I recently replaced the rear hub bearings on both sides, along with the outer axle universal joints; I had replaced the inners about 15000 miles ago, when I rebuilt the output shaft assemblies. I carefully set the wheel bearing endplay to zero, used new collapsible spacers to set the preload on the output shafts, and used Timken bearings throughout. The universal joints used were Dana Spicer 5-160 X (with grease fittings), and the crosses and caps were filled with grease on assembly. A quality job….. I literally have replaced probably 100 u-joints over the years, and 100s of wheel and differential bearings.
After driving a couple hundred miles, I thought that I would recheck everything and found that I had noticeable play in the 6-12 o’clock axis on both rear wheels. After measuring with a dial indicator and rock the wheel hub I found zero play in the hub bearings and output assemblies and about .012” play split evenly between the two universal joints, same on both sides.
Searching the internet doesn’t provide any definitive info; and Dana technical specifications do not indicate a free play spec, although they had issued a bulletin in 2015 indicating that .006” cross endplay (between caps) was normal. My Jaguar IRS experience is limited, but my Corvette C3 experience is that amount of play would be normal but would usually be due to the stub axle retention in the differential. When you think about propeller shaft u-joints, who has ever noticed longitudinal movement?
Thoughts? By the way, no unusual symptoms, no noises, no vibrations and nothing unusual noted when cornering.

 

Not me.

I've always thought of UJs as a zero play type of thing. Have I been wrong all these years?

A million years ago I had 6-12 o'clock rear wheel play as you've described....but a helluva lot more that 12-thou. More like 200-thou. Enough that the camber changes made the car behave oddly. In that case it was the stub axle bearings.

My observation is that a bit of play in the hub bearings is benign. But that's not what you have.


My gut says "Live with it. Nothing to fix here".

Let's hear what others have to say

Cheers
DD

 

Thanks, Doug- I was hoping you would reply. I value your opinion. I am with you, seems there should be zero play, but in reality there would have to be some tolerance. And they feel fine, went so far as to remove the splined end - very smooth, no binding.

 

There should be about 4 thou play in the hubs when the cebtral nut is done up tight. This is controlled by the spacer in the hub setup. Wheels off the ground, this is evident as a slight movement wheen you rock the wheels.

This gold coloured spacer should be of the correct thickness to ensure the bearings have 4 thou of endfloat in the hub assembly. If you have zero, then it would be best to buy slightly thicker spacers to protect the bearings from being too tight.

 

I recently replaced one carrier and set that one up with new Timken bearings and shimmed for zero clearance; then replaced the hub bearings on the other side with new Timken bearings and shimmed for zero clearance. After a few hundred miles I verified that nothing changed and still have zero clearance by use of a dial indicator. I also verified that there is zero movement in the output shaft assemblies. As I stated each universal has between .005-.006” end play, measured with the dial indicator.

 

Understood. As long as you understand that my point was that zero clearance is too tight on the bearings, factory spec is 4 thou endfloat.

 

Agreed. Spec is .001-.003” on some publications; others have it at .002-.004”, and some have the opinion that tapered roller bearings should be preloaded. Zero is a good compromise in my mind. Point being that the play in the wheel in the 6/12 axis is from the new Dana/Spicer u-joints….
Another datapoint is that I have two unused driveshaft u-joints still in the box, 1310 size, same Dana/Spicer product line. They have perceptible play between each cap and the cross, so must be intentional. There is a SPL Spicer line that are supposedly lifetime for a bit more money; Moss lists GKN (which was OE brand). There are also Moog here in the US that market parts that are generally “heavier duty” than OE…and Neapco which is a commercial truck supplier. Don’t know if any are “better” from a tightness standpoint. Pricing is from $8-$50 each USD.
Mostly wondering what others have used and if play was noted after.

 

Having no engineering experience I've always wondered about this. My experience is Jaguar and Corvette but I've read about other IRS rear hubs of similar design. Some call for preload while others call for free play. Even within Jaguar. My X300/XJR called for preload yet [what appears to be] the same hub design used on the late XJS call for free play.

I agree.

FWIW I've been routinely setting front bearings to zero ...and not an iota more....for years now. I do this because, IME, they always loosen a tiny bit with use.

Which I've never noticed before but, now that you've brought it up, I'll be looking for it in the future !


Conventional thinking is that the "better" or "heavy duty-ness" comes from the heft/size of the "X". Jaguar used robust UJs. Some replacements fit perfectly well but the "X" is spindly by comparison. But you may be onto something WRT tightness.

Would the Looseness or tightness in a UJ be a matter of design or a matter of manufacturing/quality control?

Cheers
DD

 

Doug, RGK
I used the top spec Dana Spicers 514101X from Rockauto. The difference was the size of the pins in the caps. The Dana had loads more thinner pins compared with the OEM GKN UJs.
I presumed that more thinner pins must equal better load spread and therefore longer life before the pins do this to the cross:


For all I know the quality of the steel may be better too. FWIW, I was able to discern no play at all in the UJs or in the assembled axle, other than the hub endfloat. And that ws still the case recently when I had to rebuild the outer lower wishbone fulcrums.

 

When I looked up the specs on Rockauto, the specs of RGKs DANA 5160X
https://www.rockauto.com/en/moreinfo...ZJ%2BRNg%3D%3D

and of the ones I used, DANA 513101X, were completely different:
https://www.rockauto.com/en/moreinfo...ZJ%2BRNg%3D%3D

So different that I suspect the former are not accurate, or I do not see how they would fit.

 

A contrary opinion....

Back in engineering school, we had a tech rep from Timken ( l'm dating myself) come and speak on taper roller bearings specifically....

From a loading standpoint, *all* tapered roller bearings are designed to be preloaded; A bearing with clearance is only loading one roller (at the top, in this case)
Under preload , all the rollers share the work.
Reflect upon how quickly a differential will destroy itself if pinion and gear carrier are not adequately preloaded...

Happy Thanksgiving 🐱‍👤

 

In this instance, clearance *increases* as the unit warms up, since aluminum (the carrier) expands faster than steel (the axle)

 

The 51310X are driveshaft “1310” size; half shafts are 1410….

 

Back in my Norton motorcycle days, engine main bearings had clearance grading, C1, C2, and C3, before they finally worked with R&M to come up with the Superblend roller (which had a larger radius on the roller ends). I wonder if when the u- joints are assembled, the tighter one become the SPL Lifetime, and the loose become the standard line, and the really loose get sold under other brands and packaging….

 

Camp Chaos Chronicles discussion of his rear suspension modifications; at 11:20 he discusses universal joint clearance, so if it’s good enough on the track, should be good enough on the road.

 

As a further data point, I have some of the GKN u joints that I removed from the half shafts, and picked a couple of trunnions and bearings caps that looked unworn. The inner bore of the caps was a consistent 1.007”, the trunnions a consistent.782”, and the bearing rollers a consistent .111”- indicating a designed in play of .003” So this would be minimum play in each joint with no wear and the shaft assembly would have.006” endplay. The hub center is approximately 5.75” above the fulcrum, so wheel movement (tip) at the top of the tire would be about 4 times that, so definitely noticeable.
I also noted that the GKN snap rings were .065” and .067”, so select fit evidently. Much thicker than the.059” copper colored Spicer….so end play could contribute.

 

Hi OP,

For what it's worth, I too have some level of free play in my recently fitted Dana UJ on the XJS rear half shafts.
I noticed the same in a regular test after refurbing everything & driving maybe 500miles - noticeable, feel-able play when rocking the rear wheel at the 6/12 o clock position in both the UJ's on both sides.
I'm confident I didn't install them incorrectly.
I used Dana Spicer JD joints without grease fittings (pre-filled with red grease as supplied)

I'm fairly sure this play didn't exist with my old UJ's, but they were a bit gummed up / seized so I don't really have a reliable 'new' comparison to make. Also hard to tell by feel over the standard rear hub bearing clearance.

My first thought was the sinking feeling of having been sold duff/possibly fake bearings, but I don't think that's the case. Would be very interested if you are able to find some facts on the expected/designed UJ axial play.

 

Dana/Spicers tech hotline wasn’t very helpful; I exchanged a couple of emails and they twice said to use there SPL series u joints because they have select fit retainers and thrust washers to control bearing cap endplay, not comprehending that my concern was bearing cap radial play. As above couple of posts seem to indicate that it is a normal condition. I did note that after I hit them with a grease gun there was less play felt, confirming that the play is internal to the joint.
This all started for me when I spun out on an icy freeway going about 70, when I inspected the car for damage I noted the play. When I tried to disassemble the axle to inspect, the stub was frozen in the hub. Pressing the hub apart destroyed the bearings and carrier…replaced the carrier with new bearings set to zero play, new u joints…and here we are. Probably was nothing wrong with the originals.

 

I had an opportunity to check an unmolested XJ-S with 58000 miles and it has similar play on both sides; also did some quick calculations and that .012” play equates to less than 1/8 degree, so am satisfied that it is a normal condition. Perhaps that is the reason that the factory specified using suspension compression locking tools when checking alignment….

 

99% of Jaguar XJS rear suspension assemblies are just plain worn out by the spec, but still seem to go down the road fine. We’ve become too picky as the XJS has transitioned from interesting used car to “classic car”. I still treat mine XJS hoard as used cars. I’m sure some of you would be horrified at the condition of cars I drive to work any day it is not actively snowing.