rambling rollie wrote:

Greetings!

At what point does one do this bearing replacement?
At the "humming" phase? (i almost believe "humming" from FL2 rear diffs is standard)...
Or do you wait until it starts whining? (a significantly higher pitched version of humming).
Or when it starts making "whirring" sounds?
Obviously once it starts making grinding noises it's probably overdue...

I ask because mine has been humming for months (maybe even a year, i cannot remember)....

Before it seizes up, which mine did while driving at 40 MPH, just before Christmas. Although to be fair, mine had no oil in it, so I'm surprised it lasted as long as it did. Hyundai Ioniq 5 Ultimate. The family car.
2009 Rimini Red SE TD4. Gone.
2006 Tonga Green i6 HSE. Gone.
Audi A5 convertible, the daily driver.
1972 Hillman Avenger GT, the project.

MRRover75 wrote:

Hi,

I have given up speculating the main reason why these pinion bearings fail. Only thing I know, is that the 9 ones I have rebuilt, have worked superb and I have no reports back from anyone that they have failed. I guess these has not been on the road long enough to build a greater mileage, but some should have failed allready if something was assembled wrong. I have only used Timken and SKF bearings. Mine has now done aprox. 50000kilometres after assembly, so probably as good as they was when new. Early diffs (-07-09) seem to hold 200000km+ before the pinion bearing fails. later ones more early for some reason but some mentioned that cheap bearing copies from China was used there !?!

Regarding lubrication, there is two oil channels to the outer pinion bearing in the casting. One lower probably below the oil level, and one higher up, probably for ventilation of some kind. Lack of frequent oil changes is probably contributing to a reduced lifetime...


I also feel the pre-load on the bearing seems a bit high, but I have not gambled on reducing this on my rebuilds. Just sticking to the OEM procedure when tightening up. Only thing is that I throw away the torque wrench. A lot more torque is needed to start compressing the collapsible spacer than the specified 250Nm.

Mine actually seized solid while doing 40 MPH just before Christmas, so my diff is now most likely scrap. The cause for this was actually due to there being nothing more than a few drops of oil in it.

I've studied the diff design, and come to the conclusion that these continual failures are probably down to a combination of factors.

The pre-load is I believe too high, for the size of the front bearing used. High pre-load, will result in delamination of the case hardening, which is what appears to happen on these bearings.

This ailment also effects the Discovery 3 front diff, which again is due to excessive pre-load, coupled with too small a bearing for the task. I've rebuilt several D3 front diffs, all suffering the same delamination of the case hardening, and all had excessive pre-load when reassembled using the factory shims, but thankfully its easy corrected before final assembly.

Lubrication, or lack of. Looking at the diff design, the oil feed to the front bearing is poor at best.
All differentials use oil for lubrication and cooling. A standard diff is designed so that oil is freely thrown around all moving parts by centrifugal action of the crown wheel. Almost all differential cases are designed to channel a large proportion of this oil directly to the outer most pinion bearing. See the size of the cast in oil gallery hump in this Freelander 1 rear diff, where the oil travels to the pinion bearing in large quantities.

The pinion is spinning fast by comparison to the crown wheel, so needs a good supply of fresh oil to replace that, oil which is displaced by the bearing rollers.

This is where the flat design of the Freelander 2 diff is lacking, as there's insufficient height for a large oil gallery to capture this centrifuged oil, limiting the oil supply to the front bearing. If the oil supply to a bearing is compromised, then the bearing will run with less than an ideal amount of lubrication, which then causes it wear faster and to run hot. This lack of lubrication, coupled with the higher than I feel is ideal pre-load, is my best guess as to the early and frequent demise of these bearings. Hyundai Ioniq 5 Ultimate. The family car.
2009 Rimini Red SE TD4. Gone.
2006 Tonga Green i6 HSE. Gone.
Audi A5 convertible, the daily driver.
1972 Hillman Avenger GT, the project.

Hi all,

Just an update and a great excuse to bump this thread

Luckily, the diff I rebuilt in the beginning of this thread is still running smoothly and in daily use. Many miles has been covered since then..... (knock on wood)....

I get contacted very randomly from people that have heard the rumors, and they need their diffs rebuilt.
no. #16 just went out the door and no.#17 is now during assembly at the workbench:



Hope all lasts

Some quick questions for MRRover75 (or anyone else who knows the answer really)...

I've just purchased a recon diff from Bell Engineering but want to add the drain plug modification...however I want to clean out any swarf which finds its way into the housing, which means removing the LH cover.

Does anyone know:
i. Can the diff cover gasket be reused, and if not, what is the part number?
ii. What is the torque for the diff cover bolts?
iii. Loctite or not on the bolts?

Thanks
Andrew

Scratch that...did a bit'o'googling and found the answers to my own questions...

i. Differential case seal = LR030846
ii. x7 Differential case bolts tightened to 29Nm
iii. Not specced in LR workshop manual (but a bit of blue wouldn't go amiss)

👍

glad you found the answers Jules

An interesting thread, and many thanks for the photo's that show all. I had one done at our local specialist. He said that the bearing fitted is under-rated and Landrover later fitted a larger nose bearing. I have had two diffs rebuilt, so far, at different specialists. I would always use EP90 Hypoid gear oil in a differential. I used to make tooling for the old Rover Company, and diff. pinion bearings were always torqued at 120 lb ft., with a four thou preload. the spacer was ground to suit. The nut was staked, to lock it. Three foot of bar on the socket, and lean all one's weight on it was usually good enough. ( nobody was over eleven stones in those days). I still have a few of the setting blocks in the toolbox. Engineers blue was used to check the gear alignment. The gears, in use, wipe across the contact area, which is why a high shear oil is used. Commercial vehicles use higher spec. oils for longer life. We have many local Lorry Companies with million mile plus vehicles around here, and money is better spent in prevention, rather than constant repairs. It is a pity Landrover has cur so many corners in recent years. Having refurbished some of their tooling in the Nineties, quality improved for a while, until the Bean Counters got too strong.

Good someone has the correct answers allready

I would like to add, that there should be no problem to reuse the differential case seal as yours is fresh and clean. An O-ring has no problem to tolerate that. I use to add a very thin line of Loctite flange sealant to the flange on mine, just as an extra seal. Many of these have some pitting/corrosion close to where the O-ring seals, so I do this in case the O-ring does not seal properly. Probably not needed to yours I guess.

Add some blue loctite to the bolts on refitting

Unbeliever wrote:

Scratch that...did a bit'o'googling and found the answers to my own questions...

i. Differential case seal = LR030846
ii. x7 Differential case bolts tightened to 29Nm
iii. Not specced in LR workshop manual (but a bit of blue wouldn't go amiss)

👍

Many thanks for the clarification MRRover75...much appreciated 👍

MRRover75 wrote:

Hi all,

I Have been busy with other things lately, but some more work has been done the last evenings. I continued with removing the pinion bearing races from the diff housing. The inner race was removed by punching it out with a suitable punch/drift sequentially side-by-side on its inner face. There is a shim between the bearing race end face and the diff housing, make sure this one don`T get damaged when punching out the bearing race. The thickness of this shim has probably been set at the factory to set the necessary pinion/crownwheel clearance?



I made a simple slide-hammer pulling tool for the outer bearings race. A suitable socket was found and mounted on the end of a M12 threaded rod. I found a good lump of steel and drilled a hole through to fit this one on the threaded rod. This made me able to pull out the bearing race:





The shaft seals was easily removed by punching out from the inside:



OK, Now everything is removed and the diff housing should be cleaned up throughly. All bearing and seal surfaces shall be inspected for damages and smoothed out with scotch brite i.e. as neccesary. Clean up the workbench and make everything ready for assembly.

Cleaned housing:

Click image to enlarge

Click image to enlarge

Hi,

I need to look back into my records to confirm your statement.

I have rebuilt both the early diffs and the later which have a larger bearing from the factory. From my mind, the difference is a wider bearing, a shorter collapsible spacer and a diff housing that has a deeper bore to accept the wider bearing.
How Bell Engineering does this, I don't know, but there might be possible that he get the bearing bore in the diff housing machined deeper to accept the later bearing.

Edit:

Copy/pasted from one of my earlier threads, the bearing designation you have mentioned is the correct one (at least size-wise) for the later diff. Here is my findings when I did my first one with the larger bearing:

On disassembly of the later diff, I found a shorter collapsible spacer than on the older ones, so I assume that the same is not used on both. The shortening in length reflects the extra width of the larger bearing. Despite a lot of searching, I could not find the part number for the shorter spacer, so I took off 4mm in a lathe of a long one to make everything fit. Attached picture shows the used spacers:

MMRover75, hi.

Thanks for confirming this. I see in this link https://parts.jaguarlandroverclassic.com/p...and-rover/ you can find the JLR part number LR082099 for VIN FROM VINBH257091:
[img]"D:\Spacer LR082099 from BH257091.png"[/img]

I'm happy then that I got quotes for the right bearing size.

You showed an SKF 32206 J2/Q https://docs.rs-online.com/639f/0900766b80b44713.pdf bearing, and from my research in comparing load ratings, you would be better off with a bearing of the same designation and higher static and dynamic load rating figures i.e. C0 and C respectively.
The SKF 32206 J2/Q which has the attached performance data, but summarised here:

Basic dynamic load rating C0 = 50.1 kN (Timken is 71.9 kN higher by 27.94%)
Basic static load rating C = 57 kN (Timken is 64.1 kN higher by 26.14%)
Fatigue load limit Pu = 6.3 kN (Timken No data, but likely comparable)

So for me the Timken bearings are the more durable, and considering the labour costs makes the most sense for the application. This is true for the Timken bearing even when comparing the smaller 30206 designation to the other manufacturers.

For interest sake, I got a quote yesterday (Sept 8th, 2025) for the Timken 32206 bearing at ZAR 336.46 excl. VAT (GBP 14.18 excl. VAT at ROE of ZAR23.73 to GBP1) in South Africa. So rather inexpensive for the effort to get it in and certainly cheaper than from the JLR dealer.

I share my rather lame comparison spreadsheet here: https://1drv.ms/x/c/42bfdbefbc55972e/ES395...g?e=9UfRfP. Link expires Sept 14th 2025.

For the non mechanical engineer, this is a simple explanation of the load ratings: https://www.skf.com/africa/en/products/pla...ad-ratings and you will see the the static load rating section they mention shock loading, which is likely the events when L359 sets off and the Active Coupling engages as power is rearward biased, before much of the power gets 'handed over' to the front axles as speed increases.

This is a great site to hunt to parts: https://parts.jaguarlandroverclassic.com/p...and-rover/

I have included the procedure 51.15.10 published Feb 11th 2019 from TOPIx here as reference: https://1drv.ms/b/c/42bfdbefbc55972e/ES6XV...A?e=Yh2FWj and expires Sept 14th 2025 also, which shows initial torque specification of 200Nm (not 250Nm as in the repair manual dated May 2011 if not mistaken). Note this is based on my VIN ending FH42###8.

Here's a lovely resource from Timken on taper roller bearing setting: https://www.timken.com/wp-content/uploads/...hure-1.pdf which I think is helpful in understanding the engineering of repairs to our vehicles. I may be the annoying customer that would give my repair clear instruction on what I expect of them making a repair, because the reality is: most workshop do not have the time to keep abreast with ANY of the data out there, and you almost certainly cannot expect the apprentice (or parts fitter if not a trained technician) to enjoy reading dry datasheets and repair manuals and questioning everything...T1kT0k is certainly more entertaining (I have no idea whether it is).

So this is my bit to help anyone eager to know what their repair (or they the DIYer, like myself) ought to do.

Hi,

Thanks for the info. And thanks for the procedure. Seems to be a newer / revised procedure than the LR procedure I get hold of earlier.
Its correct as you say, the Timken bearings have higher load ratings in the datasheets than the SKF bearings. Remember this when browsing the datasheets "back in the days". You will see some of this in my first posts in this thread. Timken should be the 1.st choice. What I find strange, when buying the complete "OEM" overhaul kit from Rimmers, I get 3 Timken bearings and 1 SKF bearing where the SKF bearing is the pinion nose bearing. No problem to swap it out for a Timken bearing but my rebuilt diffs seems to hold up pretty well (knock on wood).

For the initial torque, the 250/200 Nm will never be sufficient to crush the collapsible spacer all the way down. I used the torque wrench for this on my first handful of diffs, but found out there was no need. You have to use brute manpower on the breaker bar to get things moving. Tightening a tiny bit by bit until the play in the bearing assembly is gone, then start measuring the dynamic torque and do tiny steps of further tightening until the required dynamic torque is reached. I try to put it on the low side of the specifications as there have been speculations that the specified dynamic torque is to high from the factory.

And some advice in the end: When first doing this job, its recommended to replace all 4 bearings. I have had one diff where one of the crown wheel bearings had failed instead of the nose bearing, and one other where one crown wheel bearing failed 100.000km after the nose bearings only was replaced (that was my diff #1 referred to in the first posts.). Doing all 4 requires more tooling like spacers, flanges and sleeves to be done. Most important is that the shims behind the outer races of these 3 bearings gets back in the same place as they set the internal clearances.
I have done 19 of these for now. 2 of them had the large nose bearing from factory.


Keep going the good work, your info and findings adds to our experience and knowledge

MMRover75, hi.

Thanks for this. And yes, I think it best to replace all four bearings. Just last night I emailed Bearing Man Group to order the additional 2x Timken 32208X bearings for the differential crown wheel carriage. So advice taken!

On the procedure - I wonder then as you’ve mentioned, that with the factory preload being overdone, the 200Nm is a start from which to do the multiple (as necessary) 20deg incremental tightening the nominal 1.1Nm at 60rpm.

You’ve now done many of these repairs: after achieving the 1.1Nm within specification tolerance, what was the end float at the pinion shaft? Was hoping you perhaps recorded it for each bearing configuration i.e. with the 30206 and 32206 nose bearings respectively. Ultimately, this is what the preload is aiming to achieve i.e. desired bearing internal running clearance and, as such, the dynamic torque and end float. From your recent response there is a difference in size of the collapsable spacer.

Bearing kits - that did surprise me upon inspection that the nose bearing is an SKF (or other with a similarly lower static load rating). That is why I’d much prefer the same brand based on a datasheet selection for the same physical dimensions.

Typically, bearings are designed into a system with a reliability in excess of 99% (assuming designing for ‘infinite life’) and 100 000hrs design life. The service life will ultimately be a little less, but that is a lot of hours to endure the loads experienced during daily and harsh off roading. Definitely an engineering marvel they can last even 200 000km odometer kilometres before failure.

So, I am considering tooling up as my LR independent mechanic cannot recommend any good repairer where I am from. And it seems they either don’t have the tools, and I suspect they don’t have the specs nor would they accept it if given them. Can’t believe business survive like this and neglect training their workers.

What I’d like to know is what the size of the spline socket for the pinion shaft is i.e. external diameter and number of teeth? I saw in one of your older posts you used the old Haldex coupling female splined portion, which I do not have the luxury of having.

I’ll be having the 41mm socket made as the JLR special tool. Seems simple enough.

As for my efforts to contribute, you are all most welcome. I find technical forums a deep well to draw from when one sifts through the content…you find gems much like your write up…so thank YOU!

Well, it is a lovely spring day in Cape Town, South Africa…off to the beach!

Have a great day.

Hi,

A quick reply in a busy day...
There are no end float on the pinion when tightened up according to the procedure. The assembly is tightened up so much that the play in the conical bearings are gone and axially loaded. This friction is what are creating the dynamic torque of 1,1Nm which is an indirect measure of how much the assembly are pre-loaded. You can barely rotate the pinion with the bare hand when tightened up, but no problem when using the spline socket. Just as a simple comparison

Getting the spline tool might be hard if you do not have access to the dedicated tooling or a Haldex to scrap. I have read somewhere that the same spline dimension is used in a Ford Mondeo clutch plate for a given year/ engine size. Need to search up that.
I will take some measures at my next chance.

Appreciate your feedback and curiosity on this