Rear wheel bearings

[Peter Harrison]

Hi All, I have had a number of rear wheel bearing failures over the last couple of years with them lasting only ~5,000 miles. Is there a bad batch of SNB bearings out there or is it more likely that they are not being fitted properly? Regards, Peter.

[malcolm]

No expert me, but over-tightening is a possible cause. Maybe others have experience of correct fitting still failing

[mgcjag]

Havent heard of bad bearings...however what ones are you fitting the cheap unbranded or branded Timiken etc...are you fitting them yourself.....easy to fit correctly. ...very easy to fit incorrectly.......what is the failure?...are your water slingers and dust shields fitted?..Steve

[abowie]

Havent heard of bad bearings...

Ditto.

I would be ensuring that you have the bearing clearances set correctly and that they are well lubricated.

Endfloat should be of the order of 4 to 6 thou. The method for setting this is in the manual.

[Peter Harrison]

Thanks - the fitting was by Barney Jones in Farnham (who is pretty experienced) who said he used SNG supplies with plenty of grease. Failures were evident by excess play in the wheels when jacked up together with a continual grinding noise. The current failure also features more of a crunch that sounds like a split bearing. (sorry – not sure what slingers and dust shields are)

[mgcjag]

Im assuming that you will have the bearings replased......i would then jack up the car and check each wheel....they should turn freely with a tiny bit of end float that you can only just feel if you rock the wheel.....if there is more play or dont turn freely then the bearings havent been set up correctly.....they are very different from the fronts that you just tighten up the center nut untill you have the correct endfloat.....rears have to be set with spacers the center nut just attaches the entire hub to the driveshaft......look at any parts diagran for water slingers and dust covers they keep all the muck out of you wheel bearings......SNGB offer a range of different bearings.....if your fitter thinking there is a bearing issue then he should be taking this up with the supplier........Steve

[angelw]

Andrew Wrote:

I would be ensuring that you have the bearing clearances set correctly and that they are well lubricated.

Endfloat should be of the order of 4 to 6 thou. The method for setting this is in the manual.

Hello Andrew.
That is way too much end float for a Taper Roller Bearing. 0.001" to 0.003" is what's specified in the S3 Workshop Manual and 0.003" is too much in my opinion. Manuals for earlier cars had the End Float specified at 0.002" to 0.006", but I suspect that by the time Jaguar got to producing the S3 cars, they had wised up to the fact that the End Float being used was too great. Taper Roller Bearing rather operate in slight Pre-load, rather than End Float.

Peter Harrison Wrote:

Thanks - the fitting was by Barney Jones in Farnham (who is pretty experienced) who said he used SNG supplies with plenty of grease. Failures were evident by excess play in the wheels when jacked up together with a continual grinding noise. The current failure also features more of a crunch that sounds like a split bearing. (sorry – not sure what slingers and dust shields are)

Hello Peter,
I contacted one of the usual suppliers asking the question, whether the bearings they supplied where Chinese or from Germany, the UK, the USA or Japan. The reply came back stating that these bearings were only available in Chinese Manufacture. That certainly made my eyes roll, for I knew for certain that they were available, manufactured in the other countries stated, I was simply asking if the bearing they supplied were Chinese or not.

There ARE such things as bad bearings and I wouldn't touch a Chinese Bearings with Barge Pole.

Not that I'm suggesting that "plenty of grease" is a cause of the failure of your bearings, but most go way overboard with the amount of grease used on wheel bearings.

Most of the CNC Machining Centers I maintain, operating 24/7 on die work, have Main Spindles that in a few months, will spin more revs than an E Type Wheel Bearing will have in the cars lifetime (assuming that the wheel bearings are never changed). By the time the Spindle Bearings of these machines need changing, they will have revved many E Type lifetimes of wheel bearing rotations.

If you can now visualize the amount of air space left around the Balls of a typical Ball Bearing Race, the correct amount of grease for a Ball Bearing Spindle Bearing for a Machine Tool, is only 10% of the available air space in the confines of the bearing. When these bearings are being lubricated on assembly, a measured amount of grease is applied using a syringe.

When I grease new wheel bearings on assembly, I pack as much grease into the bearing that I can by hand and that's it; they get no further grease.

Regards,

Bill

[Peter Harrison]

Hi all, many thanks for the feedback. If there are indeed substandard Chinese bearings being supplied, there should be similar observations from other high-mileage E-typers. Regards, Peter.

[mgcjag]

I havent heard of wheel bearing failures as an issue on E type/Jag forums....You should not have an issue with unbranded/Chinese bearings as yours were supplied and fitted by your Jaguar specialist......they should know to fit well known branded bearings as available from SNGB and many other suppliers...... Steve

[abowie]

Most of the bearings have the source on them. Timken USA etc.

[abowie]

With regard to endfloat vs preload on tapered bearings. I did a bit of reading on this a while back.

If you look at lifespan of bearings vs endfloat/preload you get a graph that looks a bit like this (apologies for my laughable inability to do this with a computer program). Note that while this is very rough, the slight shift across to the preload side is accurate.





Manufacturers of tapered bearings usually specify a small amount of endfloat (looseness) cold. When the bearing comes up to operating temperature it expands and the bearing ends up with close to zero endfloat. At this point there is the best compromise between part location (movement) and bearing wear (tightness).

Depending on the application one of these factors may be more or less important than the other, hence specification differs with different applications.

What you can see though is that if the bearing is very loose or very tight the lifespan of the bearing is significantly compromised.

So for the OP I'd be ensuring that the next set of bearings are set up in spec. The graph above does agree with Bill's comments about aiming for the low end of the endfloat scale when setting up these bearings.

[angelw]

Hello Andrew,
Following is a graph grabbed from an SKF Manual and clearly indicates that bearing life is significantly improved as the bearings are placed in pre-load. Granted there is a sharp drop off in bearing life as pre-load is increased past the optimal point, but even at circa -0.0055" pre-load, it's still better than 0.005" end float. Accordingly, taking into account the longitudinal expansion of the aluminum Hub Carrier, starting at Zero End Float will only improve bearing life as the pre-load on the bearings increase.

Regards,

Bill

[abowie]

It's so hard for us non engineers to make sense of this stuff. But very interesting.

Looking at them, it strikes me that our graphs are the mirror image of each other...

This is from https://www.timken.com/wp-content/uploa ... hure-1.pdf



Vis a vis the hub carrier, that's an interesting point.

I wonder if thermal expansion would be a big issue with it. Both because any heat generated by the small bearings would be distributed into its large thermal mass, and also because that heat would be quickly dissipated by its large surface area. I haven't checked, but I reckon after a drive the hub carrier temperature would be pretty close to ambient.

[abowie]

Actually, looking at your graph, are you reading it the wrong way?

On the X axis it is is reading clearance. That is endfloat.

So I reckon your graph is identical to mine, just the other way around.

[angelw]

Andrew Wrote:

Looking at them, it strikes me that our graphs are the mirror image of each other...

Hello Andrew,
I'm not reading the graph incorrectly.

They're only mirror image in appearance, not in what they are expressing. Typically, clearance is shown as a plus value, therefore, it follows that preload would be indicated with negative values.

In your sketch, you've shown the amount of pre-load as a positive value; accordingly, you would have to show clearance as having a negative amount of interference.

They are both saying the same thing.

Regards,

Bill

[Phil P]

Having struggled recently with setting the end play to the 4 thou nominal specified by Jaguar’s manual, must I now rip it apart to go for a preload? Not likely.

Both graphs appear to say the same thing to me.

In the orange graph which has more specific data, a 0.100mm ( 4thou) end play gives a life of around about 1200hr. If you say that most journeys average 35mph then that’s a life of about 40,000 miles….not bad. Certainly not failing within 5000 miles.

My suspicion is that factors other than solely end play/preload contribute more in our cars to failure particularly whether the bearing is axially loaded by the bearing tracks and the squareness of the races in the housings.

Increase in preload due to temperature doesn’t appear to be significant. A quick calculation shows that for a 10 oC rise in temperature, preload would increase by about 3/10 thou.
Yours,
Phil