cases relative to the manual. I swapped all sorts of stuff around on our car which included bolt on hubs relative to this discussion. The “newer” style requires a different order for the seal, as it won’t clear the set up as indicated in the “old” shop manual procedure. Ask me how I know… Had to take the thing apart after doing everything to put the seal on. The old leather one apparently can be put on at a later stage with knock off, older style parts.