Is there something wrong with you? Read the text, particularly the text I’ve highlighted with the Red Rectangles. Points 85, 86 and 87 relate to the illustration to the right. Note the number 85 in the illustration; that’s a reference to point 85 in the text to the left. The point I brought attention to is 88; none of the others.
The illustration just happened to be on the same page that I scanned, along with the text for other operations involved with the assembly of the differential. I didn’t expect that the information supplied had to be so focused that everything on the page that didn’t relate to the subject had to be filtered out; I just supposed that the members of this Forum had the intelligence to read and comprehend what’s presented. For your sake I’ll be more careful in the future.
Point 88 and the Caution above it all relates to the Pinion Bearings. Take the time to read and understand a Post before putting your foot in your mouth.
We’re not talking about lots of items, only differential pinion bearings. Taper Roller Bearings rather operate in pre-load than end float. As I stated in an earlier Post, but I suspect that you either didn’t read it, or don’t understand, the pre-load of the pinion bearings diminishes with acceleration (load on Pinion and Crown Wheel). If the pre-load is light on, a shift in position by the Pinion relative to the Crown Wheel occurs and as a result, a different tooth contact and noise. Accordingly, the pre-load that you should be aiming for is the correct pre-load. Making it less is NOT being safe.
“The original bearing has not been available for many years and the usual After Market Jag parts suppliers, supply a replacement bearing that is the same OD and ID, but 2mm narrower. Two, 2mm wide spacers are supplied with this bearing, one for the outer and one for the inner race. This bearing doesn’t have a split inner race, has no pre-load and there is no scope to pre-load the bearing. Accordingly, its a poor substitute, but better than nothing.”
The later XJ6/XJS side axles will direct bolt in to replace the Series 2 & 3 style axles. There is a large o-ring in the case that you need to remove and you do not put the wide assortment of shims back. I also recommend a light polish /cleaning of the cutout in the case where the new style o-ring seals the axle housing to the diff case. The later axles do use taper rollers and a crush sleeve and operate under a preload.
Bill Angel
Referring to the out put shaft bearings you pictured, after I replaced the seals in this assembly I tightened the nut but was able to tightened it about a 1/2 turn beyond the locating marks I had made on the shaft and nut. Is this ok or should I dissemble and adjust to the original marks? The bearings seemed to be ok.
I apologize almighty one. I was confused by the illustrations being the output shaft, and the text initially speaks of the output shaft. No need to be a dick about it, just wanted to be sure no one followed a wrong procedure.
Hello Dick,
S2 XJ6 and later Output Shafts assemblies can also be used, but these parts are becoming hard to get as well; at least here in Australia they are.
There is a “metal on” condition when comparing the Output Shaft from a Type 3 differential (double row, angular contact bearing type) to the later style that uses Taper Roller bearings. Accordingly, I re-machine the earlier style to convert them to later variety and manufacture the bearing carrier anew.
Hello Thomas,
If you’re referring to the bearings in my Post #37 above, its unusual for the nut to tighten a further half turn, unless the nut was not correctly torqued up previously. Even if the balls and races of the bearing are worn, the inner faces of the inner race will not have changed, nor the overall width of the bearing.
Because of the design of the original bearing, with the split inner race and with the pre-load included in the dimensions of the bearing, you can’t actually over-tighten the nut in terms of affecting the pre-load.
Its difficult to use a conventional torque wrench that uses sockets, due to the reach required by the socket to engage with the nut. However, if you wanted to check and set the torque to that specified (90 to 110lb/ft), then you can use whatever spanner that you have to fit, arranged so that its long axis is parallel to the floor and hang a weight a set distance out from the centre of the shaft.
Well your comments would have had the opposite affect. I see this often in a CNC machining Forum I give advice on. Some make incorrect replies in such an authoritative manner, that the newbie, or uninformed, take it for fact; then argue black is white for what seems, just for the hell of it, or they’re too stupid to tell the difference.
From all the back and forth Ive seen about replacing the Pinion seal I’m concerned that it might be too much risk to attempt it?
If I would proceed, I would plan to punch-mark the nut and the pinion, remove and replace the seal and the nut and retorque the nut to the original punch-marks. Provided there is no end-play in any direction with the pinion, I’d probably stop there and call it a day.
Does this sound like a plan which can deliver success or am I missing something?
If so what additional steps can I reasonable take?
I need input from those much more knowledgeable than I on this subject.
If I were to use a dial inch lb torque meter, take me thru what I need to do. Must I remove the brake system and half shafts?
What value am I attempting to achieve as I tighten the pinion nut?
check the existing preload required to turn the entire assembly with a dial wrench
remove nut, yoke, then seal
replace seal, and (preferably Nut)
ideally, in my opinion, you should be able to retighten the nut till it is close to the preload that existed before, and near to the mark…exactly is best
to top it off, the Pinion nut itself, when the yoke is held by a big wrench or fixture, should also be around the specified torque for that application (as discussed in here, this figure is specified in the manuals, but tighter is better than looser (imo) 140ftlbs +
You should avoid exceeding the pre-existing pre-load with used bearings
that figure is the most important
( I use) blue loctite on Pinion Nuts, they have to be very tight, if you cant get things to correspond, something is wrong…but there isnt any reason that should happen.
You are only replacing the seal, (it might add a tiny bit more drag)
there is no absolute guarentee some shop wont stuff it up for ya
So it seems we are up to 51 posts on this thread. I hope the OP has his answer. But, IMO, this has raised some interesting topics and theories. I believe many, and Bill in particular, have provided a lot of good information. Now based on this information, I would like to possibly draw us to a conclusion by asking this question:
Why do some bearing sets like wheel bearings, get their preload adjusted without a shimmed spacer or a crush sleeve, and some of course do. In other words, what really does the spacer do? And here I may be wrong, but it seems crush sleeves are more common than shim packs today, then why? Modern setups seem to have superior life, are the crush sleeves part of that?
Tom
Hello Tom,
It’s a fact that Taper Roller Bearings prefer to run with pre-load than end float. But its specified by Jaguar that the wheel bearings, front and back, be set with end float.
With the front wheel bearings its for expedience in assembly and after sales service. If you have examined an E Type front stub axle, when wheel bearings needed to be replace, you will have invariably found a shallow groove the width of the inner bear from circa 4 to 7 o’clock where the bearing is located on the shaft. This is caused by the inner race of the bearing spinning on the shaft; and this is with the bearings set with End Float. If you were to set these bearings with pre-load, the bearing inner race spinning would be guaranteed, exacerbating wear of the stub axle.
If you place a spacer between the two front bearings and lets say that the setup is shimmed to give an End Float of 0.002" (0.05mm) when the nut is firmed up, then the nut can be tightened as much as you like (within reason) and the End Float will remain at 0.002". The bonus though, is that both the inner and outer bearings are locked tight between the nut and the face of grease seal disk and hence the face of the Stub Axle Carrier, preventing the bearing’s races spinning. In this arrangement, you can set the bearings with pre-load without fear of their inner races spinning.
Now the naysayers will say this is taboo. However, CNC Machine Tool Spindles bearing are setup with quite heavy pre-load, with the majority having quite a minimal amount of grease applied for life. If you can picture the amount of air space left after the inclusion of the ball bearings in a typical ball bearing race, the correct amount of grease for the precision bearings used, is only 10% of the available air space within the confines of the bearing race. And these spindles do a shite load more work in a few weeks (in terms of bearing revolution) than an E Type in its 60 year life.
With the Pinion Bearings of the differential, the inner race is firm interference fit, but the fit of the outer bearing ranges from a transition fit to a close running fit. This arrangement without a spacer between the bearings would result in at least the outer bearing’s inner race spinning on its journal and a fair chance for the inner bearing to do likewise.
Insufficient end thrust exerted by the Companion Flange Nut, is the reason one often sees evidence of Pinion bearings having spun on their journals.
Let me start with this- why? Do they prefer preload for longer life? Do they prefer preload for more precise positioning of the spinning part?
Also, should we define “preload?” Is this desired preload a desire to have a certain resistance to turning of the shaft which is created by compressing the bearings toward each other and thus results in a certain desirable bearing characteristic? Or is preload really a term that may mean “negative end float.” Or in other words, is it not so much a desire to create a turning resistance on the shaft as it is to have a tight fit of the bearings (no play/end float) even when the bearings are under load and parts may be flexing?
Someone here I am sure has seen manufacturing of differentials, so can enlighten me. I get the impression that they do not set these up at the factory but placing shims, measuring shaft torque, removing shims, etc. I get the impression they measure everything. Bearing tolerance, housing tolerance, etc. They then calculate the shims, insert them, and ship them. In other words, they are not trying to achieve a torque preload, they are calculating a negative end float. In the field, I wonder if we are simply trying to find that negative end float by a round about method called shaft rolling torque.
Tom
Hello Tom,
Negative End Float, or less than no clearance is actually how I describe pre-load to novice engineering students and the layperson.
Both excessive Pre-load and End Float can result in reduced bearing life, but moderately excessive pre-load is better tolerated than excessive End Float.
Its certainly not to simply create more drag on the bearings, but to maintain position and rigidity. As I’ve stated in earlier Posts, the pre-load of the Pinion Shaft Bearing arrangement decreases when load is applied through the differential gears. Too little pre-load results in a different tooth contact characteristic and usually unwanted noise.
I use solid spacers instead of the collapsible spacer and can get the correct setting most of the time in two assemblies. One with the Master Spacer used to determine the correct spacer length to use and two, the final assembly.
I suspect that the differential being manufacture today for supply to the vehicle manufacturing companies would be supplied ready to use. There would be too many dimension tolerance stacking to achieve the correct preload in a timely manner by measurement. This is made more difficult when using a collapsible spacer, where its unlikely that any two are going to react exactly the same when compressed.
I have read and been told a couple of times that crush sleeves are inferior,
and that modern diffs are not necessarily better, and that in fact, I have heard of several makes and models that are plagued by recalls that involve mass replacement of diffs
Okay. But it does seem that you may be agreeing the torque is not the goal, the negative end play is. So if the negative end play is the goal. And the negative end play is actually a compression of the bearings toward each other, would it really matter on a diff pinion shaft if after the negative end play were set. If the shim pack would magically disappear as long as the nut stayed tight?
Tom